(Brill's Indological Library 49) Vishwa Adluri, Joydeep Bagchee - Argument and Design - The Unity of The Mahabharata-Brill Academic Publishers (2016)
(Brill's Indological Library 49) Vishwa Adluri, Joydeep Bagchee - Argument and Design - The Unity of The Mahabharata-Brill Academic Publishers (2016)
Edited by
Johannes Bronkhorst
In co-operation with
VOLUME 49
Edited by
Vishwa Adluri
Joydeep Bagchee
LEIDEN | BOSTON
iv
Cover illustration: František Kupka (1871–1957), Hindu Motif, also known as Graduated Red (1919–23). In
the collection of the Musée National d’Art Moderne, Centre Georges Pompidou, Paris, France.
Want or need Open Access? Brill Open offers you the choice to make your research freely accessible
online in exchange for a publication charge. Review your various options on brill.com/brill-open.
Typeface for the Latin, Greek, and Cyrillic scripts: “Brill”. See and download: brill.com/brill-typeface.
issn 0925-2916
isbn 978-90-04-31139-8 (hardback)
isbn 978-90-04-31140-4 (e-book)
Contents
Foreword vii
Robert P. Goldman
Notes on Contributors xii
6 The Tale of an Old Monkey and a Fragrant Flower: What the Mahā
bhārata’s Rāmāyaṇa May Tell Us about the Mahābhārata 187
Bruce M. Sullivan
Index 471
Contents
Contents—v
Foreword
Reading and Rethinking with the Abhinavadvaipāyana—vii
Notes on Contributors —xii
Introduction
From Supplementary Narratives to Narrative Supplements—1
Vishwa Adluri
Chapter 1
Not Without Subtales: Telling Laws and Truths in the Sanskrit Epics—10
Alf Hiltebeitel
Chapter 2
On the Upatva of Upākhyānas: Is the Uttarakāṇḍa of the Rāmāyaṇa an Upākhyāna of the Mahābhārata?—69
Robert P. Goldman
Chapter 3
The Epic’s Singularization to Come: The Śakuntalā and Yayāti Upākhyānas in the Southern Recension of the Mahābhārata —83
Joydeep Bagchee
Chapter 4
Introductory Notes on the Literary Structure of the Mārkaṇḍeyasamāsyāparvan—127
Greg Bailey
Chapter 5
Of Daddies and Demons: The Rāmopākhyāna and the Rakṣovaṃśa of the Vālmīki Rāmāyaṇa—161
Sally J. Sutherland Goldman
Chapter 6
The Tale of an Old Monkey and a Fragrant Flower: What the Mahābhārata’s Rāmāyāṇa May Tell Us About the Mahābhārata—187
Bruce M. Sullivan
Chapter 7
Supernatural Conflicts, Unanimities, and Indra in the Main Story and Substories of the Mahābhārata—206
Fernando Wulff Alonso
Chapter 8
Pride and Prostitution: Making Sense of the Gālava Exhibit in the Mahābhārata Museum—237
Adheesh Sathaye
Chapter 9
The Divine Androgyne: Crossing Gender and Breaking Hegemonies in the Ambā-Upākhyāna of the Mahābhārata —275
Vishwa Adluri
Reflections on the Upākhyānas in the āpaddharmaparvan
Chapter 10
Reflections on the Upākhyānas in the Āpaddharmaparvan of the Mahābhārata—320
Adam Bowles
Chapter 11
The Status of Upākhyānas in Madeleine Biardeau’s Reflections on the Mahābhārata—359
Nicolas Dejenne
Chapter 12
Toward a Geography of the Mahābhārata’s Uñchavṛtti Brahman—378
Thennilapuram Mahadevan
Chapter 13
Upākhyānas and the Harivaṃśa—388
Simon Brodbeck
Chapter 14
The Geography of the Mahābhārata’s Upākhyānas—428
Alf Hiltebeitel
Index—471
Reading and Rethinking with the Abhinavadvaipāyana vii
Foreword
Foreword
But seriously, the record to which I refer encompasses Alf’s long and extraordi-
nary journey along the authorial paths that wend their way through the two
great epic poems of ancient India and out along the many and diverse streams
of literary, artistic, and performative cultural production that they have in-
spired from the time of their first circulation to the present day. He began this
journey as an undergraduate studying religion at Haverford College and moved
on after graduation to the University of Chicago, where he had the good for-
tune to work with such notable scholars as Mircea Eliade and J.A.B. van Buite-
nen, who guided his studies and gave him his grounding in Sanskrit. His
dissertation, “A Study of the Mahābhārata in Relation to Indian and Indo‑Euro-
pean Symbolisms,” gives a very early indication of where his scholarly interests
lay and in what direction they would lead him. He also has been profoundly influ-
enced by the work and example of the late, great epic scholar Madeleine Bi-
ardeau.
Along with only a very small cadre of academicians of his generation—schol-
ars such as Jim Fitzgerald and John Brockington—Alf has focused virtually all of
his attention on India’s extraordinary cultural legacy that is the great, ancient
Sanskrit epic poems and the incalculably enormous influence they have exerted
for millennia over the arts, literatures, religions, and societies of a vast sweep of
Reading and Rethinking with the Abhinavadvaipāyana ix
southern Asia stretching from Iran in the west to the Philippines in the east. He is
also all but unique among scholars of Vyāsa’s prodigious poem to have moved
beyond the philological and interpretive approach into some of the modern,
lived, and performed legacies of the epic, most notably the South Indian cult of
Draupadī.
Most noteworthy, I think, is Alf’s welcome call to “rethink” the epics. For, the
fact is that the grand and complex texts, in all their virtually innumerable forms,
that we classify simply as “epics” form by themselves a category that—like ani-
mals in Lévi-Strauss’s famous phrase—are “good to think with.” That is to say
that within the vast corpus of Indic texts they form a distinct, if difficult to define,
genre deeply imbricated with many other kinds of text and yet undeniably and
unmistakably of a particular type, whether we call that type itihāsa or kāvya, two
terms that have been used in various contexts of both works.
From the earliest commentaries on the poems and their fragments through
the sectarian traditions of their reception in the medieval period, down to the
huge corpora of modern scholarship and apologia in India and around the world,
the epics have presented a pertinent example of the andhagajanyāya, “the max-
im of the blind men and the elephant,” according to which people with differing
views, beliefs, and agendas impose their own restricted views on a topic that is
vast, variegated, and complex. This is especially the case with the Mahābhārata, a
work so capacious and, in its own words, so comprehensive that, when it comes
to the myths, legends, and the ideologies of religious, social, military, and political
life that formed the core of its author’s worldview, ṛṣi Vyāsa’s boast,
Students of the epic and its receptive history know how, for example, Indian
thinkers have read the text variously as a tract illustrative of advaitavedānta or
dvaitavedānta, a poem imbued with the sentiment of detached tranquility
(śāntarasa), an allegory of the conquest of the passions or an exhortation to
militant revolution. Similarly, scholars from the nineteenth century onward
have understood it variously as a nature myth, an expanded vedic legend, an
inverted anti-Pāṇḍava history, a dharmaśāstra, a confused and inconsistent
pastiche of a long tradition of bards and redactors, and as a single, unified work
x Foreword
Robert P. Goldman
William and Catherine Magistretti Professor of Sanskrit
The University of California at Berkeley
xii Notes on Contributors Notes On Contributors
Notes on Contributors
Vishwa P. Adluri
is Adjunct Associate Professor of Religion and Philosophy at Hunter College,
New York, and the author of over a dozen articles and essays on the Mahā
bhārata. His work mainly focuses on the reception of ancient thought—both
Greek and Indian—in modernity. He is the author of Parmenides, Plato and
Mortal Philosophy: Return from Transcendence (London: Continuum Publish-
ing, 2011), The Nay Science: A History of German Indology (New York: Oxford
University Press, 2014), and Philology and Criticism: A Guide to Mahābhārata Tex-
tual Criticism (London: Anthem Press, forthcoming in 2016), and has edited nu-
merous volumes on the Mahābhārata (including a two-volume edition of Alf
Hiltebeitel’s collected essays—Essays by Alf Hiltebeitel, 2 vols. [Leiden: E. J.
Brill, 2011]). Vishwa has a PhD in Philosophy from the New School for Social
Research, New York and a PhD in Indology from Philipps-Universität Marburg.
Joydeep Bagchee
has a PhD in Philosophy from the New School for Social Research, New York,
and is currently a fellow at the Freie Universität Berlin. His current research
focuses on the intersection of the textual sciences, philology, textual criticism,
and the history of science, with the aim of illumining how contemporary ideas
of scholarship and especially of historical rigor developed. Bagchee is especial-
ly interested in the textual traditions of the two Sanskrit epics. He is co-author
of The Nay Science: A History of German Indology (New York: Oxford University
Press, 2014) and Philology and Criticism: A Guide to Mahābhārata Textual Criti-
cism (London: Anthem Press, forthcoming in 2016).
Notes on Contributors xiii
Greg Bailey
is an Honorary Research Fellow in the Program in Asian Studies, La Trobe Uni-
versity, Melbourne. He has published translations and studies of the Gaṇeśa
Purāṇa, Bhartṛhari’s Śatakatraya and books on the god Brahmā, early Bud-
dhism, contemporary Australia, and many articles on Sanskrit literature. At
present he is working on the relationship between early Buddhism and the
Mahābhārata.
Adam Bowles
is Senior Lecturer in Asian Religions at the University of Queensland in Aus-
tralia. He has published three volumes concerning the Mahābhārata. Two of
these are translations of the Karṇaparvan published with the Clay Sanskrit
Library. The other, entitled Dharma, Disorder and the Political in Ancient In-
dia: The Āpaddharmaparvan of the Mahābhārata (Leiden: E. J. Brill, 2007), is
a study of one of the didactic corpora belonging to the Śāntiparvan. He has
also edited, with Simon Brodbeck and Alf Hiltebeitel, The Churning of the Epics
and Purāṇas (New Delhi: Rashtriya Sanskrit Sansthan and D. K. Printworld, in
press). In addition, he has research interests in aspects of the Maratha polity of
the seventeenth and eighteenth centuries, components of which appeared in
the co-authored monograph A History of State and Religion in India published
by Routledge in 2012.
Simon Brodbeck
is Reader in Religious Studies at Cardiff University in Wales. He is the author
of The Mahābhārata Patriline: Gender, Culture, and the Royal Hereditary (Ash-
gate, 2009) and some two dozen articles on aspects of the Mahābhārata, and
co-editor of Gender and Narrative in the Mahābhārata (Routledge, 2007, with
Brian Black) and The Churning of the Epics and Purāṇas (D.K. Printworld, in
press, with Alf Hiltebeitel and Adam Bowles). He also co-edits the journals Re-
ligions of South Asia (Equinox, est. 2007) and Asian Literature and Translation
(open access, est. 2013). He is currently engaged in translating the critically
reconstituted Harivaṃśa.
Nicolas Dejenne
is Assistant Professor in History and Textual Traditions in the Indian World at
the Université Sorbonne Nouvelle-Paris III in France. After a monographic study
about the figure of Paraśurāma (“Du Rāma Jāmadagnya épique au Paraśurāma
contemporain: Représentations d’un héros en Inde,” PhD diss., 2007), his main
field of research has remained the Sanskrit epics and Purāṇas for which he
has been a regular participant at DICSEPs and dedicated WSC-sections since
xiv Notes On Contributors
the beginning of the 2000s. He also works on the history of Indian studies in
France and is a co-editor for an ongoing French Encyclopaedic Dictionary of
Indian Literatures.
Robert P. Goldman
is the William and Catherine Magistretti Distinguished Professor of Sanskrit at
the University of California at Berkeley. He is the General Editor and a princi-
pal translator of the Vālmīki Rāmāyaṇa Translation Project, a consortial proj-
ect to provide a scholarly translation of poem’s Critical Edition as prepared by
scholars of the Oriental Institute of Baroda. The seventh and final volume of
the translation, the Uttarakāṇḍa will go to press by the summer of 2015. Gold-
man’s major areas of interest are Sanskrit epic and literary studies and Indian
cultural history. He is a Fellow of the American Academy of Arts and Sciences
and was the recipient of the President of India’s Certificate of Honour for San-
skrit (International) in 2013.
Alf Hiltebeitel
is Professor of Religion, History, and Human Sciences at the George Washing-
ton University. His publications take him back and forth between the
Mahābhārata and fieldwork on Tamil Mahābhārata “folk” traditions. From this
tandem project, his work branches out into related texts, other cults and oral
epic traditions, and recently into an attempt to understand the Indian concept
of dharma. Recently out are the following titles: Dharma (Honolulu: University
of Hawai’i Press, 2010); Reading the Fifth Veda: Studies in the Mahābhārata, Es-
says by Alf Hitebeitel, vol. 1, and When the Goddess Was a Woman: Mahābhārata
Ethnographies, Essays by Alf Hiltebeitel, vol. 2 (Leiden: E. J. Brill, 2011)—both
edited and introduced by Vishwa P. Adluri and Joydeep Bagchee; and Dharma:
Its Early History in Law, Religion, and Narrative (New York: Oxford University
Press, 2011; New Delhi: Oxford India, 2014).
Notes on Contributors xv
Thennilapuram Mahadevan
is an adjunct faculty member in the English Department at Howard University
in Washington, D. C. His research interests center around the Vedic Oral Tradi-
tions and their texts, both metrical and pravacana style, the rituals and the oral
agencies behind them, eventually the historical Brahmans. He has studied the
dissemination of the Vedic schools and the later migrations of Brahmans of
different schools to the peninsula, as attested by both epigraphy and fieldwork.
He has shown that the Northern Recensions of the Sanskrit epics arrived in the
peninsular India with Brahman migrations, from which were made their
Southern Recensions. He has also studied the textual history of the Southern
Recensions of the epics and their journeys to Poona and Baroda and the Criti-
cal Editions. His work appears in the Electronic Journal of Vedic Studies.
Adheesh Sathaye
is Associate Professor of Sanskrit Literature and South Asian Folklore at the
University of British Columbia. He is the author of Crossing the Lines of Caste:
Viśvāmitra and the Construction of Brahmin Power in Hindu Mythology (New
York: Oxford University Press, 2015). His research concerns Hindu mythologi-
cal literature, the cultural history of caste, Sanskrit poetry, drama and story
literature, and the broader intersections between performance, textual pro-
duction, and traditional culture in South Asia.
Bruce M. Sullivan
(Professor of Comparative Study of Religions and Asian Studies at Northern
Arizona University) is a specialist in Hinduism and Sanskrit literature, and the
author of four books, including a study of Vyāsa in the Mahābhārata (Leiden:
E. J. Brill, 1990). Two books were coauthored with a scholar in India, N. P. Unni,
each a translation and study of a Kūṭiyāṭṭam drama (Delhi: 1995 and 2001). Sul-
livan also edited and contributed to Sacred Objects in Secular Spaces: Exhibiting
Asian Religions in Museums (London: Bloomsbury, 2015). His educational back-
ground includes a PhD in South Asian Languages and Civilizations and the
History of Religions from the University of Chicago.
xvi Notes On Contributors
Introduction 1
Introduction
The premise for this volume was that the Mahābhārata is a work of literature
and that its upākhyānas (subtales or, perhaps more accurately, proximate nar-
ratives) are central to its literary project. The volume’s starting point was Alf
Hiltebeitel’s 2005 essay “Not without Subtales: Telling Laws and Truths in the
Sanskrit Epics,”1 which was the first article to argue that rather than being
extraneous to the main epic (a putative “Bhārata” or a Mahābhārata without its
ancillary narratives) the upākhyānas may well be central to the epic’s literary
design.2 In this essay, Hiltebeitel argued “it should no longer be enough to tell
its [the Mahābhārata’s] main story, especially with the suggestion that its main
story would have been an original ‘Bhārata’ with the rest making it a ‘Mahāb
hārata.’ Even though it must require shortcuts, one owes it to this grand text to
attempt to block out the main story against the backdrop of its archetypal
design, which includes its frame stories, upaparvans, upākhyānas, and the
enigma of the author.”3
In response to Hiltebeitel’s provocative suggestion, a group of epic scholars
gathered at the Forty-First Annual Conference on South Asia in Madison,
Wisconsin in October of 2012 to discuss and debate the status of the Mahā-
bhārata’s upākhyānas, especially as they influence our understanding of the
text as a work of literature. The contributions were not restricted to the sixty-
seven narratives Hiltebeitel identifies in his list as upākhyānas: many of the
participants in the Mahābhārata Upākhyāna Project took Hiltebeitel’s invita-
tion to consider the status of the Mahābhārata’s secondary narratives literally,
and identified other material that fit the category of “upākhyāna” understood
as a narrative that either elaborates on some aspect of the main narrative or
1 Alf Hiltebeitel, “Not Without Subtales: Telling Laws and Truths in the Sanskrit Epics,” Journal
of Indian Philosophy 33 (2005): 455–511.
2 For this reason, this article has been reprinted here as the first of the essays in the book. Along
with the concluding “The Geography of the Mahābhārata’s Upākhyānas,” this essay brackets
this collection of essays dedicated to furthering Hiltebeitel’s thinking on the epic.
3 Ibid., 479 and see p. 35 in this volume.
clarifies a specific dharma point.4 Thus, Greg Bailey discusses the Mārkaṇḍeya-
samāsyāparvan, a narrative unit of about forty-two chapters in the Āraṇya‑
kaparvan. In his words,
4 As will be seen, the Mahābhārata upākhyānas have a special relationship to its dharma dis-
course. In this sense, rather than being subtales they are “super”-tales in the sense of narratives
that function as powers function in mathematics: Ku. It is this special relation of Ku that, of
course, gives the Mahābhārata’s dharma discourse its special texture (Ku = D).
5 Gregory M. Bailey, “Introductory Notes on the Literary Structure of the Mārkaṇḍeyasamāsyā
parvan,” 127–28.
6 Robert P. Goldman, “On the Upatva of Upākhyānas: Is the Uttarakāṇḍa of the Rāmāyaṇa an
Upākhyāna of the Mahābhārata?” 69.
Introduction 3
7 Joydeep Bagchee, “The Epic’s Singularization to Come: The Śakuntalā and Yayāti
Upākhyānas in the Southern Recension of the Mahābhārata,” 84.
8 Ibid., 84, n. 6.
9 Bruce M. Sullivan, “The Tale of an Old monkey and a Fragrant Flower: What the Mahā
bhārata’s Rāmāyaṇa may Tell Us about the Mahābhārata,” 188.
10 See Alf Hiltebeitel, Rethinking the Mahābhārata: A Reader’s Guide to the Education of the
Dharma King (Chicago: University of Chicago Press, 2001), 20 and 167–69.
11 In fact, much of my own work has been dedicated to this idea, that is, to work out the
recognizable narrative patterns or units such as the eighteen-parvan architecture, the cir-
cular composition, the doubled beginning, the return to the beginning with the ascent of
the main characters to heaven in the Svargārohaṇaparvan, and so on that hint at a care-
fully conceived philosophical program. See Vishwa Adluri, “Frame Narratives and Forked
Beginnings: Or, How to Read the Adiparvan,” Journal of Vaishnava Studies 19, no. 2 (2011):
143–210 and “Literary Violence and Literal Salvation: Śaunaka Interprets the Maha
bharata,” Exemplar: The Journal of South Asian Studies 1, no. 2 (2012): 45–68, and see also
Vishwa Adluri, “Hermeneutics and Narrative Architecture in the Mahabharata,” in Ways
and Reasons for Thinking about the Mahabharata as a Whole, ed. Vishwa Adluri (Pune:
Bhandarkar Oriental Research Institute, 2013), 1–27.
4 Adluri
define other societies and their productions as inferior to Western canons, but
also discussing these perspectives and their implications. The values that sup-
ported aesthetic valuations, for example, were full of assumptions about the
existence of a single (classical) aesthetic and rationality and automatically
implied assessments about the capabilities of the societies that produced such
texts or works of art.”12 Against the proponents of “an accumulative Mahābhā
rata,” Wulff points to a major unifying theme of the epic: the world constructed
or inhabited by the epic poets, in which the macrocosm is understood as a dou-
ble of the microcosm and in which the difference between gods and humans
first comes into play. In contrast, Adheesh Sathaye invokes a museological
metaphor to characterize the Mahābhārata. He suggests that rather then
accept theories of random accretion and expansion over centuries, we might
look at the epic’s narratives as carefully “curated” exhibits within a larger pan-
Indian project. My own article focuses on how the Ambā-Upākhyāna of the
Udyogaparvan, narrated just prior to the commencement of the great battle,
elucidates a central aspect of the epic’s cosmological narrative: the primor-
dial conflict between the creative, outward force of śṛṣṭi, emanation, and the
absorptive, inward pull of laya, dissolution. Taking up Madeleine Biardeau’s
concept of the upākhyānas as “récit-miroir,” “mirror story” narratives that are so
called because they reflect the “proximate narrative circumstances of the princes
and princess to whom they are recited,”13 Adam Bowles examines a sequence of
three narratives in the Āpaddharmaparvan of the Mahābhārata. He finds that
the upākhyānas do not just reflect these circumstances, but much like two mir-
rors arranged in parallel, create an infinite series of reflections, all inviting us
to enter and to lose ourselves in reflection(s) on the epic’s central theme of the
proper dharma.14 Extending the concept of the “mirror story,” Sally Sutherland
goes on to ask in what way the Rakṣovaṃśa of the Vālmīki Rāmāyaṇa might be
reflective of the Rāmopākhyāna of the Mahābhārata (or vice versa). Nicolas
12 Fernando Wulff Alonso, “Supernatural Conflicts, Unanimities, and Indra in the Main
Story and Substories of the Mahābhārata,” 206–207.
13 Adam Bowles, “Reflections on the Upākhyānas in the Ā paddharmaparvan of the Mahāb
hārata,” 327.
14 “In the Mahābhārata, this discursive structure is doubled, since in the narrative frame
embedding Bhīṣma’s counsels, King Janamejaya is receiving precisely the same (or almost
the same) wise counsel from the brahman Vaiśaṃpāyana. (And we might ask in a manner
that poses the problem of the context of the Sanskrit Mahābhārata’s composition and
reception, who were the homological parities of these royal listeners?) The point, how-
ever, is to simply note that the stories told in the Ādhp, which in many cases are animal
fables, are ‘mirrors’ being held up to potentates ‘in’ the narrative, and most likely ‘outside’
of the narrative, too. And, if their narratives are not so closely tied to the ‘narrative’ of the
Mahābhārata, they still invite readings against their varied Mahābhārata contexts
together with reflections on the art of governance.” Ibid., 328.
Introduction 5
Dejenne’s contribution takes us back to the person probably responsible for the
contemporary surge of interest in the upākhyānas: Madeleine Biardeau, whose
work on the Nalopākhyāna Hiltebeitel credits with triggering his own reflec-
tions on the upākhyānas. T.P. Mahadevan, in contrast, prefers to look further
back: at the possible Vedic origins of the Mahābhārata’s composers, who he
thinks may have written themselves into the epic as the eponymous uñchavṛtti
brāhmaṇa of the Mudgala Upākhyāna. Two concluding pieces take us back to
the status of the upākhyānas in the Mahābhārata’s overarching project: Simon
Brodbeck’s “Upākhyānas and the Harivaṁśa,” which discusses how the various
theories of the Mahābhārata’s upākhyānas might apply to its khila or appen-
dix, the Harivaṁśa, and Alf Hiltebeitel’s “The Geography of the Upākhyānas,”
which looks at how the Mahābhārata might be using upākhyānas to “to con-
struct what Mikhail Bakhtin calls a ‘chronotope.’” More specifically, he asks,
“how do upākhyānas function in the epic with regard to cosmology—includ-
ing not only space but time, and time not just with regard to the main story’s
pasts and futures, but what the Mahābhārata might mean by its primary genre
term, itihāsa or ‘history’?”15 As Brodbeck’s contribution is the final article in
this sequence of essays that take a substantive look at specific upākhyānas,
it seems appropriate to take a closer look at his conclusions here. Summing
up his analysis of the Dhanya-Upākhyāna, the upākhyāna that in terms of
its placement in the epic is located at the extreme end from the Śakuntalā-
Upākhyāna, Brodbeck notes:
At the time Alf and I launched the Mahābhārata Upākhyāna Project, we could
neither have anticipated that the responses would be so positive nor, indeed,
the wealth of evidence that would be raised in support of the project’s thesis.
We had simply been formulating our ideas on the upākhyānas independently
of each other (though, of course, in dialogue) and thought that it would
be interesting to see what our colleagues thought about the topic. My own
work was based on the study of specific segments of the epic, segments I had
begun to identify as imbuing the epic with “‘Archimedean points’ for musings
about events with the ‘detachment’ implicit in wisdom.”17 These included the
Mudgala-Upākhyāna (3.246–47), the Jāpaka-Upākhyāna (12.189–93), and the
Uñchavṛtty-Upākhyāna (12.340–53) (as a trio of upākhyānas concerned with
the uñchavṛtti brahmans, the ultimate fate and message of the epic); the Rāma-
Upākhyāna (3.257–76) and the Nārāyaṇīye Hayaśira-Upākhyāna (12.335) (as a
duo of upākhyānas concerned with the themes of incarnation and descent);
and the Ambā-Upākhyāna (5.170–93) and the Tripura-Upākhyāna (8.24) (two
upākhyānas I found fascinating for their clues to the theology of the Goddess
and Rudra, and their roles in inaugurating the destruction of the Kurus). On
the basis of my reading of these upākhyānas, I had tentatively advanced the
hypothesis that
But how or why the other upākhyānas did so was an open question, requiring
wider investigations. Fortunately, thanks to the contributors to this volume, we
are now in a position to understand this issue more clearly. We are now begin-
ning to see, if not a complete upākhyāna map, at least the first outlines of one:
17 Vishwa Adluri, “The Divine Androgyne: Crossing Gender and Breaking Hegemonies in the
Ambā-Upākhyāna of the Mahābhārata,” 281.
18 Ibid., 280–81.
Introduction 7
a new way of navigating the epic, using not the so-called heroic epic as our
guide19 but the epic’s own musings about itself and its characters, provided
self-reflexively via its narration of its upākhyānas. Indeed, as I received and
edited the articles for this volume, I was struck by the similarity between the
contributors’ conclusions. In launching the project, Alf and I had been clear
that there were to be no dogmas. Everyone was free to interpret the topic as
they liked; everything was open to question (including the hypothesis of the
meaningfulness of the upākhyānas); there would be no guidelines for thinking.
Yet, surprisingly enough, almost all of the scholars who contributed to the
project arrived at fairly similar conclusions: (1) the upākhyānas are meaning-
ful; (2) their inclusion (or addition, if one prefers) in the epic follows a design;
(3) together, they yield an argument; and (4) this argument shows the Mahāb
hārata to be a highly self-conscious work of literature, a dharma text from its
inception and not a Kuru epic with didactic interpolations, as had long been
suspected.20 These conclusions were often extended beyond what Alf in his
2005 article had defined as strictly upākhyānas. It seems that all sorts of didac-
tic and philosophic, reflective, and recapitulating material was now being seen
as part of the “main” Kuru narrative, undermining the claim of this narrative to
being the main theme of the epic.
As the “last frontier in theorizing an integral Mahābhārata,”21 the upākhyā
nas thus compel our attention. They compel our attention both because of
what they contain in themselves and because they are part of what it means to
read the epic (and the coda “as a whole” is only necessary because so far so
19 This reading of the epic, long the backbone of Western scholarly approaches, has become
untenable in light of much new work, especially the reconstruction of its genesis under-
taken in Vishwa Adluri and Joydeep Bagchee, The Nay Science: A History of German Indol-
ogy (New York: Oxford University Press, 2014). While the idea of an original heroic epic
with later accretions may have given an earlier generation of Orientalists, unfamiliar with
and untrained in Indian philosophy, literature, and poetics as they were, a handle on this
gigantic epic, it is scarcely credible when present-generation scholars revive Lassen’s
hypothesis of invading white Aryans or Holtzmann’s hypothesis of an “inversion” in the
epic’s sympathies. Fortunately, the number of scholars still searching for, as Sukthankar
puts it, “the lost paradise of the primitive Kṣatriya tale of love and war,” is now in the sin-
gle digits and still dwindling. V.S. Sukthankar, On the Meaning of the Mahābhārata (Bom-
bay: Asiatic Society, 1957), 31.
20 For a brief overview of the roots of this suspicion, see Vishwa Adluri and Joydeep Bagchee,
“Introduction,” in Reading the Fifth Veda: Studies on the Mahābhārata, Essays by Alf Hilte-
beitel, vol. 1 (Leiden: E.J. Brill, 2011), xi–xxxvi and see also Vishwa Adluri and Joydeep
Bagchee, “Introduction,” in When the Goddess Was a Woman: Mahābhārata Ethnogra-
phies, Essays by Alf Hiltebeitel, vol. 2 (Leiden: E.J. Brill, 2011), xi–xxxiii.
21 Hiltebeitel, “The Geography of the Upākhyānas,” 428.
8 Adluri
many have tried to read the epic without bothering to read it as a whole22). If
Brodbeck’s observations are correct, the upākhyānas are important not only in
some abstract sense but specifically important to us, that is, to how we are to
interpret and “take” the narrative. It may be that the paradigmatic recipient of
the upākhyānas is not Yudhiṣṭhira (who, according to Hiltebeitel’s calculations,
is either the primary or secondary auditor of forty-nine of them), but the reader
of the Mahābhārata himself. Like Yudhiṣṭhira, he is meant to undergo a peda-
gogy in what it means to be “the Dharma king” and, like Yudhiṣṭhira, he is
meant to understand that the narrative of what happens to him may ultimately
be less important than the narratives of how he can take and what he should
think about the narrative of what befalls him. A Mahābhārata without its
upākhyānas would not only be an abridgment in the sense of being a shorter
version but it would also be an abridgment in the sense of being a Mahābhā
rata shorn of its pedagogic, philosophic, and transformative functions; it would
be an epic that had lost its stated purpose: of being in size and weight, a text
rivaling the four Vedas as a source of salvation.23
This volume, with its fourteen chapters arranged sequentially,24 of course
can only make a beginning with this project of reading the Mahābhārata as a
22 Reading the epic “as a whole” obviously does not imply that scholars have to have read the
whole epic or that they can rule out the possibility of surprise, of wonder, of bafflement, of
things that appear not to fit, and so on. If that were the case, then they would not be read-
ing the epic; at most, they would be rereading it. By “as a whole,” I simply mean the her-
meneutic principle that one has to assume the unity of the epic if one is to read it as a
literary work. Even if one wishes to maintain that certain sections were added later to the
epic, one still has to understand why just these additions in just these places were made
and this means, one still has to approach the epic as a literary unity. The task of under-
standing cannot be circumvented, as text-historical scholars have tried to do for so much
of their history, by saying, “well, then, they added these sections because they wished to
write their ideology (or theology or history or sectarian views) into the text, and they did
it here because it was convenient (or they were lazy, or they were cunning, or they were
careless and overlooked the possibility that eight hundred years later astute scholars from
Germany would identify the insertion and lay bare their subterfuge).”
23 See 1.1.208a–209c for the story of the Mahābhārata’s weighing against the four Vedas:
catvāra ekato vedā bhārataṁ caikam ekataḥ |
samāgataiḥ surarṣibhis tulām āropitaṁ purā |
mahattve ca gurutve ca dhriyamāṇaṁ tato ’dhikam ||
mahattvād bhāravattvāc ca mahābhāratam ucyate |
niruktam asya yo veda sarvapāpaiḥ pramucyate ||.
24 This is true only of the central eight contributions. The first two contributions, as men-
tioned, are theoretical and contexualizing essays. Then follow eight contributions from
the Śakuntalā-Upākhyāna (1.62–69) to the Kapota- (12.141–45) and Kṛtaghna-Upākhyānas
Introduction 9
work of both argument and design, namely, as literature. It offers the reader a
first orientation to the Mahābhārata’s upākhyānas, their placement, their
names, their significance, and how they tie up with each other. But beyond
that, there are many things that remain to be discovered about the upākhyā
nas: how they relate to the epic’s stated project of being a dharma and a mokṣa
text,25 how they make it possible for the Mahābhārata to claim encyclopedic
comprehensivity,26 how they enable the Mahābhārata to expand in its various
local incarnations while still retaining the same basic narrative architecture,27
and so on. After the failure of the oral epic and Indo-European approaches to
the Mahābhārata, which arose from the German Indologists’ need to posit a
special proximity between their country and ancient India that would confirm
their status as privileged interpreters and translators of Indian Antiquity,28 the
upākhyānas offer the best approach for rethinking what the Mahābhārata is
and how it might live up to its claim to containing encyclopedic wisdom—the
mataṁ kṛtsnaṁ (Mahābhārata 1.55.2c; 1.56.12c), as it likes to call itself, of the
seer Vyāsa. It is with this hope in mind that we, the thirteen contributors and
participants in the Mahābhārata Upākhyāna Project, would like to present this
volume to the reader.
Chapter 1
This chapter on India’s two Sanskrit epics, the Mahābhārata and Rāmāyaṇa,
will address four topics:1 how they have been defined by scholars and by them-
selves; how each conceptualizes the relationship between its whole and its
parts, and particularly its subtales;2 how subtales figure in their main stories;
and how each creates grand narrative out of this configuration. This article
favors the priority of the Mahābhārata and will be presented from that
standpoint.3
The Mahābhārata describes itself as “sprung from the oceanic mind (rnanaḥ
sāgarasambhūtām)” (1.53.34a) of its author Vyāsa and to be his “entire
thought” (1.1.23; 1.55.2) in a text of a hundred thousand couplets (ślokas)
(1.56.13). Although no known edition reaches that number, when the Mahāb
hārata describes texts of that size it denotes their originary vastness. As one
lost prototype,4 it mentions that aeons ago, seven sages known as the Citra
śikhaṇḍins, “having become of a single thought, promulgated5 a supreme
1 Many of the ideas in this chapter, along with fuller synopses of both epics, appear as separate
entries on each epic in Stanley Wolpert, ed., Encyclopedia of India (New York: Scribners/
Macmillan, 2006). I thank the editor for permission to develop that material further for this
article.
2 I will favor the translation “subtale” for upākhyāna, with perhaps a hint of subtext.
3 I agree with Madeleine Biardeau’s chronological positioning of the Mahābhārata as older than
the Rāmāyaṇa (Madeleine Biardeau, Le Mahābhārata: Un récit fondateur du brahmanisme et
son interprétation, vol. 1 [Paris: Éditions du Seuil, 2002], 700–701 and ff., 726; Le Rāmāyaṇa de
Vālmīki [Paris: Gallimard, 1999], xxxiii–xxxv, though I see a shorter time between them. See
Alf Hiltebeitel, Rethinking the Mahābhārata: A Reader’s Guide to the Education of the Dharma
King (Chicago: University of Chicago Press, 2001), 15–31, 165; see also Alf Hiltebeitel, “The
Emergence of Kāvya out of the Two Sanskrit Epics,” paper presented at the International
Seminar on the Origins of Mahākāvya, Università degli Studi di Milano, June 4–5, 2004.
4 It was destined to be lost after the golden age reign of King Vasu Uparicara (12.322.48).
5 For proktam, see Christopher Z. Minkowski, “Janamejaya’s Sattra and Ritual Structure,” Journal
of the American Oriental Society 109, no. 3 (1989): 402, 411–12 concerning pra + vac/.proktaḥ as
having Vedic overtones, with “the sense of an original utterance”; cf. Hiltebeitel, Rethinking
the Mahābhārata, 98–99.
6 “Loom” for tantra, or more prosaically, perhaps, “course.”
7 The Pune Critical Edition has 1995 “chapters” or “lessons” (adhyāyas).
8 Lévi seems to suggest that the designation “hundred thousand-verse” “had been consecrated
since the fifth century,” citing its appearance on a fifth century inscription. But that inscription
would be quoting Mahābhārata 1.56.13 and 12.331.2, where the claim is made in the epic itself.
There is no reason to think that the Mahābhārata number was reached only by the fifth cen-
tury and announced at that point.
9 Sylvain Lévi, “Tato Jayam Udirayet,” trans. L.G. Khare, Annals of the Bhandarkar Oriental
Research Institute 1 (1917): 13–20 (translation slightly modified). Lévi’s Prajñāpāramitā se-
quencing is uncertain; see Edward Conze, The Prajñāpāramitā Literature (The Hague: Mouton,
1960). My view would be that his instinct is likelier for the Mahābhārata, with which the
Buddhist texts would have “caught up.” On Lévi’s hypothesis, the Śatasāhasrikā would have
“lent itself to this process without difficulty: it was only necessary to cut down the tautologies,
the redundancies, and the repetitions which swelled it and gave it an enormous bulk …. In its
steady effort for the beautiful [or perfect], India has manifestly passed through the intermedi-
ate stage of the colossal. Before relishing and realizing the beauty that consists in a harmoni-
ous equilibrium of lines, the mind of man first permits itself to be carried away by mere mass.”
Lévi, “Tato Jayam Udirayet,” 19. Evolutionary claims aside, Lévi’s conclusion could be said to
anticipate what this article has to say about the more local stance of Vālmīki vis-à-vis Vyāsa.
12 Hiltebeitel
grown from oral origins into a massive “encyclopedia”—a text of such monu-
mental self-sufficiency that it could have considered itself to have absorbed
everything that would have haphazardly come its way as a reflex of its snowball
descent through centuries.10 Many such scholars cite another verse in support
of this theory, which says that Vyāsa “composed a Bhārata-collection [saṃhitā]
of twenty-four thousand couplets without the subtales [upākhyānair vinā]; so
much is called Bhārata by the wise” (1.1.61).11 Although a hundred thousand-
verse Bhārata is also mentioned (12.331.2), translators have sought to help the
developmental argument along by adding that Vyāsa composed this shorter
version “first”12 or “originally.”13 But the verse says nothing about anything com-
ing first. Since “without” implies a subtraction, and since the passage describes
Vyāsa’s afterthoughts, the twenty-four thousand-verse Bhārata would probably
10 This approach gained authoritative status in Edward Washburn Hopkins, The Great Epic
of India: Its Character and Origin (Calcutta: Punthi Pustak, 1969; first published in 1902 by
Charles Scribner’s Sons) and has recently been slightly refined (bardic background, a
post-Aśokan first written redaction, and a normative redaction under the Guptas yielding
a “library”) in James L. Fitzgerald, “Mahābhārata,” in The Hindu World, ed. Sushil Mittal
and Gene Thursby (New York: Routledge, 2004), 52–55, 68–70. For counter-arguments, see
Hiltebeitel, Rethinking the Mahābhārata: that the epic would have been composed over a
shorter period (1–31), with nothing required from the Guptas (ibid., 25–26); and that the
term “encyclopedia” has been misleadingly applied to the Mahābhārata, particularly with
reference to the “Whatever is here may be found elsewhere …” verse, which, rather than
defining the exhaustiveness of the text, is pitched toward an “ontological debate” (162–
63).
11 Sukthankar illustrates the lengths to which scholars have gone in fitting the 24,000 verse
“Bhārata without upākhyanas” into their theories of the text, notably his theory of
Bhṛguization: “in my opinion we should have no hesitation in concluding that in our ver-
sion of the Mahābhārata there is a conscious—nay deliberate—stitching together of the
Bhārata legends with the Bhārgava stories (author’s italics). The question how precisely
this Bhārgava element, which we find concentrated mostly in the upākhyānas came into
the cycle of the Bhārata legends … is largely a matter of speculation. Even according to the
traditional view, it was not the work of Vyāsa, the reputed author of the Mahābhārata,
because the diaskeuasts have been fortunately frank enough to admit that his work, the
Bhārata, which originally consisted merely of 24,000 stanzas, had no episodes to speak of.”
V.S. Sukthankar, “Epic Studies VI: The Bhṛgus and the Bhārata: A Text-Historical Study,”
Annals of the Bhandarkar Oriental Research Institute 18, no. 1 (1936): 70. There is nothing
demonstrably “traditional” about this view of Vyāsa, and the frankness of the diaskeuasts
is a fancy. On “Bhṛguization,” see Hiltebeitel, Rethinking the Mahābhārata, 105–18.
12 J.A.B. van Buitenen, “Introduction,” in The Mahābhārata, vol. 1. 1. The Book of the Beginning
(Chicago: University of Chicago Press, 1973), 1 and 22; cf. Aurobindo Ghosh, “Notes on the
Mahābhārata,” in On the Mahabharata (Pondicherry: Sri Aurobindo Ashram, 1991), 9.
13 Kisari Mohan Ganguli, trans. The Mahabharata, vol. 1 (New Delhi: Munshiram Manohar-
lal, 1970 [1884–96]), 6.
Not without Subtales 13
14 David Shulman, “Toward a Historical Poetics of the Sanskrit Epics,” in The Wisdom of
Poets: Studies in Tamil, Telugu, and Sanskrit (Delhi: Oxford University Press, 2001), 25, see-
ing the fallacy of arguing that the 24,000-verse “Bhārata” came first, takes the verse to
imply that Vyāsa “compressed” the 100,000 verses “by eliminating the various minor sto-
ries (upākhyāna) into a mere 24,000 verses.” But first credit on this recognition may go to
Sastri (see P.P.S. Sastri, ed. The Mahābhārata (Southern Recension) Critically Edited by
P.P.S. Sastri [Madras: Ramaswamy Sastrulu & Sons, Madras, 1931], x–xi), who states that
the verses in question in his attempt to critically edit the Southern Recension “unmistak-
ably state that the computation of the Mahābhārata is 100,000 verses if counted along
with the minor narratives, and only 24,000 verses if the minor narratives are excluded. It
is not at all meant that 24,000 verses alone were originally composed and the remaining
76,000 verses were added later to complete the 100,000 verses.”
15 Fitzgerald still asks “at what stage was a putative Bhārata story recast as the ‘Mahā’
Bhārata …?” (“Negotiating the Shape of ‘Scripture’,” 272), and in a footnote to this continu-
ing question cites “the purported historical observation of Vyāsa at 1.1.61 … [as] evidence
that at least some in ancient India distinguished larger and smaller versions of this epic”
(ibid., 272, n. 19). While not claiming directly the shorter Bhārata’s precedence, he implies
it (it would be no use to his developmental argument were it not earlier). Suggesting by
the move to a footnote that this Bhārata might be the same as “a putative Bhārata story”
that would have this priority over the Mahābhārata, he ignores the question of what kind
of text it would be “without the upākhyanas.” But that element of the verse in question is
where one must begin (as Sukthankar did, claiming that the original “had no episodes to
speak of”; see n. 11 above) in considering how the epic “distinguished larger and smaller
versions of” itself (rather than gratuitously attributing the distinction to “some in ancient
India,” implying “some Brahmans,” whom Fitzgerald frequently invokes with the sugges-
tion that sociological divisions would have produced the textual variations in which he
usually sees additions). Fitzgerald tries to relate 1.1.61 to his argument for centuries of
sequential development in the composition of Books 12 and 13, and to explain “sectors” of
the Bhārgava Rāma dossier by associating the Vaiśaṁpāyana and Ugraśravas frames with
two of “four, or more, distinct poetic or redactional efforts” (James L. Fitzgerald, “The
Rama Jāmadagnya ‘Thread’ of the Mahābhārata: A New Survey of Rāma Jāmadagnya in
the Pune Text,” in Stages and Transitions: Temporal and Historical Frameworks in Epic and
Purāṇic Literature, Proceedings of the Second Dubrovnik International Conference on
the Sanskrit Epics and Purāṇas, Aug. 1999, ed. Mary Brockington [Zagreb: Croatian Acad-
emy of Sciences and Arts, 2002], 115, 99–100, 104–7, 112–13)—an opportunistic (I believe)
association between frames and redactions also made by Sukthankar and some others
(see Hiltebeitel, Rethinking the Mahābhārata, 105, n. 47 and above n. 11).
14 Hiltebeitel
a double “etymology” (nirukta) for one and the same huge text.16 Yet despite
nothing surviving of this shorter Bhārata, scholars have used it to argue for an
originally oral bardic and heroic story that would have lacked not only subtales
but frame stories, tales about the author both in the frames and elsewhere,
didactic additions, and devotional passages with “divinized” heroes. Some have
assumed that Kṛṣṇa would have been “divinized” before the introduction of
still “later” passages glorifying Śiva and even the Goddess; and there were even
those who wanted to argue that Kṛṣṇa was not original to the earliest bardic
version. Although these ideas dominated Western scholarship only in the late
nineteenth and early twentieth centuries, they are still vigorous. It is, however,
no longer possible to find a serious scholar who wants to argue for an originally
Kṛṣṇa-less epic, and there are now those who see principles of ongoing design
as guiding the Mahābhārata’s agglutination rather than historical accident.
New developments have thus complicated this profile. These include inter-
textual studies positioning the Mahābhārata in relation to both Indo-European
and Indian texts; genre study, including the history of kāvya or Sanskrit “poetry”
composed according to classical aesthetic norms; debate on the likely period
of the Mahābhārata’s composition in written form; and the completion of the
Pune Critical Edition, along with wider recognition of the Mahābhārata’s
design. For a notable result of the Mahābhārata’s Pune Critical Edition is its
establishment of a textual “archetype.” There remains debate as to whether
this archetype takes us back to the text’s first composition, or to a later redac-
tion that would put a final stamp on centuries of cumulative growth. This essay
favors the first option. In either case, this archetype includes a design of eigh-
teen Books or parvans,17 nearly all the epic’s one hundred “little books” or
16 Cf. 1.56.31: “The Mahābhārata, they say, is the great Birth of the Bhāratas (bharatānām
mahoj janma): he who knows this etymology [nirukta] is rid of all his sins.”
17 Schlingloff’s contrary claims (Dieter Schlingloff, “The Oldest Extant Parvan-list of the
Mahābhārata,” Journal of the American Oriental Society 89 [1969]: 334–38) about the
Mahābhārata’s “oldest extant parvan-list” based on the Kuṣāṇa period “Spitzer manu-
script” found in east Turkestan have been revived by Franco (Eli Franco, The Spitzer Man-
uscript: The Earliest Philosophical Manuscript in Sanskrit, 2 vols. [Vienna: Verlag der
Österreichischen Akademie der Wissenschaften, 2004]), with some additional informa-
tion and suggestions: that it may have come from “the Great Gandhara area” and been
written using a broad-nibbed copper pen (vol. 1, 11); that it is probably a Sarvāstivādin text
(19) from “around the second half of the third century” (33); that it included a refutation
of God in one fragment (18–19); and that its reference to some Mahābhārata units and
brief encapsulation of the Rāmāyaṇa “may have been occasioned by a discussion of the
Buddha’s omniscience” (17). If the last two things are true, it hardly seems that the Bud-
dha’s omniscience was directed toward the “extant” totality of either epic. Indeed, not
Not without Subtales 15
knowing the context, we cannot know what the units were listed for, why both parvans
and subparvans were selected, why in some cases they are apparently listed out of
sequence and in others with one inclusive of another, why the Mahābhārata is digested by
(selected) components and the Rāmāyaṇa as a (minimalist) consecutive narrative, or
even that the four fragments mentioning these features were all on the same page. No
Mahābhārata scholar using the find as evidence of a once-shorter text (see John L. Brock-
ington, The Sanskrit Epics [Leiden: E.J. Brill, 1998], 131–32; Fitzgerald, “Negotiating the
Shape of ‘Scripture’,” 270–71, nn. 15 and 17 has tried to explain what kind of “Bhārata” it
would have been with the odd assortment of units mentioned. With such uncertainties,
notions that the Virāṭā- and Anuśāsana-Parvans would not yet have been extant (Schlin-
gloff, “The Oldest Extant Parvan-list of the Mahābhārata,” 338; Franco, The Spitzer Manu-
script, vol. 1, 10; Brockington and Fitzgerald as cited) must be taken cum grano salis.
Regarding Book 4, the only evidence is that no Virāṭaparvan is mentioned between a unit
beginning with a or ā, for which Schlingloff (“The Oldest Extant Parvan-list of the Mahāb
hārata,” 338) proposes a(raṇeyaṃ) “or perhaps a(jagara)”—both subparvans of Book 3—
and (ni)ryyaṇaṃ for the Abhiniryāṇa subparvan of Book 5. But a could provide
a(jñātavāsa), the “residence incognito” widely used to describe the Virāṭaparvan (see Alf
Hiltebeitel, “Śiva, the Goddess and the Disguises of the Pāṇḍavas and Draupadī,” History
of Religions 20 [1980]: 148, n. 4), or a(bhimanyu-vivāha), the main adhyāya name (4.66–67)
in Book 4’s concluding subparvan.
18 Robert P. Goldman and Sally Sutherland Goldman, trans. Bālakāṇḍa, vol. 1 of The
Rāmāyaṇa of Vālmīki: An Epic of Ancient India, ed. Robert P. Goldman and Sally Suther-
land Goldman (Princeton, NJ: Princeton University Press, 1984), 63.
19 G.H. Bhatt and U.P. Shah, eds., The Vālmīki Rāmāyaṇa: Critical Edition. 7 vols. (Baroda: Ori-
ental Institute, 1960–1975).
20 Sheldon I. Pollock, “Ātmānam mānuṣaṃ manye: Dharmākūtam on the Divinity of Rāma,”
Journal of the Oriental Institute of Baroda 33 (1984): 505–28.
16 Hiltebeitel
21 Robert P. Goldman, trans. Sundarakāṇḍa, vol. 5 of The Rāmāyaṇa of Vālmīki, ed. Robert P.
Goldman and Sally Sutherland Goldman (Princeton, NJ: Princeton University Press,
1996).
22 Sheldon I. Pollock, trans. Ayodhyākāṇḍa, vol. 2 of The Rāmāyaṇa of Vālmīki: An Epic of
Ancient India, ed. Robert P. Goldman (Princeton, NJ: Princeton University Press, 1986),
38–42.
23 Madeleine Biardeau, “Some Remarks on the Links between the Epics, the Purāṇas and
their Vedic Sources,” in Studies in Hinduism: Vedism and Hinduism, ed. Gerhard Oberham-
mer (Vienna: Verlag der Österreichischen Akademie der Wissenschaften, 1997), 77–119.
Not without Subtales 17
and that the Rāmāyaṇa poet is not only familiar with the Mahābhārata’s design
but intent upon refining it.
Such a relation can be exemplified by the two epics’ frame stories, which are
opened at the beginning of the first Books and left pending into the denou-
ments. In both epics the frames are three-tiered. In the Mahābhārata, there are
in fact three frames. Initially, Vyāsa recites the epic to his five Brahman disci-
ples, first to his son Śuka and then to the other four, including Vaiśampāyana
(Mahābhārata 1.1.63). Second, Vaiśaṁpāyana recites it at Vyāsa’s bidding to
King Janamejaya, a descendant of the Pāṇḍavas, at his snake sacrifice so that
he can hear the story of his ancestors. And third, the Bard Ugraśravas, who
overheard Vaiśaṁpāyana’s narration, brings it to Śaunaka and the Ṛṣis of the
Naimiṣa Forest and recites it there in 18 Books (2.3.71).24 Unlike the Mahābhā
rata’s three frame stories, which present a serial layering of the first three
recitals of supposedly the same text that are scattered over its first 56 chapters
and resumed in late portions of its twelfth Book, the Rāmāyaṇa frame, in only
its first four chapters known as the upodghāta or preamble, presents two pro-
gressive unfoldings of the story—the first by the sage Nārada to the hermit
Vālmīki; the second by Vālmīki himself, now a poet—that trace its ripening
into the third full unfolding, the Vālmīki Rāmāyaṇa itself.
In the first, in answer to Vālmīki’s opening question whether there is an
ideal man in the world today (Rāmāyaṇa 1.1.2–5), Nārada satisfies the question
with a brief and entirely laudatory account of Rāma’s virtues and adult life,
presumably to date (1.1.7–76). Saying the minimum about Rāma’s killing of the
monkey Vālin (1.1.49, 55), Nārada hardly hints at anything problematic in
Rāma’s life and omits both Sītā’s fire ordeal and her banishment. Among the
great Ṛṣis or seers Rāma encounters, he mentions only Vasiṣṭha (29) and
Agastya (33–34).
In the second sarga, once Nārada has left, Vālmīki witnesses the cries of
grief of a female Krauñca bird (probably the large monogamous sarus crane
over the slaying of her mate by a “cruel hunter,”25 and is provoked into the
spontaneous utterance that creates “verse” (and thus poetry) out of “grief”
(śloka out of śoka; 1.2.9–15). As this verse is said to mark the origins of poetry,
the Rāmāyaṇa is called the ādikāvya or “first poem”—a term that does not
occur in the Baroda Critical Edition, though it probably should since it occurs
24 The eighteen parvans are mentioned again at 1.2.244ab toward the end of the Parvasaṃ
graha or “Summaries of the Books”—as if to say that what we get is this Naimiṣa Forest
edition.
25 See Julia Leslie, “A Bird Bereaved: The Identity and Significance of Vālmīki’s Krauñca,”
Journal of Indian Philosophy 26 (1998): 455–87.
18 Hiltebeitel
in a universally attested sarga where, after Sītā has vanished into the earth, the
god Brahmā encourages Rāma to hear the rest of this ādikāvya (7, Appendix I,
No. 13, lines 31–39). Now, however, the same Brahmā appears (22–36) to prompt
Vālmīki to tell the story he has just heard from Nārada, and gives him the
insight to see what he did not know and what is still yet to happen—with,
moreover, the confirmation that his poem will endure so long as the rivers and
mountains last on earth and that it will all be true (1.2.33–35). Brahmā thus
assures Vālmīki that he will know things omitted from Nārada’s encomium.
Upon Brahmā’s vanishing, Vālmīki now conceives the idea of composing “the
entire Rāmāyaṇa poem [kāvya] in verses such as these” (1.2.40d), that is, such
as the śloka he has just uttered.
In the third sarga, Vālmīki meditatively enters into this project for the first
time (1.3.2) and has a sort of preview of the story (3–28): not a retrospective
table of contents like the Parvasaṃgraha (PS)—the lengthy “Summaries of the
Books” that forms the Mahābhārata’s second adhyāya and second upapar-
van—but a kind of first glimpse and unfolding of what his poem will contain.
Here he provides the first reference to some of Rāma’s encounters with impor-
tant Ṛṣis (he will hear Viśvāmitra’s stories [4], face Rāma Jāmadagnya [5], and
hear Bharadvāja’s instructions [8]). Most important, while adding nothing
problematic on the slaying of Vālin (15–16) and without having mentioned
Sītā’s fire ordeal, he closes with Sītā’s banishment (28).
Then, looking back upon the poem’s completion, the fourth sarga hints at
the context in which Vālmīki’s Rāmāyaṇa will finally be told by the twins Kuśa
and Lava to their father Rāma. Just as information on the Mahābhārata’s frame
is resumed with further revelations about Vyāsa, Śuka and his co-disciples, and
the Naimiṣa Forest is in Book 12,26 the Rāmāyaṇa’s frame will be picked up in
Book 7 when Kuśa and Lava do just that: pick up the frame in the Naimiṣa
Forest (Rāmāyaṇa 7.82,14b)!27 The main difference is that when the Rāmāyaṇa
frame is reentered in Book 7, it is not just a matter of further revelations about
the composition that are difficult to relate to the main story. Vālmīki’s dramatic
entry into the main story presents the occasion to reveal the poetic heart of the
whole poem through its effects on its hero and its heroine. Nonetheless, in
both epics there is a moment where the author emerges from the frames to
speak directly to the epic’s main listener. In the Rāmāyaṇa this occurs at this
26 In stories about Śuka including the Śukotpatti or “Birth of Śuka” (Mahābhārata 12.310–20),
and in the Nārāyaṇīya (12.321–39); see the final section of this chapter “Upākhyāna Prece-
dence and the Essence of them All.”
27 The site where the twins recite the Rāmāyaṇa at Rāma’s aśvamedha sacrifice. On the coin-
cidence of the sites for the third narrations, which suggests a nod to Mahābhārata prece-
dence, see Hiltebeitel, Rethinking the Mahābhārata, 285–86.
Not without Subtales 19
climactic moment when Vālmīki addresses Rāma and confirms Sītā’s veracity
before she enters the earth. In the Mahābhārata, in a much less noticed but still
quite dramatic passage, it occurs in the one time that Vyāsa addresses Jana
mejaya directly to tell the subtale (upākhyāna) of the Horse’s Head in answer
to a culminating question of the Nārāyaṇīya.28
Vālmīki thus gets a triple inspiration—from Nārada, the krauñcī, and
Brahmā. Yet the upodghāta leaves us in suspense as to when Sītā came to his
ashram. Was it before or after the Krauñca bird incident? The poem never tells
whether Vālmīki’s response to the female bird comes before or after his famil-
iarity with Sītā’s grief at her banishment. But in either case, now that Vālmīki
knows the whole story from Brahmā, he could connect Sītā’s banishment with
the cry of the krauñcī whenever she arrived. What we do know is that, having
had pity (karuṇā, Rāmāyaṇa 1.2.11d) for the female bird, Vālmīki will compose
his poem with pity as its predominant aesthetic flavor (aṅgīrasa) in relation to
grief (śoka) as its underlying sthāyibhāva or “stable aesthetic emotion.” The
Mahābhārata provides no such developmental inspiration story for its author
Vyāsa, although I believe the father-son story of Vyāsa and Śuka is its analogue.29
Even though Ugraśravas seems to tell the Pauloma and Āstīka subparvans on
his own, there is no suggestion that they are anything but the “entire thought”
of Vyāsa, and there is no hint at any growth process either in the poet’s mind or
in the performances by either of the narrators.30 The Rāmāyaṇa frame is thus
shorter, more developmental, more focused, and more poetically traceable
into the main narrative and the whole poem.
Indeed, once past the upodghāta, the Rāmāyaṇa’s main story begins imme-
diately with a brief praise of the Rāmāyaṇa itself and the Ikṣvāku dynasty
(Rāma’s ancestors) that quickly narrows down to the country of Kosala, its
capital city of Ayodhyā, and the current reign there of Rāma’s father Daśaratha
(Rāmāyaṇa 1.5.1–9), all presumably as it was composed by Vālmīki and imparted
to be recited by Kuśa and Lava to Rāma. So it continues to its end—again,
28 See Alf Hiltebeitel, “The Nārāyaṇīya and the Early Reading Communities of the Mahābhā
rata,” in Between the Empires: Society in India, 300 bce to 400 CE, ed. Patrick Olivelle (New
York: Oxford University Press, 2006), 227–56.
29 Both include poignant bird stories; see Hiltebeitel, Rethinking the Mahābhārata, 279–322
and Alf Hiltebeitel, “More Rethinking the Mahābhārata: Toward a Politics of Bhakti.” Indo-
Iranian Journal 47 (2004): 203–27. As in the Rāmāyaṇa, the Śuka story is presented in a
way that appears disjointed from the main story, and in the Mahābhārata’s denouement
rather than its preamble. It is thus much more difficult to trace into the main story.
30 See Hiltebeitel, Rethinking the Mahābhārata, 99–105). Ugraśravas recital in eighteen par-
vans (see above at n. 16) would not add anything but rather be his way of arranging the
“whole” to meet the sacrificial timetable of the Naimiṣa Forest Ṛṣis’s 12-year sattra
(1.1.1–2).
20 Hiltebeitel
unlike the Mahābhārata, which has the overriding device of presenting its
multiple frame stories as intertwining dialogues between its narrators and
their listeners.31 Although the upodghāta concludes with Rāma, as chief-audi-
tor-to-be, inviting his brothers to join him in listening to Kuśa and Lava, whom
he is yet to recognize as his sons, he interrupts their narration to question them
only once: when, having listened for some time, he asks them who authored
this poem (kāvya) (7.85.19). Otherwise, until he recognizes them soon after this
and wants to see their banished mother (86.2–6), he is the rapt and silent lis-
tener. Yet note the concluding words of the upodghāta with which he launches
their recital: “Moreover, it is said that the profound adventure [mahānubhāvaṃ
caritam] they tell is highly beneficial even for me. Listen to it” (1.4.26d). Who
has said this? Why beneficial to Rāma? The preamble leaves us with such
implicit and subtle questions. The point seems to be that listening to Vālmīki’s
poem will awaken Rāma to recall Sītā after he has banished her.
In these passages, we see two of the three leading terms by which the Rāmāyaṇa
describes itself: kāvya (poem) and carita (adventure), the third being ākhyāna
(tale, narrative). Let us look at how these and other terms are used by each epic
to identify itself and to define the relationship of its whole to its parts.
Most frequently, the Mahābhārata characterizes itself fourteen times as a
“narrative” (ākhyāna; Mahābhārata 1.1.16a; 1.2.29b, 235c, 238a, 239b, 240b, and
241b; 1.53.31d and 32a; 1.56.1c, 30c, 32c; 12.337.10a; 18.45.53a) and eight times as a
“history” (itihāsa; 1.1.17a, 24d, 52c; 1.2.237a; 1.51.16c; 1.56.18c and 19a; 1.93.46c).32
But it also calls itself a work of “ancient lore” (purāṇa; 1.1.15b; 1.56.15d), a “story”
(kathā; 1.56.2a), a “collection” (saṃhitā; 1.1.19.1c and 61b), a “fifth Veda” (1.57.74ab;
12.327.18ab), the “Veda that pertains to Kṛṣṇa” (Kārṣṇa Veda, probably referring
primarily to Kṛṣṇa Dvaipāyana Vyāsa—1.1.205a; 1.56.17c), a “great knowledge”
(mahaj-jñāna; 1.1.25b and 49a), a “treatise” (śāstra; 1.56.21; indeed, in this verse,
a dharṃaśāstra, arthaśāstra, and mokṣaśāstra; and probably 12.238.13c),33 an
upaniṣad (1.1.191a), a “biography” or “adventure” (carita; 1.56.1d),34 a “victory”
they were once at the heart of long debates centered on an “ākhyāna theory” of
the origins of Vedic poetry itself.39 The one non-Vedic exception seems to be
upākhyāna—a term that may have been given its first life by the authors of the
Mahābhārata.40 They present a topic whose significance—for both epics—
has not been sufficiently appreciated.41
Walter de Gruyter, 1996), 197–98 on early appearances of ākhyāna, itihāsa, and purāṇa in
the Aitareya and Śatapatha Brāhmaṇas, with the use of itihāsa to interpret Ṛg Vedic
saṃvāda hymns, leading to an aitihāsika “school” of interpretation; mention of ākhyāna
vids as “those who know the stories”; Chāndogya Upaniṣad 7.1.2 where itihāsapurāṇa
refers to a fifth Veda; and 202 on the question of the coherence of Ṛg Vedic saṃvādas
independent of ākhyānas and itihāsas. See now also Stephanie Jamison, The Rig Veda
between Two Worlds. Le Ṛg Veda entre deux mondes (Paris: Boccard, 2007), 120–50 on
Ṛgvedic usages of kāvya in relation to later ones.
39 See Patton, Myth as Argument, 195–214.
40 That is, as far as I can ascertain it is a non-Vedic term: see Monier-Williams (Monier
Monier-Williams, A Sanskrit-English Dictionary [Oxford: Clarendon Press, 1964], 212), cit-
ing nothing earlier than the Mahābhārata. Barbara Gombach introduces some uncer-
tainty here. While positing throughout her dissertation that the Mahābhārata’s “ancillary
stories make the epic a Veda” and “Vedicize” the main story—as if the main story were not
filled with Vedic allusions itself—she lists upākhyāna among terms “known from earlier
Vedic literature” along with itihāsa, ākhyāna, gāthā, and saṃvāda. Barbara Gombach,
“Ancillary Stories in the Sanskrit Mahābhārata” (PhD diss., Columbia University, 2000),
part 1, 345 and 346. But this is without citation. Gombach has done immensely valuable
work in mapping the number and variety of the epic’s interior sub-narratives, but the
term “ancillary” carries for her the general imputation of “addition” and “interpolation”
(ibid., 24, 184, 319) through “centuries of compilation” (ibid., 302), in particular with the
suggestion that the clustering of stories in Books 1, 3, 12, and 13 “might help explain differ-
ent degrees of interpolation” (ibid., 24). All this is said while granting “that some of the
ancillary material was inspired by the epic itself” (ibid., 165).
41 Sukthankar partially tracked the Mahābhārata’s upākhyānas (those that have something
to do with Bhārgavas) with the assumption that upākhyānas are not only “episodes” but
“digressions” (Sukthankar, “Epic Studies VI: The Bhṛgus and the Bhārata,” 14, 17, 33, 35, 44,
65; see n. 11 above); Dange (Sadashiv Ambadas Dange, Legends in the Mahābhārata [Delhi:
Motilal Banarsidass, 1969]) brings a Frazerian comparative folklore approach to many
upākhyānas among the legends he discusses; Van Buitenen (1975, 111, 182) also introduces
the “episodes” in Book 3 as more “pious” than those in Book 1, and “extraneous to the main
story” yet “preserved in the library that is The Mahābhārata” because they each have
“their own interest”—a view he extends elsewhere, as will be noted. Sri Aurobindo
claimed to be able to identify two upākhyānas—Sāvitrī and Nala—as Vyāsa’s by “the ulti-
mate test of style.” Ghosh, “Notes on the Mahābhārata,” 12, 44-54. Richest are Jamison’s
(Stephanie Jamison, Sacrificed Wife, Sacrificer’s Wife: Women, Ritual and Hospitality in
Ancient India [New York: Oxford University Press, 1996]) and Parida’s (Sarat Chandra
Parida, Hospitality in Changing Indian Society: Vedic Age to Puranic Age [Delhi: Bharatiya
Vidya Prakashan, 2004], 47–172) treatments of numerous upākhyānas mainly around the
theme of hospitality: Jamison discussing nineteen of those listed below (numbers 1, 2, 6,
Not without Subtales 23
1. Śakuntalā-Upākhyāna 1.62–69
2. Yayāti- Upākhyāna 1.70–80
3. Mahābhiṣa-Upākhyāna 1.91
4. Aṇimāṇḍavya-Upākhyāna 1.101
5. Vyuṣitāṣva-Upākhyāna 1.112
6. Tapatī-Upākhyāna 1.160–163
7. Vasiṣṭha-Upākhyāna 1.164–68, 173
8. Aurva-Upākhyāna 1.169–172
9. Pañcendra-Upākhyāna 1.189
10. Sunda-Upasunda-Upākhyāna 1.201–4
11. Śārṅgaka-Upākhyāna 1.220–25
12. Saubhavadha-Upākhyāna 3.15–23
13. Nala-Upākhyāna 3.50–78
14. Agastya-Upākhyāna 3.94–108
15. Ṛśyaśṛṅga-Upākhyāna 3.110–13
16. Kārtavīrya/ Jāmadagnya-Upākhyāna 3.115–17
17. Śukanyā-Upākhyāna 3.122–25
18. Māndhātṛ-Upākhyāna 3.126
19. Jantu-Upākhyāna 3.127–28
20. Śyena-Kapotīya-Upākhyāna 3.130–31
21. Aṣṭāvakrīya-Upākhyāna 3.132–34
22. Yavakrīta-Upākhyāna 3.135–39
23. Vainya-Upākhyāna 3.183
24. Matsya-Upākhyāna 3.185
25. Maṇḍūka-Upākhyāna 3.190
26. Indradyumna-Upākhyāna 3.191
27. Dhundhumāra-Upākhyāna 3.192–95
28. Pativratā-Upākhyāna 3.196–206
29. Mudgala-Upākhyāna 3.246–47
30. Rāma-Upākhyāna 3.257–76
31. Sāvitrī-Upākhyāna 3.277–83
7, 13-15, 29, 31, 34-35, 40, 48, 50-51, 55, 57, 62, and 67), and Parida twenty-one of them (num-
bers 1, 2, 7, 10, 13-15, 17, 20, 29-33, 42, 47, 48, 50, 55, 64, and 67)—the latter, while still brack-
eting them among “interesting episodes” that were “inserted” as “this Epic grew to a great
extent.” Parida, Hospitality in Changing Indian Society, 76.
24 Hiltebeitel
42 There are three of these: numbers 27 (see Mahābhārata 3.195.37c), 33 (see 5.18.16a), and 39
(see 9.42.28a). The first two are also named upākhyānas in the colophons and headings;
the third only in passing.
43 The Parvasaṁgraha mentions two in Book I: numbers 9 (at 1.2.87) and 10 (at verse 90);
four in Book 3: numbers 20 (115ab), 21 (auddalakīya = Aṣṭāvakrīyam) and 23 as two cited
together (126ab), and 32 (127ed); and one in Book 5: number 35 (mentioned twice at
54a and 150f). Curiously, the PS’s description of the Rāma-Upākhyāna is “the very de-
tailed Rāmāyaṇa upākhyāna” (rāmāyaṇam upākhyānam … bahuvistaram; Mahābhārata
1.2.126ed).
44 While highlighting some of those discussed below, those with the best upākhyāna cre-
dentials for both N and S are numbers 7 (Vasiṣṭha), 8, 10 (Sunda-Upasunda), 11, 13 (Nala),
15 (Ṛśyaśrṅga), 27, 28, 30 (Rāma, usually as Rāmāyaṇa-Upākhyāna in S), 31 (Savitṛ), 34, 49
(Kṛtaghna), 50 (Japaka, on which see V.M. Bedekar, “The Place of Japa in the Mokṣadharma
Parvan (Mahābhārata 12.189–93) and the Yogasūtras,” Annals of the Bhandarkar Oriental
Research Institute 44 [1963]: 63–74), 51 (Cirakāri; on which see V.M. Bedekar, “The Legend
of Cirakārin in the Skanda Mahāpurāṇa and the Mahābhārata,” Purāṇa 11, no. 2 [1962]:
210–28; Gombach, “Ancillary Stories in the Sanskrit Mahābhārata,” part 1, 209–317; Fitzger-
ald, “The Rama Jāmadagnya ‘Thread’ of the Mahābhārata,” 112), 52, 54 (Uñchavṛtti), 60,
61, 66, and 67 (Nakula); those called upākhyānas only in N are numbers 3 (Mahābhiṣa),
9 (Pañcendra), 12, 16 (Kārtavirya), 18, 20, 36, 37, 43 (Rāma[-Jāmadagnya]), 45, 46, and 65;
those only in S, numbers 29 (Mudgala), 47 (Sumitra, called Ṛṣabha-Gītā in N), 48, 53
(Hayaśiras), and 59. Mudgala has an apparent descendent named Mavutkalliyar (Maudg-
alya) Muni who, in Tamil Draupadī cult stories, was married to Draupadī in her previous
life; see Alf Hiltebeitel, The Cult of Draupadī, vol. 2: On Hindu Ritual and the Goddess (Chi-
cago: University of Chicago Press, 1991), 484–85.
45 See Gombach, “Ancillary Stories in the Sanskrit Mahābhārata.”
26 Hiltebeitel
adhyāyas and larger narrative units. This is not the case in the Rāmāyaṇa, for
which I consider the number of upākhyānas to be zero. The Rāmāyaṇa men-
tions no upākhyānas in passing or in the upodghāta, which would be the
closest analogue to the PS; and the practice of giving names to sargas and larger
units differs from that for adhyāyas and larger units in the Mahābhārata in one
important respect. The Rāmāyaṇa’s Baroda Critical Edition shows that while
Northern Recension colophons do name a few units primarily as upākhyānas,46
the Southern Recension gives none as either primary or secondary names. In
fact, the Southern Recension gives hardly any sarga or larger unit names at all.47
The extensive absence of Southern Recension sarga names, especially the total
absence with regard to the few units called upākhyānas in the Northern
Recension, confirms that there would be no upākhyānas in the Rāmāyaṇa inso-
far as they could be counted as such in the Baroda Critical Edition, which
would require some parity across recensions. I take this as evidence that the
few instances of naming units upākhyānas in Northern Rāmāyaṇa manuscripts
is late and probably affected by the usage in the Mahābhārata. Indeed, it would
appear that whereas the Mahābhārata’s names for adhyāyas and larger units
tend to be genre-related, the Rāmāyaṇa’s names for sargas and larger units tend
to be mainly descriptive of events that transpire in the sarga. Most important,
the Rāmāyaṇa has neither a colophon discourse about upākhyānas nor a prac-
tice of using the term in passing that could have given rise to the few Northern
usages one finds.
Ākhyāna and upākhyāna are thus both among the multigenre terms by
which the Mahābhārata characterizes itself and its varied components. If
ākhyāna—even ahead of itihāsa—is the term used most frequently to describe
48 See, e.g., 1.2.124c-25d, where, after reference to “the series or upākhyānas” told in “the
encounter with Mārkaṇḍeya,” one finds one of them, Indradyumna Upākhyāna, referred
to as an ākhyāna.
49 By Gombach’s count (“Ancillary Stories in the Sanskrit Mahābhārata,” part 1, 10–22), it has
six “ancillary stories,” with the fifth having its own substory about the two Jaratkārus, male
and female, on which see Hiltebeitel, Rethinking the Mahābhārata, 174–76, with a sugges-
tion that this story offers a key as to how the Mahābhārata’s tales and subtales “fit.”
50 One could make an exception for the inclusion of the Aurva-Upākhyāna (1.169–72) within
the Vasiṣṭha-Upākhyāna (1.164–73), on which, see Sukthankar, “Epic Studies VI: The
Bhṛgus and the Bhārata,” 14: “a digression within a digression.” But I would stress as over-
riding factors the close thematic relation between the two stories and the uninterrupted
character of the upākhyāna material itself, which includes Vasiṣṭha not only in these two
upākhyānas but in the Tapati-Upākhyāna that immediately precedes them (1.160–63).
28 Hiltebeitel
narrates three (6–8) to Arjuna and the Pāṇḍavas; Śalya tells two: one (33) to
Yudhiṣṭhira, the other (38) to Karṇa and Duryodhana; Vyāsa tells one to
Draupadī’s father Drupada (9) and another to the Pāṇḍavas (29); and six are
told by single-time speakers: Kuntī to Pāṇḍu (5; the only upākhyāna spoken by
a woman), Nārada to the Pāṇḍavas (10); Bṛhadaśva to the Pāṇḍavas (13);
Akṛtavraṇa to the Pāṇḍavas (15; interrupting Lomaśa’s skein); Rāma Jāmadagnya
to the Kauravas (34); and Duryodhana to Karṇa and Śalya (47). As to auditors,
of the 56 that are addressed to main characters, 49 are told primarily to
Yudhiṣṭhira, 48 of these to him and his Pāṇḍava brothers, and 44 of these also
to their wife Draupadī (all of these told once the Pāṇḍavas and Draupadī are in
the forest). On the Kaurava side, three are addressed to Duryodhana and two to
Karṇa. Adding the 10 told to Janamejaya and one narrated to Pāṇḍu by Kuntī,
one finds that 65 of the 67 upākhyānas are addressed directly to members of
the larger Kaurava household to which all these listeners belong, and of which
Yudhiṣṭhira is clearly the chief listener. Not irrelevant to this pattern is the one
in which King Drupada hears upākhyāna 9 as an explanation of how his daugh-
ter can marry into that household. And likewise not irrelevant would be the
last upākhyāna in this tally, the anomalous number 53 known in S colophons
(see n. 31) as the Nārāyaṇīye Hayaśira-Upākhyāna. Here the primary narrator is
Ugraśravas, who answers a question by Śaunaka (speaking for the Naimiṣa
Forest Ṛṣis) about the Horse’s Head, a form of Viṣṇu, by quoting what Vyāsa
told Janamejaya about that subject.51 With Janamejaya as one of the two listen-
ers, one can now say that all the upākhyānas are addressed in one way or
another to those with ties to the Kaurava household. Moreover, with Vyāsa,
author of the outermost frame, addressing Janamejaya for once in the inner
frame, where he otherwise sits silently and leaves the recitation to Vaiśaṁ
pāyana, and to have all this further reported by Ugraśravas to the Naimiṣa
Forest Ṛṣis, means that this upākhyāna cuts across the Mahābhārata’s three
frames.52 Further, that the Naimiṣa Forest Ṛṣis are, this one and only time, an
interested party to a subtale suggests their proximity to this mysterious Veda-
reciting form of Viṣṇu that resides in the “great northern” or milky ocean,
51 The Critical Edition has suppressed this anomalous feature, overriding the preponderant
manuscript evidence at 12.335.1 and 9 to have Janamejaya cite Yudhiṣṭhira as the one
addressed by Vyāsa. See Hiltebeitel, “The Nārāyaṇīya and the Early Reading Communities
of the Mahābhārata.”
52 On the Narāyaṇīya context of this exchange, see Hiltebeitel, “The Nārāyaṇīya and the
Early Reading Communities of the Mahābhārata.”
Not without Subtales 29
53 See Hiltebeitel, Rethinking the Mahābhārata, 158 and passim: naimiṣa, “twinkling,” seems
to evoke the heavenly night sky, at least in the Mahābhārata. On the Horse’s Head and its
location, see Mahābhārata 12.330.36-39; 335.3, 27, and 34, and Hiltebeitel, “The Nārāyaṇīya
and the Early Reading Communities of the Mahābhārata.”
54 Barend A. Van Nooten, The Mahābhārata, Twayne’s World Authors Series (New York:
Twayne Publishers, 1971), 50: “about 73,900”; Brockington, The Sanskrit Epics, 4: “nearly
75,000.” A count has to be approximate because the Mahābhārata contains prose pas-
sages. One also has to count all couplets as “ślokas.”
55 These are the 151-verse Uttara-Yayāta (1.81–88), which continues upākhyāna number 2; the
368-verse continuation of no. 24 at 3.186–188; the 203-verse sequel to no. 62 in the
Cyavana-Nahuṣa-Saṃvāda (13.51–56); and the 58-verse Maitreya-Bhikṣā (13.121–23) which,
rather more loosely than the other cases, continues no. 65. I consider it more meaningful
to add upākhyāna-sequels than their prequels.
56 Gombach, “Ancillary Stories in the Sanskrit Mahābhārata,” part 1, 5, and 24.
57 Ibid., part 1, 194, 225.
30 Hiltebeitel
with two; Śalya (Book 9), with two; and Udyogaparvan (Book 5), with three are
comprised of 54%,58 28%, and 17% ancillary story material respectively.
Of the 67 upākhyānas, 57 thus occur in parvans 1, 3, 12, and 13 where “stories
cluster” most densely. There are, however, two major differences in the ways
upākhyānas are presented in the two early Books from the two later ones.
Whereas Books 1 and 3 provide multiple narrators for their 32 upākhyānas, all
but 3 of the 25 in Books 12 and 13 are spoken by one narrator, Bhīṣma (who has
told two upākhyānas earlier, one each in Books 5 and 6 [numbers 35 and 36], to
Duryodhana). And whereas Books 1 and especially 3 show a tendency to cluster
their upākhyānas (two in a row are told by Vaiśaṁpāyana and three in a row by
the Gandharva Citraratha in Book 1; nine, five, and two in a row by Ṛṣis whom
the Pāṇḍavas encounter while pilgrimaging in Book 3), in Bhīṣma’s run of four
hundred and fifty adhyāyas in Books 12 and 13, he tends to present his twenty-
one upākhyānas there only intermittently. Yet there is one run, from the end of
Book 12 through the first third of Book 13, where he concentrates nine of them.
These two books run together the totality of Bhīṣma’s postwar instructions to
Yudhiṣṭhira in four consecutive upaparvans, which James Fitzgerald calls “four
large anthologies.”59 Both Books abound in dialogues (saṃvādas), “ancient
accounts” (itihāsam purātanam),60 and other genres. Why then does Bhīṣma
intensify his upākhyānas at this juncture? This question will be taken up in the
section “Upākhyānas in the Mahābhārata.”
The upākhyānas’ content should also be important, and allow us to identify
certain themes that recur in them in meaningful patterns. But for now, the best
way to register their content would be by their primary personages or protago-
nists. This approach makes it possible to break the 67 down into no less than
ten groupings: 17 about leading lights of the great Brahman lineages,61 fifteen
58 This is by Gombach’s account which, I think, dubiously includes the Anugītā (14. 16.12-
19.60).
59 James L. Fitzgerald, “Introduction,” in The Mahābhārata, vol. 7. 11. The Book of Women. 12.
The Book of Peace, Part 1, trans. James L. Fitzgerald (Chicago: University of Chicago Press,
2004), 79–80.
60 Two pre-war upākhyānas, numbers 10 (1.201.1) and 33 (5.9.2), cite this phrase to describe
their content, as do fourteen post-war ones: nos. 41 (12.29.12), 44 (75.3), 46 (12.122.1), 49
(162.28), 50 (189.6, 192.2), 52 (263.2), 55 (13.2.4), 57 (12.2), 59 (28.6), 61 (40.2), 64 (70.2), and
67 (14.95.4).
61 Numbers 7, 8, 14, 15, 16, 17, 21, 22, 23, 43, 50, 51, 58, 61, 62, 64, 66. All eight of the major Brah-
man lineages are featured at least once: fifteen (2, 8, 16, 17, 18, 26, 34, 35, 43, 56, 60, 61, 62,
66, 67) feature Bhārgavas; nine (3, 6, 7, 8, 30, 44, 53, 56, 65) feature Vasiṣṭhas, including
Vyāsa); six (1, 22, 28, 50, 56, 62) feature Vaiśvāmitras; four (1, 15, 16, 43) feature Kāśyapas;
three (14, 33, 67) feature Agastya; three (23, 47, 49, 51) feature Gautamas; two (22, 60)
Not without Subtales 31
about heroic kings of varied dynasties,62 eleven about animals (some divine),63
seven about gods and demons,64 four (including the first two) about early kings
of the main dynasty,65 four about women,66 three about the inviolability of
worthy Brahmans and hurdles to attaining that status,67 three about revela-
tions concerning Kṛṣṇa,68 two about current background to the epic’s main
events,69 and one about the Pāṇḍavas as part of the main story.70 From this, the
only useful generalization would seem to be that such content is represented
as being of interest to the rapt audiences that listen to these tales. But here an
important point has been conceded by certain scholars who have been prone
to correlate such variety with lateness and textual strata. Regarding the most
famous of all the Mahābhārata’s upākhyānas, the Nala-Upākhyāna, Fitzgerald
regards “Nala” and some other non-upākhyāna stories as “good examples of
passages that do exhibit an inventive freedom suggestive of ‘fiction.’”71 More
feature Bharadvāja; and two (23, 54) feature Ātreyas (the latter, apparently, as suggested
by somānraye at 12.341.2). The number featuring Bhārgavas could be raised to 22 if we
note, as Sukthankar does (“Epic Studies VI: The Bhṛgus and the Bhārata,” 28–29), that
Mārkaṇḍeya is a Bhārgava. But these numbers would not suggest that the upākhyānas are
primarily Bhārgava material. See n. 11 above.
62 Numbers 13, 18, 19, 20, 26, 27, 29, 30, 41, 42, 44, 46, 47, 55, 57.
63 Numbers 11, 24, 25, 38, 45, 48, 49, 53, 54, 65, 67.
64 Numbers 4, 9, 10, 33, 37, 39, 52. Most frequently mentioned are: Indra in 26 upākhyānas (5,
6, 9, 10, 13, 14, 20, 21, 29, 33, 39, 41, 42, 43, 46, 48, 49, 50, 51, 54, 55, 57, 59, 61, 66, 67); Agni in
12 (11, 13, 27, 30, 33, 41, 43, 46, 48, 55, 57, 66); Brahmā in 12 (9, 10, 14, 24, 27, 29, 30, 45, 46, 49,
50, 53); Dharma in eight (4, 9, 32, 47, 50, 55, 66, 67); Viṣṇu in eight (9, 14, 27, 30, 33, 46, 50,
53); Śiva in six (9, 10, 14, 35, 46, 58); Yama in five (9, 13, 31, 46, 50); Varuṇa in five (13, 23, 30,
33, 46); Kubera in five (13, 33, 35, 44, 46); Bṛhaspati in five (2, 33, 41, 46, 67); Hayaśiras in
three (46, 47, 53); Soma in three (5, 33, 46); four gods in two each: Sūrya (6, 54), Gaṅgā (9,
41), Mṛtyu (46, 50), Krodha (50, 67); and numerous gods in one: Vāyu (30), Śrī (9), Śacī (33),
Viśvakarman (10), Aśvins (17), Tvaṣṭṛ (33), Upaśruti (33), Earth (43), Sarasvatī (46), Skanda
(46), Nirṛti (46), Kāla (50), Vedamāta Sāvitri (50), Svarga (50), Kāma (50), Kuṇḍadhāra
(52).
65 Numbers 1, 2, 3, 6.
66 Numbers 5, 28, 31, 40. But women figure centrally in at least 10 others, notably 1, 6, 17, and
35, which feature women in their titles and could have been counted in this category.
67 Numbers 56, 59, 60.
68 Numbers 34, 36, 63.
69 Numbers 12, 35. These are the only two upākhyānas where a character in the main story
tells about other characters in the main story. But cf. 11.27.6–11, Kuntī’s short account of
Karṇa’s mysterious birth, called an ākhyāna in the Parvasaṃgraha at 1.2.188a.
70 Number 32. On this anomaly, see below.
71 James L. Fitzgerald, “The Many Voices of the Mahābhārata,” review of Rethinking the
Mahābhārata: A Reader’s Guide to the Education of the Dharma King, by Alf Hiltebeitel,
32 Hiltebeitel
Journal of the American Oriental Society 123, no. 4 (2003): 803–818 discussing Hiltebeitel,
Rethinking the Mahābhārata, chapter 6 on “Nala,” and mentioning the Śuka story as well.
72 Madeleine Biardeau, “Nala et Damayantī. Héros épiques. Part 1,” Indo-Iranian Journal 27
(1984): 247–74; “Nala et Damayantī. Héros épiques. Part 2,” Indo-Iranian Journal 28 (1985):
1–34.
73 Gombach, “Ancillary Stories in the Sanskrit Mahābhārata,” 1 and 73.
74 Ibid., part 1, 164–65.
75 With his test of style (see n. 28), Sri Aurobindo sought to salvage the Nala- and Sāvitrī-
Upākhyānas for Vyāsa as works of “the very morning of Vyāsa’s genius, when he was young
and ardent.” Ghosh, “Notes on the Mahābhārata,” 44. On compilers and redactors, see my
discussion of this point with regard to Nārāyaṇiya scholarship in Hiltebeitel, “The
Nārāyaṇīya and the Early Reading Communities of the Mahābhārata” and Gombach’s for-
mulation that “the Mbh’s editors and redactors took pains to archaicize the epic” (“Ancil-
lary Stories in the Sanskrit Mahābhārata,” 1300) through the ancillary stories, as if
“interpolators” were editors and redactors and compilers who came along later than
authors, and as if there were not such archaizing as well in the presumably prior “epic.”
76 Biardeau, Le Mahābhārata, vol. 1, 412–13.
77 1, Appendix 1 line 4; 6.3709*. I thank Shubha Pathak, “The Things Kings Sing: The Religious
Ideals of Poetic Rulers in Greek and Sanskrit Epics” (PhD diss., University of Chicago,
2005), 50 for these references and for making available to me her further charting of the
Not without Subtales 33
kāvya in the Mahābhārata’s Critical Edition; as if the two texts were in early
agreement to yield one of these terms to the other. Neither does purāṇa
(ancient lore) describe the Rāmāyaṇa,78 which evidently places itself outside
the itihāsa-purāṇa tradition that Chāndogya Upaniṣad 7.1.2 links with Nārada
as a fifth Veda.79 Similarly, upākhyāna, “subtale,” is used only in the Mahābhā
rata, although there is an interpolated verse in the Rāmāyaṇa’s aśvamedha
recital scene where the twins begin singing the poem and tell Rāma that the
Rāmāyaṇa has twenty-four thousand verses and a hundred upākhyānas
(Rāmāyaṇa 7.1328*, following 7.85.20)—suggesting Mahābhārata influence.
Other words the Mahābhārata uses to define itself such as śāstra (treatise) do
not define the Rāmāyaṇa at all.
A distinctive point about the usage of kāvya is that it is used only at the
Rāmāyaṇa’s two framing points:80 nine times in the upodghāta, four in the two
chapters of the aśvamedha recital scene (Rāmāyaṇa 7.84–85) where the hints
left at the end of the upodghāta are picked up as the frame finally enters the
story (or where the story finally returns to the frame). It thus has a kind of
bookend function of describing the work as poetry, most notably that “it is
replete with” all the “poetic sentiments” or rasas (1.4.8).81 In contrast to kāvya,
carita implies the “movement” (√car) of the main narrative. Of its four usages
in the upodghāta to characterize the Rāmāyaṇa, two present a juxtaposition.
The first has Brahmā enjoin Vālmīki to “compose the whole adventure of
two epics’ terms for themselves in the star passages and appendices of their critical edi-
tions.
78 Though, curiously, the only instance where purāṇa describes a story of any kind occurs
when Rāma hears from the fallen vulture Sampāti how the Ṛsi Niśakara once told him
that, “in an ancient legend” (purāṇe) he once heard (Rāmāyaṇa 4.61.3), Rāma’s life was
foretold with some strange and exceptional twists (4–13).
79 Clearly the Mahābhārata does not do this. I would even suggest that Vālmīki might be
registering a distrust of this tradition, which the Rāmāyaṇa puts under question right at
the beginning when Nārada describes Rāma as the ideal man. Note that in the Mahābhā
rata, Nārada includes Rāma among the sixteen great kings of old whose past glories he
recounts (twice) in the soḍaśarājakīya, the second version being upākhyāna number 41,
as quoted by Kṛṣṇa.
80 Aside from kāvya, the only other words to describe the Rāmāyaṇa at the aśvamedha
recital are carita and, as the twins now sing it, gītā: “Having heard the sweetness of the
song [gītāmādhuryam], he [Rāma] returned to the sacrificial pavilion” (Rāmāyaṇa
7.85.23).
81 On the rasas in the upodghāta, see the rich discussion in Pathak, “The Things Kings Sing”;
similarly, in the Sundarakāṇḍa, see Robert P. Goldman, trans. Sundarakāṇḍa, vol. 5 of The
Rāmāyaṇa of Vālmīki: An Epic of Ancient India, ed. Robert P. Goldman and Sally Suther-
land Goldman (Princeton, NJ: Princeton University Press, 1996), 35–37.
34 Hiltebeitel
Rāma” (rāmasya caritaṃ kṛtsnaṃ kuru; 1.2.30cd). The second, once it is implied
that Vālmīki has composed it, calls “the whole Rāmāyaṇa poem [kāvya] the
great adventure of Sītā [sītāyāś carītam mahat]” (1.4.6). This suggests that
although Rāma’s adventure is Vālmīki’s starting point the complete poem is
also about Sītā’s adventure. The “profound adventure” that Rāma prepares
himself to hear at the end of the upodghāta would thus include the two adven-
tures intertwined (4.26). This sense of carita as ongoing double adventure
carries through the narrative. For instance, when Hanumān begins to tell
Bharata about “Rāma’s whole adventure in the forest” (sarvaṃ rāmasya caritaṃ
vane; 6.114.4cd) near the end of Book 6, he tells mainly of the separate adven-
tures of Rāma and Sītā once the latter is abducted (cf. 2.54.18). Carita is also the
main word to describe the Rāmāyaṇa’s adventures in course (2.54.18; 6.114.4)—
and even in the course of hearing it. When the twins begin reciting the poem
and Rāma asks who composed it, they reply, “The blessed Vālmīki, who has
reached the presence of the sacrifice, is the author [kartā] by whom this adven-
ture is disclosed to you without remainder [yena idaṃ caritaṃ tubhyam aśeṣaṃ
sampradarśitam]” (7.85.19).
Meanwhile, ākhyāna is used four times in the upodghāta. It describes the
benefits of hearing the tale’s recital (Rāmāyaṇa 1.1.78), that it is “unsurpassed”
as a “tale exemplary of righteousness [dharmyam ākhyānam uttamam]” (1.4.11),
that it is a “wondrous tale told by the sage” that he “completed in perfect
sequence” as “the great source of inspiration for poets [kavīnām]” (1.4.20), and
that Rāma urged his brothers to “listen to this tale whose words and meanings
alike are wonderful as it is sweetly sung by these two godlike men” (1.4.25). It is
also the first term to describe the Rāmāyaṇa as the recital of its main story
begins: “Of these kings of illustrious lineage, the Ikṣvākus, this great tale is
known as the Rāmāyaṇa. I will recite it from the beginning in its entirety, omit-
ting nothing. It is in keeping with the goals of righteousness, profit, and
pleasure and should be listened to with faith” (1.5.3–4). Ākhyāna can also be
used for tales told in course, most notably for the “glad tidings” that Hanumān
brings at various points to others (5.57.1, 59.6; 6.101.17, 113.40).82 It is thus com-
plementary to both kāvya and carita. It links the narrative to poetry and the
inspiration of poets while also bringing listeners into the unfolding of the hero
and heroine’s double adventure, the reiterations of that adventure by
Hanumān, and the blessings that it brings when heard with faith.
82 In contrast to these four consistent usages and the four in the upodghāta, it is certainly in
a minor key that ākhyāna is used just once—for the “tale of the descent of the Gaṅgā”
(Rāmāyaṇa 1.43.30)—for a unit within the whole.
Not without Subtales 35
The Rāmāyaṇa thus makes very selective use of limited terms. In contrast to
the Mahābhārata, they are used strategically rather than definitionally, and are
not used to emphasize the interplay between the Rāmāyaṇa’s parts and its
whole. Emerging from and flowing back into the passages that frame the
Rāmāyaṇa (the upodghāta and the aśvamedha recital scene), side-stories fall
within a single poetic narrative that is portrayed as being addressed uninter-
ruptedly (the one exception noted) to Rāma. The Rāmāyaṇa does not have
multiple audiences in a thrice-told stacking of dialogical frames.
We must now see how these findings relate to the different manners in
which subtales figure in the two epics’ main stories.
83 See now Alf Hiltebeitel, “The Southern Recension’s Śakuntalā as a First Reading: A Win-
dow on the Original and the Second Reading by Kālidāsa,” in Revisiting Kālidāsa’s
Abhijñāna Śākuntalam: Land, Love, Languages: Forms of Exchange in Ancient India, ed.
Deepika Tandon and Saswati Sengupta (Delhi: Orient BlackSwan Edition, 2011), 17–37 on
the importance of the Śakuntalā story’s primacy among upākhyānas, though only in the
Northern Recension, and its considerable (and probably early) rehandling in this and
other regards in the Southern Recension. Although I neglected to reread Biardeau’s 1979
article on the Śakuntalā-Upākhyāna in researching this chapter, I find now that this study
36 Hiltebeitel
youths of the main heroes, with heightened attention to the three generations
before them. This narrative widening begins with the Mahābhiṣa-Upākhyāna
(number 3),84 about how Mahābhiṣa, a royal sage residing in heaven, boldly
gazes up the windblown skirt of the heavenly river Gaṅgā and is cursed to
earthly birth, whereupon, as King Śantanu, he marries Gaṅgā85 their union
resulting in the birth of Bhīṣma as their eighth86 and sole surviving son and
Gaṅgā’s departure once Śantanu asks why she drowned the first eight lead-
ing to Śantanu’s second marriage to Satyavatī, now a fisher-princess, upon her
father’s obtaining Bhīṣma’s double vow to renounce kingship and women,
for which Śantanu gives Bhīṣma the boon to be able to choose his moment
of death; Bhīṣma’s abduction of three sisters, two as brides for Śantanu and
Satyavatī’s second son, who dies soon after becoming king, leaving the two as
widows, and the third, the unwedded Ambā, with thoughts of revenge against
Bhīṣma; Satyavatī’s determination to save the line by getting the two widowed
queens pregnant, first by asking Bhīṣma, who refuses to break his vow of celi-
bacy, and then, admitting her premarital affair, recalling her first son Vyāsa;
Vyāsa’s unions with the two widowed sisters, cursing the first to bear a blind son
because she had closed her eyes at his hideous ascetic ugliness and the second
to bear a pale son because she had blanched; the births of the blind Dhṛtarāstra,
the pale Pāṇḍu, plus a third son, Vidura, sired with the first widow’s low caste
maidservant—and behind Vidura’s birth, the Aṇimāṇḍavya-Upākhyāna (the
fourth), named after a sage who learns that he was impaled as the result of a
childhood sin in his previous life and curses the god Dharma—lord of post-
mortem punishments and thus “functionally”87 tantamount in this, his virtual
is returning to an argument she makes there about upākhyānas in the Mahābhārata: “the
apparently secondary accounts with which the epic is stuffed, far from being what one
lately calls interpolations, are the reprise under a symbolic form of the dominant message
of the principal account, which they thus aide to decipher, all while contributing to the
progression of the intrigue.” Madeleine Biardeau, “Śakuntalā dans l’épopée,” Indologica
Taurinensia 7, Dr. Ludwig Sternbach Felicitation Volume, Part I (1979): 120 (my translation).
84 One of the epic’s “three beginnings” (see Mahābhārata 1.1.50); these are probably recom-
mended for performance purposes like the 24,000-verse Bhārata.
85 See Andrea Custodi, “Dharma and Desire: Lacan and the Left Half of the Mahābhārata”
(PhD diss., George Washington University, 2004), 155–203 on the theme of the gaze in this
upākhyāna.
86 This corrects earlier versions of the article, where I erroneously had Bhīṣma as the “ninth”
son.
87 On this episode, see Hiltebeitel, Rethinking the Mahābhārata, 192–95. As Kantawala says,
Dharma, seated on his throne and meting out “justice,” is here “a functional name given to
Yama.” S.G. Kantawala, “The Legend of Aṇī Māṇḍavya,” in Modern Evaluation of the
Not without Subtales 37
epic debut, to Yama, god of the dead—to suffer Vidura’s low-caste human birth.
From here, one enters upon the generation of the main heroes. Dhṛtarāṣṭra’s
marriage to Gāndhārī yields the hundred Kauravas, incarnate demons headed
by Duryodhana. And once Pāṇḍu becomes impotent after his marriages to
Kuntī and Madri, Kuntī tells him the Vyuṣitāṣva-Upākhyāna (number 5) about
a queen made pregnant by her husband even after he was dead as part of the
build-up to her disclosure that she has the means to induce pregnancy by gods,
which results in Pāṇḍu choosing Dharma to sire his first son, Yudhiṣṭhira, and
so on. Already we see how impoverished the Mahābhārata would be “without
upākhyānas,” the first five of which tie in with the main story through a train
of curses and boons having to do with sex and with death, identify dharma/
Dharma as death/Yama, and, while stirring up such undercurrents below
the surface, lay the groundwork for the birth of Dharma’s son, Dharmarāja
Yudhiṣṭhira.
After some youthful trials, the Paṇḍavas must conceal their survival from
the Kauravas, which they do disguised as Brahmans, and Vyāsa appears to
direct them to Pañcāla where they will meet their destined bride. On the way
Arjuna defeats the Gandharva Citraratha who had challenged him. Citraratha
tells the Pāṇḍavas they are vulnerable without keeping a priest and holy fires,
and then relates three upākhyānas in succession: the Tapati-Upākhyāna (about
another of their ancestresses Tapatī, daughter of the Sun and mother of the
eponymous Kuru), and the Vasiṣṭha- and Aurva-Upākhyānas (about Brahmans),
all three of which prepare them for forthcoming adventures while imparting
some positive and negative information on marriage and sexuality.88 Then,
when the five Pāṇḍavas—still disguised as Brahmans—marry Draupadī,
Vyāsa, who “by chance arrived” (Mahābhārata 1.187.32d), sanctions the mar-
riage by telling Draupadī’s father Drupada the Pañcendra-Upākhyāna. At a
sacrifice performed by the gods at Naimiṣa Forest, Yama was consecrated as
the śamitṛ priest assigned to putting victims to death, which detained him
from killing humans for the rite’s duration, making the gods edgy until they
learned from Brahmā that the rite would strengthen Yama for this job once it
was done. As one of the attendees, Indra then saw golden lotuses floating down
the Gaṅgā and traced them upriver to the tears of the goddess Śrī, who was
Mahābhārata. Prof. P.K. Sharma Felicitation Volume, ed. S.P. Narang (Delhi: Nag Publish-
ers, 1995), 104–5.
88 Said in correspondence to the point made by Sutherland Goldman (2004, 72) cited below
in connection with Rāma and Lakṣmaṇa’s approach to Mithilā. Tapatī is a positive ances-
tress of the Pāṇḍavas, mother of Kuru: Vasiṣṭha’s encounter with the cannibal Kalmāṣapāda
builds up to the latter’s near attack of a pregnant woman.
38 Hiltebeitel
weeping at the river’s source over the fall of four former Indras, her former
husbands, into a cave as the result of their arrogance toward Śiva. Once the
current Indra has suffered the same fate, Vyāsa reveals that the Pāṇḍavas are
the five Indras, cursed by Śiva to become mortals and marry Draupadī, who is
Śrī incarnate, which gives the marriage a resemblance of monogamy; they will
regain Indroloka only after performing “unbearable” (aviṣahya) and lethal
karma. Further, Kṛṣṇa and Balarāma have joined them incarnated from two
hairs of Viṣṇu. The Pañcendra-Upākhyāna thus shifts the emphasis from the
five gods who sire the Pāṇḍavas, beginning with Dharma, to the Pāṇḍavas
being additionally five Indras. Vyāsa says nothing explicit to indicate that
Yama’s death-dealing mission is to be carried out in the person of Yudhiṣṭhira
at the battle of Kurukṣetra. But if Dharma and Yama are “functionally” the
same, this conclusion would be inevitable.89 The ninth upākhyāna, spoken by
the author himself, thus deftly knots together threads we detected in the first
five.90
After some amends are made between the two camps, the seer Nārada
arrives at the Pāṇḍavas’ new capital, Indraprastha, to tell the Sunda-Upasunda-
Upākhyāna about two demonic brothers who kill each other over a woman,
thereby warning the Pāṇḍavas to regulate their time with Draupadī and pro-
viding them with an inverse mirror story to their own situation—and the very
rule that will send Arjuna into a period of exile in which he will marry three
other women. Two upākhyānas in a row are thus concerned with the marriage
of Draupadī: through the first, Vyāsa tells Draupadī’s father Drupada the divine
secret that makes it legal, and through the second Nārada tells the Pāṇḍavas
how to regulate it. Finally, when Arjuna and Kṛṣṇa burn the Khāṇḍava Forest to
satisfy Agni and clear the ground for the construction of Indraprastha, Book 1
closes with the Śārṅgaka-Upākhyāna about some precocious birds reminiscent
of the four Vedas91 who escape the blaze.
Book 3, the Āraṇyakaparvan, relates 21 upākhyānas during the Pāṇḍa-
vas’ residence in the forest. Early on, Kṛṣṇa tells its first upākhyāna—the
Saubhavadha-Upākhyāna—to explain his absence from the dice match.
Thereafter, most of the upākhyānas relate to events in the Book’s second
through fifth upaparvans and its final three. The opening sequence tells
92 Durvāsas also appears three other times in this role. He is a cranky and demanding guest
in the household of Kuntibhoja, who leaves Durvāsas’s high maintenance to his adopted
daughter Kuntī (1.104; 1.113–14; 5.142.19). In the Durvāsa-Māhātmya (13.144), Yudhiṣṭhira
directs a question about honoring Brahmans to Kṛṣṇa rather than Bhīṣma and learns how,
when no one else would invite Durvāsas as he went about uttering a verse proclaiming his
troublesome reputation, he came to stay with Kṛṣṇa and Rukmiṇī and ate voraciously and
harassed them in other ways until he finally granted them boons for keeping their temper.
And in a Northern passage justly rejected by the Critical Edition, he grants Duryodhana,
after being well fed, the boon of appearing with his horde of ten thousand disciples before
Draupadī to demand food just after she has fed the Pāṇḍavas and gone to take rest, where-
upon Kṛṣṇa comes to her rescue, filling the horde’s bellies from one gram of leftover rice,
whence Durvāsas withdraws lest he provoke the Pāṇḍavas by not finishing the meal pro-
vided (3, App, I, No. 25).
40 Hiltebeitel
As noted, three upākhyānas in Book 3 stand out as what Biardeau calls “mir-
ror stories”: the Nala-Upākhyāna—the love story about Nala and Damayantī
told by the seer Bṛhadaśva while Arjuna is visiting Śiva and Indra and Draupadī
misses this favorite of her husbands; the Rāma-Upākhyāna—a “Mahābhāra
ta-sensitive” version of the Rāma story93 focused on Sītā’s abduction and
told to all five Pāṇḍavas and Draupadī by Mārkaṇḍeya just after Draupadī’s
abduction; and the Sāvitrī-Upākhyāna—the story of a heroine who saved her
husband from Yama, told by Mārkaṇḍeya just after the Rāma-Upākhyāna when
Yudhiṣṭhira asks, having already heard about Sītā, if there ever was a woman as
devoted to her husband(s) as Draupadī (this implicit slighting of Sītā is rather
curious). Finally, the “Firesticks Subtale” then closes Book 3 as it began with
the encounter of a monster who appears first as a speaking crane and for the
moment “kills” the four youngest Pāṇḍavas at a lake where they have gone to
slake their thirst. But whereas the first monster, Kirmīra, was a rākṣasa, this
crane turns into a one-eyed Yakṣa before he reveals himself, after question-
ing Yudhiṣṭhira, to be Yudhiṣṭhira’s own father Dharma in disguise. Gratified at
his son’s subtle answers to his puzzling questions, Dharma revives Dharmarāja
Yudhiṣṭhira’s brothers and promises him success in disguising himself in Book
4 during the thirteenth year in exile, which soon inspires Yudhiṣṭhira to take on
the name “Heron” (an “eater of fish,” like the first “crane” disguise of his father)
and to introduce himself to King Virāta of Matsya (the kingdom of “Fish”) as a
dicing master thanks to his having received the boon of “the heart of the dice”
after hearing how this skill saved Nala in the Nala-Upākhyāna.94 One may also
suspect that a subcurrent runs between the Sāvitrī-Upākhyāna in which Yama
restores life to Sāvitrī’s husband Satyavan and the “Firesticks Subtale” in which
Yudhiṣṭhira restores life to his brothers, for such a parallel between Yama and
Yudhiṣṭhira would not only hark back to their already established connections
through Dharma, but anticipate Book 4, which will speak of the Pāṇḍavas’
year in concealment in Matsya as a rebirth from the womb (Mahābhārata
93 See Alf Hiltebeitel, “Authorial Paths through the Two Sanskrit Epics, Via the Rāmo
pākhyāna,” in Epic Undertakings: Papers of the 12th World Sanskrit Conference, vol. 2, ed.
Robert P. Goldman and Muneo Tokunaga, (Delhi: Motilal Banarsidass, 2009), 169–214.
94 Moreover, all five Pāṇḍavas and Draupadī adopt disguises that could be cued from the
Nala-Upākhyāna; see Hiltebeitel, Rethinking the Mahābhārata, 228–29). On Yudhiṣṭhira-
Kaṅka and the Matsyas, see ibid., 197 and n. 562; Madeleine Biardeau, “Études de mythol-
ogie hindoue V: II. Bhakti et avatāra (suite),” Bulletin de l’École Française d’Extrême-Orient
65 (1978): 99–101, 107; Madeleine Biardeau, “Un certain Kīcaka,” in Lex et litterae: Studies in
Honour of Professor Oscar Botto, ed. Siegfried Lienhard and Irma Piovano (Alessandria:
Edizioni dell’Orso, 1997), 44–47. The foregoing sentence corrects an error in Hiltebeitel,
“Not Without Subtales,” 484, before n. 67.
Not without Subtales 41
4.66.10cd). We shall find later support for reasoning that the Parvasaṁgraha
calls the “Firesticks Subtale” an upākhyāna, even though it is the only one that
is part of the main story, just because so many upākhyāna crosscurrents run
through it.
Looked at from the standpoint of its three upākhyānas, Book 5, the Udyoga
parvan, presents surprising symmetries and asymmetries. The initial upaparvan
traces how both sides try to secure alliances. Arjuna and Duryodhana come to
Dvārakā to seek aide from Kṛṣṇa, who says bafflingly that his relation to each is
equal. But since he saw Arjuna first he gives him the first choice of two options:
Kṛṣṇa as a noncombatant charioteer, or a whole army division. Arjuna chooses
Kṛṣṇa and Duryodhana departs content. Then the Madra king Śalya, brother of
the twins’ mother Mādrī, sets out to join the Pāṇḍavas but has his mind turned
after he finds elegant way-stations en route prepared for him by Duryodhana.
Travelling on, he tells Yudhiṣṭhira that he has sided with Duryodhana, and
Yudhiṣṭhira, foreseeing that Śalya will be Karṇa’s charioteer, asks him to destroy
Karṇa’s confidence (tejas,) in combat. Telling Yudhiṣṭhira that even Indra had
ups and downs, Śalya consoles him with Book 5’s first upākhyāna, a cycle of
three ultimately triumphant Indra stories called the Indravijaya-Upākhyāna
(Mahābhārata 5.9-18): both a rear-view mirror story comparing Draupadī’s suf-
ferings with Damayantī’s (58.34cd) and Yudhiṣṭhira and Draupadī’s tribulations
in Book 4 with those of Indra and Sāci, and a prophetic lens through which to
see aspects of Yudhiṣṭhira’s war conduct anticipated by Indra’s.95
As negotiations proceed, events come to center on the lengthy middle upa-
parvan 54, titled “The Coming of the Lord,” in which Kṛṣṇa as divine messenger
comes as the Pāṇḍavas’ last negotiator with the Kauravas while a host of celes-
tial seers descend to watch the proceedings and tell stories: one of them an
upākhyāna about the arrogant king Dambhodbhava that Rāma Jāmadagnya
tells to warn Duryodhana that Arjuna and Kṛṣṇa are the ancient indomitable
seers Nara and Nārāyāṇa.96 Arbitrations break down when Duryodhana tries,
futilely to capture Kṛṣṇa, and end when the Kauravas send Śakuni’s son Ulūka
(Duryodhana’s mother’s brother’s son, who thus has the same relation to
Duryodhana that Kṛṣṇa has to the first three
Pāṇḍavas) with a last abusive message to the Pāṇḍavas. Book 5 then closes with
95 Parallel themes in the Indravijaya-Upākhyāna and the Mahābhārata war include destruc-
tion of opponent’s tejas, breach of friendship (sakhya), reliance on stratagems supplied by
Viṣṇu/Kṛṣṇa, and concluding sin-cleansing aśvamedhas.
96 Other important stories related here are Kaṇva’s story of Mātali (95–103) and Nārada’s
about Gālava (104–21), the latter called the Gālava-Carita and “this great incomparable
ākhyāna” (idam rnahākhyanam anuttamam; Mahābhārata 5.121.22a) in closing.
42 Hiltebeitel
97 See Custodi, “Dharma and Desire: Lacan and the Left Half of the Mahābhārata,” 204–63
on the constraints on gender transformation in this upākhyāna.
98 Albeit charmingly, Van Buitenen, typically (see nn. 28 and 67 above) finds this upākhyāna
“epigonic” and “absurd.” J.A.B. van Buitenen, “Introduction [to Book 5. The Book of the
Effort],” in The Mahābhārata: 4. The Book of Virāṭa; 5. The Book of the Effort (Chicago: Chi-
cago University Press, 1978), 175, 178. While offering the convincing formulation “I assume
that it developed within the Mahābhārata” (ibid., 176), he takes it to have been added as
“instant tradition” toward the end of the epic’s “half millennium of … composition.” Ibid.,
178. Along with his own “monologue intérieur” to account for such a belated creation
(ibid., 177), he offers such erroneous or misleading statements and details as: Rāma
Jāmadagnya’s appearance in the story is “posthumous” (ibid., 175); King Drupada is “once
more … sonless” (far more likely, Śikhaṇḍin is old enough, and certainly older than
Dhṛṣṭadyumna and Draupadī, to be among the “accursed brood” [dhig bandhūn; 1.155.3b;
Van Buitenen trans.] of sons Drupada speaks of when he desires one who will be up to
killing Droṇa); and with five references to it elsewhere in the epic, the story of Śikhaṇḍin’s
sex change is “astonishingly underplayed.” Ibid., 176.
99 See Alf Hiltebeitel, “The Two Kṛṣṇas on One Chariot: Upaniṣadic Imagery and Epic
Mythology,” History of Religions 24 (1984): 15 and n. 42, where I tried for the last time to
develop the notion of “background story” as a way to handle some of the issues raised by
this chapter: in particular, the relation of certain myths (mostly, however, not upākhyā
nas) to certain parvans.
Not without Subtales 43
100 The way Vaiśaṁpāyana sets the scene is remarkable, as are Kṛṣṇa’s use of humor (hāsya
rasa) and Yudhiṣṭhira’s momentary openness to it; see Alf Hiltebeitel, “On Reading
Fitzgerald’s Vyāsa,” Journal of the American Oriental Society 125, no. 2 (2005): 241–61.
44 Hiltebeitel
able to recall that Vyāsa himself had exalted the same practice, along with the
merits of giving that meager fare to guests, toward the end of Book 3 in the
Mudgala-Upākhyāna. Moreover, since Book 3 ends with the “Firesticks Subtale”
in which Dharma appears disguised as a crane and a puzzle-posing Yakṣa, it
would appear that one strain of the epic’s upākhyānas carries a major subcur-
rent through such puzzle pieces, especially in that they frequently punctuate
the ends of major units. Moreover, with one such story ending the Śāntiparvan,
we have reached the juncture mentioned earlier where Bhīṣma is launching
his only concentrated stretch of upākhyānas.
Book 13, the Anuśāsanaparvan, begins with Bhīṣma’s fourth anthology, on
Dānadharma, comprising his closing “further instructions”101 to Yudhiṣṭhira
on “the law of the gift” (upaparvan 87). Here we must consider Fitzgerald’s
hypothesis that the four anthologies demonstrate decreasing “tautness” and
increasing relaxation as the result of “a progressive loosening of editorial
integration”102 over centuries, from the second century bce down to the fourth-
to-fifth century CE.103 Fitzgerald’s point is buttressed by the general impression
scholars have had that the Anuśāsanaparvan is loose and late. R.N. Dandekar,
the Critical Edition editor of this edition’s last-to-be-completed parvan, per-
haps puts it best:
The scope and nature of the contents of this parvan were such that liter-
ally any topic under the sun could be broached and discussed in it …. This
has resulted in poor Yudhiṣṭhira being represented as putting to his
grandsire some of the most elementary questions—often without rhyme
or reason. Not infrequently, these questions serve as mere excuses for
introducing a legend or a doctrine fancied by the redactor, no matter if it
has already occurred in an earlier part of the Epic, not once but several
times.104
101 The Parvasaṁgraha makes this connection by calling the Dānadharmaparvan the
Ānuśāsanika, “Further Instruction,” at Mahābhārata 1.2.65b.
102 Fitzgerald, “Introduction,” 147–48.
103 Ibid., 114.
104 R.N. Dandekar, “Introduction,” in The Anuśāsanaparvan for the First Time Critically Edited,
part 1 (Poona: Bhandarkar Oriental Research Institute, 1961), xlvii.
105 Dandekar is quite insistent, going on: “the redactors must have seen in this parvan per-
haps their last opportunity to introduce into the epic various miscellaneous topics which
Not without Subtales 45
had occurred to them at the last moment” (ibid., lxxiv–lxxv); and rephrases this point in
his Introduction to the Critical Notes (R.N. Dandekar, ed. The Anuśāsanaparvan for the
First Time Critically Edited, part 2 [Poona: Bhandarkar Oriental Research Institute, 1961],
1051): “Some of the questions put into Yudhiṣṭhira’s mouth are so elementary that they
show the venerable king to be a naive person. They make one wonder if Yudhiṣṭhira had
not learnt even commonplace things in the course of his long and eventful life.”
106 Dandekar, “Introduction,” lix.
107 See Hiltebeitel, “On Reading Fitzgerald’s Vyāsa.”
108 Dandekar, “Introduction,” lvii–lviii.
109 Dandekar, “Introduction,” lviii supplies this compound. The text speaks of atithipūja,
“honoring or worshiping guests” (Mahābhārata 13.2.68–69, 91), as does the Uñchavṛtti-
Upākhyāna just before it (12.347.3) ending the Mokṣadharma, and the Kapota-Upākhyāna
(12.142.39–40, 143.8) in the Āpaddharma. Cf. atithi-vratin, one who is “devoted to guests,”
in the Mudgala-Upākhyāna (3.246.4 and 15).
46 Hiltebeitel
110 The atithi is not just an ordinary guest (abhyāgata) but, etymologically, “one who has no
fixed day in coming” (Monier-Williams, A Sanskrit-English Dictionary, 14), that is, an unin-
vited and “date-less” (a-tithi) guest. Thus Ganguli’s delightful note on the Uñchavṛtty-
Upākhyāna: “the word ‘atithi’ which is rendered ‘guest’ here and elsewhere, means a
person who enters without invitation the abode of a householder. Such an individual is
adorable. All the deities reside in his person. He is supposed to favor the householder by
giving him an opportunity of performing rites of hospitality …. [But] he cannot expect to
be served with food till the householder has done his best for serving him as sumptuously
as possible …. Hence, by the time the food is placed before him, the guest becomes very
hungry.” Ganguli, trans. The Mahabharata, vol. 10, 630 n. 1 (second ellipsis Ganguli’s). My
thanks to Balaji Hebbar for pointing out the abhyāgata/.atithi distinction.
111 At the very moment Sudarśana arrives home to hear Dharma’s report from the bedroom
that he is there with his wife, Death (Mṛtyu) is standing over Sudarśana with a concealed
iron mallet (kūtamudgara; Mahābhārata 13.2.66) to test his reaction. A study of the inter-
activities between Yama, Mṛtyu, Kāla, and Dharma in the epics would be revealing.
112 Indeed, both discussants might have things to ponder here. Yudhiṣṭhira would know that
he owes his very birth and rule to the fact that his father Pāṇḍu gave his mother Kuntī to
Dharma first among the gods he selected to sire his sons, and might also ponder whether
he gave his own wife Draupadī to dharma at the dice match; and Bhīṣma now prolongs his
life on a bed of arrows, overcoming death to this extent thanks to a boon given by his
father for renouncing marriage—a sort of gift of his wife to the dharma of his father.
113 Jamison, Sacrificed Wife/Sacrificer’s Wife, 96, 283, n. 221.
114 Alf Hiltebeitel, “Weighting Orality and Writing in the Sanskrit Epics,” in Epics, Khilas, and
Purāṇas: Continuities and Ruptures, Proceedings of the Third Dubrovnik International
Conference on the Sanskrit Epics and Puranas, September 2002, ed. Petteri Koskikallio
(Zagreb: Croatian Academy of Sciences and the Arts, 2005), 81–111.
Not without Subtales 47
115 For the stretch from 12.340 to 13.51, Bhīṣma recounts upākhyānas at a clip of 35 out of 64
adhyāyas or 54.7%. By the same rough measure, that contrasts with 64 out of 450 adhyāyas
or 14.22% over the four anthologies as a whole: about the same clip as that for upākhyānas
in the epic at large (see at nn. 40 and 41).
116 On which see the discussion below at nn. 97–98 in connection with the Rāmāyaṇa.
117 See Sukthankar, “Epic Studies VI: The Bhṛgus and the Bhārata, 45: “the third repetition of
the birth of Jamadagni,” etc.
48 Hiltebeitel
ogy approach allows. As the return to upākhyānas indicates, Book 13 goes well
beyond the rules of transformation that Fitzgerald offers as explanation for
such a progression.
With Book 14, the Āśvamedhikaparvan, Yudhiṣṭhira, now adding Bhīṣma’s
demise to his guilt over the war, agrees to perform a sin-cleansing Horse
Sacrifice at Vyāsa and Kṛṣṇa’s bidding. While the Pāṇḍavas prepare for it, Kṛṣṇa
wants to see his people at Dvārakā, and on the way meets the sage Uttaṅka for
the multistoried Uttaṅka-Upākhyāna. Arjuna then has many adventures guard-
ing the horse. But immediately upon the rite’s completion an angry half-golden
blue-eyed mongoose appears from his hole to disparage the grand ceremony as
inferior to a gleaner’s hospitality to a ravenous guest. With this incident comes
the Mahābhārata’s final upākhyāna: this time a double puzzle piece that reveals
the mongoose to have been Dharma in disguise when testing the “pure gift”
(śuddha dāna; Mahābhārata 14.93.57) of the gleaner, an uñchavṛtti Brahman;
but before that, Dharma had been Anger (Krodha) as a mysterious guest who
tested the absence of anger in the Ṛṣi Jamadagni and been cursed by the lat-
ter’s ancestors to become the mongoose. It addresses the question of whether
a king’s giving to Brahmans and others in sacrifice is comparable to the glean-
er’s “pure gift,” done with devotion and faith and without anger, to Dharma,
that ever-demanding guest who would harbor not only this trace of anger but,
from his Mahābhārata debut, the “functional” identity of Yama. Again, a major
unit ends with an upākhyāna puzzle piece on this theme of dharma’s disguises.
Moreover, it brings to culmination the cycle of substories about gleaners with
the hungry guest finally being not just Durvāsas but Dharma—who will have
one remaining disguise by which to test Yudhiṣṭhira in the last adhyāya of the
epic’s penultimate Book 17: that of the dog whom Yudhiṣṭhira, because of his
“non-cruelty” (ānṛśaṃsyam; 17.3.7d), will refuse to abandon even at the cost of
heaven. Yet this last lesson is not really finished until Yudhiṣṭhira curses
dharma/Dharma out of anger (18.2.42-45) at seeing Duryodhana in heaven,
which brings home the mongoose’s last lesson about how even Dharma leaves
his anger behind him. Just so, as Yudhiṣṭhira puts his human feelings behind
him, he bathes in the heavenly Gaṅgā and becomes “freed of enmity” (18.3.26–
27, 38–40). Here Dharma, finally in his own form, reveals in this final third test
of Yudhiṣṭhira’s human heart that he had earlier been the one testing him as
the Yakṣa and the dog.118
Clearly this beginning of a thematic analysis of the underlying values or
messages of the Mahābhārata’s upākhyānas takes us beyond our earlier classi-
fication of their content by their primary protagonists. Although a fuller
From the Rāmāyaṇa’s seven books, only a few matters bear summary in any
detail: the stories of great Ṛṣis who are the subject of upākhyānas in the Mahāb
hārata; and the relation of these Ṛṣis to other Ṛṣis, including Vālmīki. Attention
will thus be restricted to portions of Books 1–4, 6, and 7.
Book 1, the Bālakaṇḍa, opens, as we have seen, with the upodghāta, which
leads into a description of the Ikṣvāku dynasty, narrowing down to the one
defect in the long reign of its current monarch, Daśaratha: he is sonless. At
this time the Gods and Ṛṣis are alarmed by Rāvaṇa, who harasses the Ṛṣis in
their hermitages. With the help of a descendant of the sage Kaśyapa named
Ṛśyaśṛṅga (whose story is told in the Mahābhārata’s Ṛśyaśṛṅga-Upākhyāna),
Daśaratha’s three wives bear four sons, all partial incarnations of Viṣṇu. Once
the boys start their Vedic education, the Ṛṣi Viśvāmitra (whose story is told
in the Mahābhārata’s Vasiṣṭha- and Viśvāmitra-Upākhyānas and is mentioned
119 In Nala at 367.15; in the Pativratā-Upākhyāna at 3.198.87, 203.41, and 206.33; and in the
“Firesticks-Subtale” at 3.297.55, 71 (doubly), and 74, and 298. See Hiltebeitel, Rethinking
the Mahābhārata, 202–14, 230–31, 268–70.
120 Notable in upākhyānas 48 (Kapota) and 49 (Kṛtaghna).
121 See nn. 78 and 79 above. The guest/hospitality theme figures prominently in the upākhyā
nas of Books 12 and 13: notably in numbers 48 (Kapota), 49 (Kṛtaghna), 51 (Cirakāri), 54
(Uñchavṛtti), 55 (Sudarśana), and 65 (Kiṭa), as also in number 67 (Nakula) in Book 14. For
its still wider range, see Jamison, Sacrificed Wife/.Sacrificer’s Wife and Parida, Hospitality in
Changing Indian Society, as cited in n. 27 above.
50 Hiltebeitel
122 Sally J. Sutherland Goldman, “Gendered Narratives: Gender, Space, and Narrative Struc-
tures in Vālmīki’s Bālakāṇḍa,” in The Ramayana Revisited, ed. Mandakranta Bose (New
York: Oxford University Press, 2004), 72.
123 Also called the Jāmadagnya-Upākhyāna.
124 12.48–49; I insert “Jāmadagnya” in parentheses to distinguish this upākhyāna from the
Rāma-Upākhyāna about Rāma Dāśarathi, but the colophons give both of them the name
Rāma-Upākhyāna.
Not without Subtales 51
125 See Goldman and Sutherland, Bālakāṇḍa, 60 endorsing, especially with reference to the
Ṛśyaśṛṅga episode, long held views of the “‘purānic’ quality of the Bālakaṇda, as con-
trasted with the more ‘epic’ quality of Books Two through Six.” Cf. Brockington, The San-
skrit Epics, 132 for whom “the Ṛśyaśṛṅga episode probably has an independent origin,” and
the Bhārgava Rāma encounter is an “interpolated” “grotesque story.” Ibid., 478–79. In
another Brockington study, the Bharadvāja and Agastya episodes are “examples of inter-
polated episodes.” John L. Brockington, Epic Threads: John Brockington on the Sanskrit
Epics, ed. Gregory M. Bailey, and M. Brockington (Delhi: Oxford University Press, 2000),
299. Cf. Rosalind Lefeber, trans. Kiṣkindhā-Kāṇḍa, vol. 4 of The Vālmīki Rāmāyaṇa (Prince
ton, NJ: Princeton University Press, 1994), 346 and 349 on the Niśākara episode’s “late” and
“decidedly purāṇic type of story.”
126 Śatānanda is also a Gautama, providing a male presence from this family that the story
told does not provide.
127 See Hiltebeitel, “Authorial Paths through the Two Sanskrit Epics” for a fuller treatment of
this pattern as a “map” through the text, with discussion of the substitutions of Ṛśyaśṛṅga
for Kaśyapa and Rāma Jāmadagnya for Jamadagni.
52 Hiltebeitel
to Citrakūṭa. There, after long discussion, Bharata agrees to be regent for the
duration of Rāma’s exile.
Soon sensing disquiet among the Citrakūṭa Ṛṣis. Rāma learns that Rāvaṇa’s
younger brother Khara has been cannibalizing ascetics in nearby Janasthāna.
The sages retreat to a safer ashram and Rāma moves on to the ashram of Atri,
where Atri’s wife Anasūyā tells Sītā the duties of a faithful wife and gives her
apparel and jewels.128 Rāma gets his next directions from the ascetics there,
who recommend, all other routes being treacherous, “the path through the for-
est that the great Ṛṣis use when they go to gather fruits” (Rāmāyaṇa 111.19).
With this close of Book 2, adding Bharadvāja and Atri, Rāma has now been
linked with seven of the eight pravara Ṛṣis or their descendants. These original
seven, who together constitute the northern constellation of the Seven Ṛṣis
(the Big Dipper), have pointed Rāma south.
The first line of Book 3, the Araṇyakāṇḍa, finds the trio entering the “vast
wilderness” of Daṇḍaka. As they move on from a circle of ashrams, the rākṣasa
Virādha looms before them and seizes Sītā. Pained by seeing her touched,
Rāma fills Virādha with arrows and the brothers each break off an arm to
release her. Virādha realizes he has been slain by Rāma, which relieves him
from a curse. Before going to heaven, he tells Rāma that the great Ṛṣi Śarabhaṅga
“will see to your welfare” (Rāmāyaṇa 3.3.22–23). Śarabhaṅga relays Rāma to the
hermitage of Sutīkṣṇa, who offers his ashram as a residence; but Rāma says he
might kill the local game. The trio lives happily for ten years in another circle
of hermitages before returning to Sutīkṣṇa (10.21–26). Storytellers have now
told Rāma about Agastya’s āśrama and he asks Sutīkṣṇa how to find it in so vast
a forest (29–30). Sutīkṣṇa heads him due south, and along the way Rāma tells
Lakṣmaṇa stories told about Agastya that also occur in the Mahābhārata’s
Agastya-Upākhyāna. Rāma intends to live out the remainder of his exile with
Agastya (Rāmāyaṇa 3.10.86), but Agastya, after meditating a moment, says that
he knows Rāma’s true desire and directs him to a lovely forest called Pañcavaṭī
near the Godāvarī River where Sītā will be comfortable and Rāma can protect
her while safeguarding the ascetics (12.12–20). These words of the eighth, last,
and southernmost of the great pravara Ṛṣis resound with forebodings, as does
the trio’s meeting on the way to Pancavaṭī with the vulture Jaṭāyus, who offers
to keep watch over Sītā whenever Rāma and Lakṣmaṇa are away. However
128 On this theme, see Alf Hiltebeitel, Sītā Vibhūṣitā: The Jewels for Her Journey,” Indologica
Taurinensia 8–9, Ludwik Sternbach Commemoration Volume (1980–81): 193–200. The arti-
cles given by Anasūyā are additional to those given to Sītā by Daśaratha to cover the bark
that Kaikeyī has contemptuously given Sītā to wear over her silk (Rāmāyaṇa 2.33.5–12;
34.15–18).
Not without Subtales 53
kindly, a vulture is normally a bad omen (3.22.4). At Pañcavaṭī, the trio is soon
visited by Rāvaṇa’s sister Śūrpaṇakhā, and there, after one thing leads to
another, Sītā is carried off by Rāvaṇa while Jāṭayus is sleeping (3.48.1). Once
Rāvaṇa has met Jaṭayus’s challenge and picked Sītā up to continue on his way,
Brahmā, seeing this outrage with his divine eye, says, “What is done was to be
done,” and the Daṇḍaka Forest Ṛṣis are “thrilled” (prahṛṣṭāḥ) at the same sight
(50.10–11).
Jaṭāyus, unwinged after his fight with Rāvaṇa, soon tells the brothers before
he dies that Rāvaṇa abducted Sītā and went south. The brothers head south on
an “untrodden path” (Rāmāyaṇa 3.65.2), passing into the Krauñca Forest, still
hoping to find Sītā. Instead they run into a dānava-turned-rākṣasa, Kabandha:
“Headless trunk,” but also a name for a sacrificial post. He guards the way past
him as Virādha did for the Daṇḍaka Forest at this Book’s beginning (and as
Kirmīra and the Yakṣa do at the beginning and end of the Mahābhārata’s Book
3). Kabandha is a headless torso with a single-eyed129 face in his stomach, a
huge devouring mouth, and long grabbing arms that suddenly seize the broth-
ers, who quickly sever them. Realizing that this amputation by Rāma ends a
long curse, Kabandha tells his story, and after Rāma has asked if he knows any-
thing about Rāvaṇa and has cremated the demon, Kabandha rises lustrously
from his pyre to say that Rāvaṇa’s abode may be found if Rāma allies with the
monkey Sugrīva, whom Rāma should quickly make a friend and “commisera-
tor” (vayasva). Kabandha then directs them to Sugrīva’s haunt on Mount
Ṛṣyamūka. This path takes them through Mataṅga’s Wood to Mataṅga’s āśrama,
where all the Ṛṣis have passed away except the “mendicant woman” Śabarī
(“the Tribal Woman”). As Śabarī soon corroborates, Mataṅga and his disciples
ascended to heaven just when Rāma reached Citrakūṭa, but Śabarī has awaited
Rāma’s arrival so that she can go to heaven after seeing him. For this, Rāma
permits her to enter fire (70.26)—indexing an association between fire-entry
and purification that will also apply to Sītā.
Book 4, the Kiṣkindhākāṇḍa, then begins with Rāma exploring Mount
Ṛṣyamūka, being met by Hanumān, and making Sugrīva his friend and com-
miserator just as Kabandha had advised.130 In offering to find Sītā, Sugrīva
129 So too Dharma as the Yakṣa. One might connect this with their penetrating insight into
what Rāma and Yudhiṣṭhira need for their next adventures: friendship with Sugrīva and
the heart of the dice.
130 In precise detail: Kabandha had advised, “Quickly make him a comrade [vayasya], having
gone there now, Rāghava, sealing your compact in the presence of blazing fire to shun
all trickery” (Rāmāyaṇa 3.68.13); and now, “Sugrīva and Rāghava entered into vayasya
by reverently circling the blazing fire” (4.5.16). This fairly unusual term vayasya, literally
meaning “contemporary” but used only by Vālmīki in the sense of “commiserator” or
54 Hiltebeitel
expresses a one-sided willingness to die for his bond with Rāma (Rāmāyaṇa
8.9), and begins to give his side of a story that Vālin wronged him, which Rāma
accepts even before fully hearing it and promises to kill Vālin. The first part of
Sugrīva’s tale concerns his falling out with Vālin after Vālin had killed the
demon Māyāvin, which Rāma accepts without question. But behind this story
lies another by which Sugrīva discloses why Mount Ṛṣyamūka provides him
asylum. Māyāvin opposed Vālin because he had killed Māyāvin’s older brother,
“a buffalo named Dundubhi” (4.11.7), whom Vālin crushed until blood oozed
from his ears, hurling away the carcass. But “blood drops from the wounds fell
out from its mouth and were lifted by the wind toward Mataṅga’s hermitage”
(41). There Mataṅga cursed Vālin to be unable to enter his Wood on pain of
death. Sugrīva now points to Dundubhi’s bones, which Rāma kicks off to a
great distance with just his big toe. Mataṅga’s departure thus defines his her-
mitage, along with Mount Ṛṣyamūka, as a place cursed for its pollution. Albeit
that Mataṅga is a Ṛṣi, he is not a Vedic Ṛṣi or even a Brahman. Rather, just as
Śabarī’s name denotes the Tribal, his denotes the Untouchable. As with the
Mahābhārata’s Mataṅga-Upākhyāna,131 Dundubhi’s killing has behind it a buf-
falo sacrifice—a quite archaic one, with death by wrestling rather than the
sword—in which this “untouchable Ṛṣi” takes on the pollution of this non-
Vedic villagey rite.132
“sympathizer,” will continue to define Rāma and Sugrīva’s bond. I note Vāmīki’s unique
development of Sugrīva’s and Rāma’s friendship around this “sentiment” (which is not to
be found in the Rāmopākhyana) in Hiltebeitel, “Authorial Paths through the Two Sanskrit
Epics.”
131 There a she-ass discloses the unwelcome news to the young Mataṅga, who thinks himself
a Brahman, that because his Brahman mother slept with a Śūdra barber he is by birth a
Caṇḍāla or Untouchable, and Mataṅga spends the rest of his days doing tapas, unsuccess-
fully, to become a Brahman (Mahābhārata 13.30.13-14). It would seem that he cannot be
the same Mataṅga: being denied brāhmaṇya by Indra, he would likely be denied the pos-
sibility of being a Ṛṣi.
132 A myth linked with the village buffalo sacrifice in Karnataka shows that an old nexus may
link the stories of these two Mataṅgas: Back in the Tretāyuga when all of south India was
under the rule of Rāvaṇa and Brahmans had to perform ceremonies in secret, a Brāhmaṇī
discovered that her husband and his relatives were meat-eating and liquor-imbibing
Untouchables, and that her own two daughters shared their fare. Before submitting her-
self to flames and becoming the village goddess, she cut off her husband’s penis and put
it in his mouth, making him the prototype victim of the buffalo sacrifice at which his
maternal uncle would become the Potrāj charged with bearing off the rite’s pollution, and
his younger brother Gavanga the chief of the Ranigya musician-choristers (Āsādis)
charged with reviling the goddess with abusive songs—all of which would take place with
participation of Brahmans and local landlords in the place of Kṣatriyas. See Sir Walter
Not without Subtales 55
Rāma thus forges his friendship with Sugrīva in a place that is both cursed
and beyond the range of the Vedic Ṛṣis, who up to now have marked his trail.
On the one hand, since leaving Agastya, Rāma’s interventions have brought
grace and salvation to Kabandha, Śabarī, and Vālin, and a timetable for
Mataṅga to have vacated his hermitage and go to heaven before Rāma’s
arrival—of these, only Mataṅga is thus denied Rāma’s saving presence. On the
other, since meeting Jaṭāyus and the cannibal-post Kabandha, Rāma has met
only impure or inauspicious beings, including monkeys (according to Sītā
when she first sees Hanumān and thinks she is dreaming, “a monkey in a dream
is held by all the śāstras to be inauspicious” [Rāmāyaṇa 5.30.4; similarly 32.21]).
This pattern recurs toward the end of Book 4, where one learns that 8,000 years
earlier (4.59.9), the Ṛṣi Niśākara (“Night-Maker”) welcomed Jaṭāyus’s vulture
brother Sampāti, wingless after a misadventure, to his āśrama where wild ani-
mals—bears, deer, tigers, lions, elephants, and snakes—surrounded him as
they would a benefactor (dātṛ). Like Mataṅga, he went to heaven rather than
wait to see Rāma (61.15), and, although he could have restored Sampāti’s wings,
he left him wingless so he would have to stay on the spot until it was time to
benefit Rāma. Sampāti is thus there to see the monkey search party that is
looking for Sītā in the south stop and think of fasting to death because they
have failed to find her. Thinking better of eating this tempting prey, Sampāti
tells the monkeys that he and his son saw Rāvaṇa taking Sītā to Laṅkā—a
vulture-Ṛsi collaboration that thus cues Hanumān’s leap to Laṅkā.133
Elliot, Aboriginal Caste Book, vol. 2 (London: India Office Library, 1821–60), 675–81 for this
rich and little known version summarized in Alf Hiltebeitel, “Sexuality and Sacrifice: Con-
vergent Subcurrents in the Fire-walking Cult of Draupadī,” in Images of Man: Religion and
Historical Process in South Asia, ed. Fred W. Clothey (Madras: New Era Publications, 1982),
88–91 and 109 nn. 81–82. The Rāmāyaṇa’s Mataṅga is thus like the Potrāj a handler of the
impurity of a buffalo sacrifice (see Alf Hiltebeitel, “Rāma and Gilgamesh: The Sacrifices of
the Water Buffalo and the Bull of Heaven,” History of Religions 19 [1980]: 200–23). And the
Mahābhārata’s Mataṅga becomes a kind of chorister, for after he fails to become a Brah-
man, he asks Indra to be able to rove at pleasure through the heavens honored by Brah-
mans and Kṣatriyas and able to assume any form at will, and Indra gives him the boon of
becoming Chandadeva, God of Chandas verses, and of being adored by women (Mahāb
hārata 13.30.13–14).
133 It is interesting that Nightmaker points the way to Laṅkā, since Hanumān’s leap to Laṅkā
follows a route “adorned with planets, constellations, the moon, the sun, and all the hosts
of stars … thronged with hosts of great seers” (Goldman, trans. Sundarakāṇḍa, 112, trans-
lating 5.1.161). One wonders whether the animals that went around (parīvārya upagac-
chanti, Rāmȳaṇa 5.59.15c) Niśākara’s ashram do not hint at the constellations. Cf. n. 78
above on Niśākara, whose knowledge of Rāma’s future from “of old” might suggest that it
had been “in the stars.” On Laṅkā as an “astronomical conundrum” and location, see Alf
56 Hiltebeitel
In the Yuddhakāṇḍa (Book 6), Rāvaṇa seems unable to focus on Rāma or the
war until his wise maternal grandfather Mālyavān, counseling peace with
Rāma and Sītā’s return, says the gods and Ṛṣis desire Rāma’s victory, differenti-
ates dharma and adharma as divine and demonic, alludes to the Mahābhārata
idea that the king defines the age (yuga),134 says that throughout the regions
the Ṛṣis are performing fiery Vedic rites and austerities that are damaging the
rākṣasas, foresees the rākṣasas’ destruction, and concludes, “I think Rāma is
Viṣṇu abiding in a human body” (6.26.6–31), Getting it right, Mālyavān calls
attention to the Ṛṣis’ labors to affect the war’s outcome and provides analogs to
features of the Bhagavadgītā: a theology for the war about to happen; a predic-
tion of its outcome; and a disclosure of the hidden divinity behind it—in this
case, hidden so far mainly from himself.
Once the war is won and Rāma has accepted Sītā after her fire ordeal, after
finally learning that he is Viṣṇu, everyone (monkeys and rākṣasas included)
heads toward Ayodhyā on the Puṣpaka chariot, stopping along the way at
Bharadvāja’s āśrama where Bharadvāja recounts the trio’s whole adventure,
which he knows by his penances (Rāmāyaṇa 6.112.14). Rāma is at last enthroned
in the presence of his rejoicing family and people and the monkeys, rākṣasas,
and ṛṣis. Twice it is said that he ruled for 10,000 years (82, 90), the second time
in this Book’s very last words—surely sounding like a happy ending, as many
western scholars and some Indian vernaculars have taken Book 6 to be.
But the Uttarakāṇḍa (Book 7) opens with Rāma just consecrated and a series
of departures and dismissals. First, the Ṛṣis come to his palace—Agastya and
the original Seven among them (Rāmāyaṇa 7.1.3–4). Rāma asks about the
rākṣasas he conquered, launching their former near-neighbor Agastya on a
lengthy rākṣasa genealogy, with tales of Rāvaṇa’s boon and his violations of
women, including what some Northern manuscripts call the Vedavatī-
Upākhyāna (7.17) (see above n. 46). Rāma is repeatedly filled with wonder.
Then “all the Ṛṣis went as they came” (36.46). Rāma also dispatches a hundred
kings, and the rākṣasas, monkeys, and bears—Hanumān parting with the
famous words: “As long as I hear Rāma-kathā on the face of the earth, so long
will my breaths reside in my body” (39.16). Next Rāma dismisses the Puṣpaka
chariot while keeping it on call. And next he dismisses Sītā who will not remain
on call. All these dismissals subtract down to a great unraveling.
Hiltebeitel Rethinking India’s Oral and Classical Epics: Draupadī among Rajputs, Muslims,
and Dalits (Chicago: University of Chicago Press, 1999), 89–90 and nn. 5 and 6, 93–94 and
n. 18.
134 See Madeleine Biardeau, “Études de mythologie hindoue IV: II. Bhakti et avatāra,” Bulletin
de l’École Française d’Extrême-Orient 63 (1976): 155–71.
Not without Subtales 57
After some happiness between Rāma and Sītā, there comes the news that
Ayodhyā’s citizens gossip about her time in captivity, and Rāma banishes her to
protect his royal reputation even after she has announced that she is pregnant.
Painfully, Lakṣmana leaves her at Vālmīki’s hermitage. Next Rāma hears that
there are still some ascetics who live in fear of a rākṣasa named Lavaṇa.
Śatrughna goes to tackle Lavaṇa, and stops over in Vālmīki’s leafy hut on the
night Sītā gives birth to the twins.135 At dawn he resumes his journey,136 kills
Lavaṇa, and establishes a kingdom at Mathurā. Twelve years later he decides to
visit Ayodhyā. On the way, in a passage rejected by the Critical Edition even
though it appears in all the manuscripts collated, he stops at Vālmīki’s, over-
hears the twins’ elegant recitals, and promises that he and his army will keep
their birth secret (7, Appendix 1, no. 9137). When Śatrughna sees Rāma, he men-
tions nothing about Vālmīki, Sītā, or the twins.
Finding himself once again in Agastya’s hermitage after going by the Puṣpaka
to behead the Śūdra Śambūka, Rāma listens to more of Agastya’s stories,138 and
returns to Ayodhyā, again dismissing the Puṣpaka. He now tells Bharata and
Lakṣmaṇa he wishes to perform a Rājasūya sacrifice, but Bharata tells him a
horse sacrifice is less destructive and Lakṣmaṇa that the aśvamedha removes
all sins and purifies (Rāmāyaṇa 75.2). Rāma approves the aśvamedha. He orders
Lakṣmaṇa to make invitations to the monkeys and rākṣasas, and to the regional
ṛṣis and their wives, and to prepare a vast sacrificial enclosure in the Naimiṣa
Forest. Bharata is to lead a procession trailed by all the mothers from the inner
apartments and “my golden wife [kāñcanīm mama patnīm] worthy of conse-
cration [dīkṣā] in sacrificial rites” (19). Sītā thus has a replacement-statue even
while still alive.139 With the sacrifice proceeding, Vālmīki suddenly arrives with
his disciples (84. 1) and directs the twins to sing “the whole Rāmāyaṇa poem at
the gate of Rāma’s dwelling” (3–5)—twenty sargas a day (9). Rāma hears the
boys sing the first twenty sargas beginning “from the sight of Nārada [nārada-
135 Earlier that evening he hears Vālmīki tell what some Northern manuscripts call the
Saudāsa-Upākhyāna (Rāmāyaṇa 7.57).
136 Stopping at an āśrama on the Yamunā, he hears from Cyavana what some Northern man-
uscripts call the Mandhātṛ-Upākhyāna (Rāmāyaṇa 7.59).
137 U.P. Shah, ed. The Uttarakāṇḍa: The Vālmīki Rāmāyaṇa (Baroda: University of Baroda,
1975), 26–27.
138 Including what some Northern manuscripts call the Śveta-Upākhyāna (Rāmāyaṇa 7.69),
which builds up to an account of the Daṇḍaka Forest.
139 Amid these preparations, Rāma, with a smile, tells Lakṣmaṇa what some Northern manu-
scripts call the Iḍa/.Ilā-Upākhyāna (Rāmāyaṇa 7.78–79) about a king who alternates being
male and female. Perhaps in Sītā’s absence, Rāma suggests a curiosity about bisexual self-
sufficiency.
58 Hiltebeitel
darśanāt]” (11), that is, from the beginning of the upodghāta on. Once the twins
tell Rāma who authored this poem that contains his whole adventure (19), they
offer to continue singing it at intervals in the rite (21). After many days, Rāma
recognizes them, misses Sītā, and summons her to attest to her purity by oath
in the midst of the great Ṛṣis, rākṣasas, and monkeys, plus unnamed kings and
the four castes in thousands (87.6–7). But when Vālmīki brings Sītā he attests
to her purity himself (19), and tells Rāma only that “she will give proof of her
fidelity” (15, 20). No longer demanding the oath just announced, Rāma accepts
Vālmīki’s word as tantamount to being Sītā’s: “Surely I have proof of fidelity in
your stainless words. Surely Vaidehī gave proof of fidelity formerly in the pres-
ence of the gods” (88.2–3)—who by now have also come to witness (5–7).
Indeed, in a phrase that occurs nowhere else in either epic, this conclave occurs
“in the middle of the universe [jagato madhye]” (1, 4). Not demanded to make
an oath, Sītā nonetheless makes one implicitly in her only and last words: “If I
have thought with my mind of none other than Rāma, let the goddess Mādhavī
[Earth] give me an opening …” (10). Rāma, who had hoped for “affection” (priti)
from Sītā (4), has thus accepted the author’s word as Sītā’s only to be over-
whelmed with grief and horror by what her word—and the poet’s—actually is.
This is the moment at which he comes to realize what it means to be caught up
in his own story, which, if he heard it from the frame on, as we are told, he
would know to have also been Sītā’s story and to have been inspired by the grief
of a female bird. Rāma now threatens to destroy the Earth unless she returns
Sītā intact (7, Appendix I, No. 13, lines 18–20) until Brahmā repeats what he
told him after Sītā’s fire ordeal, that he is Viṣṇu, and invites him to listen with
the great Ṛṣis to the rest of this “first poem,” which will now tell what is still to
happen (21–40). Once Brahmā returns to heaven, the Ṛṣis in Brahmaloka
obtain his permission to return for the rest as well (43–49). The heavenly Ṛṣis
of Brahmaloka thus come to an earthly Naimiṣa Forest to hear the end of the
Rāmāyaṇa, whereas in the Mahābhārata the Ṛṣis of the Naimiṣa Forest seem to
be in the heavens when they have the Mahābhārata at last brought to them.140
Though the Critical Edition rejects this sarga, it does so only on the grounds
that without it “the continuity of the narration … is not hampered and appears
in a better order.”141 For Rāma, the relation between Sītā’s two ordeals seems to
be that whereas his first self-recognition as Viṣṇu emerges out of a human
identity crossed with uncertainty and confusion as to his own all-too-human
emotions, his second comes after he has learned of his divinity and has repeat-
edly pared his life down to a perfect rule through his repeated dismissals of
others, yet without consideration of what this has cost him since the banish-
ment of his wife—not to mention what it has cost her. If so, the poem could be
saying that Vālmīki’s initial question to Nārada—whether there is an ideal man
today—was not really convincingly answered.
Once the aśvamedha ends, Rāma finds the universe empty without Sītā and
again dismisses the kings, bears, monkeys, and rākṣasas (Rāmāyaṇa 89.1). The
ṛṣis seem to take care of themselves. Rāma never remarries, but at all his sacri-
fices there is a golden Sītā (jānakī kāncanī; 4). For 10,000 years he rules a
harmonious kingdom. Finally Death or Time (Kāla) comes to him as a messen-
ger from Brahmā and tells him they must meet alone; anyone hearing them
must be killed. While Rāma posts Lakṣmaṇa at the door, “Time who destroys
all” (94.2) tells Rāma it is time to return to heaven as Viṣṇu. As the two con-
verse, the congenitally ravenous “blessed Ṛṣi Durvāsas” (95.1b), familiar from
the Mahābhārata’s Mudgala-Upākhyāna,142 tries to barge in, threatening to
curse the kingdom if he is prevented. Lakṣmaṇa chooses his own death rather
than allowing that of others and admits him. Durvāsas only wants something
to eat after a thousand-year fast, which Rāma happily provides. At Vasiṣṭha’s
advice Rāma then banishes Lakṣmaṇa as equivalent to death, and Lakṣmaṇa,
meditating by the Sarayū River, is taken up to heaven. After Rāma divides
Kosala into two kingdoms to be ruled by Kuśa and Lava, he enters the Sarayū
and resumes his divine form, followed in this by Bharata and Śatrughna.
as an epitome of the Rāmāyaṇa, since it lacks the structure that the Rāmāyaṇa
shares with the Mahābhārata.
Second, this article holds that it is helpful to reflect on how upākhyāna
material is used in both epics. As observed, the Rāmāyaṇa uses this term only
in an interpolation and in Northern Recension colophons. Rather than having
stand-out “subtales,” the Rāmāyaṇa folds all its secondary narratives into one
consecutively unfolding poem. This is especially noteworthy in its stories
about the eight great Ṛṣis encountered by Rāma, many of which include mate-
rial that the Mahābhārata relates in its upākhyānas. Other than mentioning
Vasiṣṭha, a fixture in the Ikṣvāku house, the Rāmopākhyāna does not know
these Ṛṣis. It has no Ṛśyaśṛṅga, Viśvāmitra, Gautama and Ahalyā, Rāma Jāma
dagnya, or for that matter Vasiṣṭha involved in the stories from birth through
marriage; just this: “In the course of time [Daśaratha’s] sons grew up very vigor-
ous, and became fledged in the Vedas and their mysteries and in the art of
archery. They completed their student years, and took wives” (Mahābhārata
3.261.4–5). It has no Bharadvāja; just this of Bharata: “He found Rāma and
Lakṣmaṇa on Mount Citrakūṭa” (216.63). And from Citrakūṭa on, there is not a
peep from Atri and Anasūyā or Agastya. There is also no Vālmīki, Mataṅga, or
Niśākara. It is improbable that the Rāmopākhyāna would have strained out all
these figures and episodes if it were a Rāmāyaṇa epitome. Vālmīki would seem
to have worked such upākhyāna material into something he claims to be new:
kāvya, “the first poem.” And this would seem to be the best way to think about
what he did with the Rāmopākhyāna: go beyond it to author a poem in which
Rāma and Sītā move through their double adventure along paths signposted
by Ṛṣis who impart Vedic authority to new values about dharma centered on
bhakti as a servant-master relation of subjects to a divinity-embodying king.
These knowing Vedic Ṛṣis represent “all the Ṛṣis” high and low who motivate
this divine incarnation to cleanse the world of noxious rākṣasas, and ulti-
mately come to hear out his story to the end. And they in turn are represented
by Vālmīki himself who frames all the paths that Rāma and Sītā take as ones
that begin with his inspiration to tell their adventures in a poem that will lead
them ultimately back to him.
As to the Mahābhārata, we began with the question of what a “Bhārata”
without upākhyānas might have signified, and looked at how and where
upākhyānas are woven into the Mahābhārata. There is, however, one other ref-
erence to the epic’s upākhyānas that is yet to be plumbed. It occurs toward the
end of Book 12 in the highly devotional Nārāyaṇīya, and takes us back where
we began: to the “oceanic mind” of the author, and also to the Āstīkaparvan
Not without Subtales 61
substory called “The Churning of the Ocean” (1.15 17).143 One may also recall
that on the last day of the war, Duryodhana, hiding from the Pāṇḍavas to recu-
perate, finds his last relief by magically concealing himself in an otherwise
unheard of Dvaipāyana Lake (9.29.53a), that is, a lake bearing the name of the
author.144
About one third of the way through the Nārāyaṇīya, itself an eighteen-
chapter epitome of the Mahābhārata (although the Critical Edition splits
a chapter and makes it nineteen [Mahābhārata 12.321–39]),145 Bhīṣma says
that the story he has just told Yudhiṣṭhira about Nārada’s journey to “White
Island” (Śvetadvīpa)—an island somewhere on the northern shore of the milky
ocean146—is a “narrative [ākhyānam] coming from a seer-based transmission
[ārṣeyam pāramparyāgatam] that should not be given” to anyone who is not
a Viṣṇu devotee (12.326.113), and, moreover, that it is the “essence” of all the
“other upākhyānas” he has transmitted:
143 Of which the Rāmāyaṇa has a short version as well (Rāmāyaṇa 1.44.13–27). The Mahābhā
rata provides no genre term or independent title for its “Churning of the Ocean” story; see
Gombach, “Ancillary Stories in the Sanskrit Mahābhārata,” part 2, 11, n. 9.
144 Just after Vyāsa suddenly appears on the battlefield to rescue Saṃjaya so that his all-see-
ing bard can return to the city and continue narrating events to Dhṛtarāṣṭra (9.28.35–39),
Saṃjaya meets Duryodhana alone and tells him of his narrow escape “through the grace
of Dvaipāyana” (42–43). When Saṃjaya keeps Duryodhana uninformed about his three
remaining allies, even though Saṃjaya has just seen them, Duryodhana tells him to tell
his father he has entered a lake (47–49), which he then does, solidifying the waters by his
māyā or power of illusion (52). Eventually, goaded by Yudhiṣṭhira, Duryodhana breaks up
through the solidified waters shouldering his iron mace (31.36). See Hiltebeitel, Rethinking
the Mahābhārata, 59–62.
145 On this point, which calls for a correction of the Critical Edition, see Hiltebeitel, “The
Nārāyaṇīya and the Early Reading Communities of the Mahābhārata.”
146 12.322.8; 323.21; 326.126, placing it in the general vicinity of the Horse’s Head, the subject
of the upākhyāna narrated later in the Nārāyaṇīya by Ugraśravas to the Naimiṣa Forest
Ṛṣis about Viṣṇu’s manifestation (Mahābhārata 12.335) and also mentioned in two other
upākhyānas (see n. 47 above).
62 Hiltebeitel
Hearing this, Yudhiṣṭhira and all the Pāṇḍavas become Nārāyaṇa devotees
(Mahābhārata 12.326.121). This suggests that one could count the “White Island”
story as a sixty-eighth upākhyāna. Furthermore, Bhīṣma holds that it is the
essence of them all. He has also used ākhyāna and upākhyāna interchangeably
with each other and with kathā (story). And when he speaks of the “hundreds
of other virtuous upākhyānas that are heard from me,” he probably implies not
only those he has just told Yudhiṣṭhira in the Śāntiparvan, but all the others he
has told or will tell elsewhere, and those that have been recited by others,
which Bhīṣma, given his many heavenly and earthly sources,147 would almost
certainly know as well.
Still within the Nārāyaṇiya, just after its next major narrative on pravṛtti and
nivṛtti, Śaunaka148 says to Ugraśravas:
O Sauti, very great is the narrative [ākhyāna] recited by you, having heard
which, the sages are all gone to the highest wonder.149[…] Surely having
churned the supreme ocean of knowledge by this hundred-thousand
(verse) Bhārata narrative with the churning of your thought [idaṃ
śatasahasrād hi bhāratākhyāna vistarāt / āmathya matimanthena jñāno
dadhim anuttamam]—as butter from milk, as sandal from Mount Malaya,
and as Āraṇyaka (forest instruction) from the Vedas, as nectar from
herbs—so is this supreme nectar of story [kathāmṛtam] … raised up [as]
spoken by you, which rests on the story of Nārāyaṇa [nārāyaṇakathā
śrayam]. (Mahābhārata 12.331.1-4)
147 See Alf Hiltebeitel, “Bhīṣma’s Sources,” in Vidyārṇavavandanam: Essays in Honor of Asko
Parpola, ed. Klaus Karttunen and Petteri Koskikallio, Studia Orientalia, no. 94 (Helsinki:
Finnish Oriental Society, 2001), 261–78 making the point that Bhīṣma’s youth spent with
his mother, the heavenly Gaṅgā, may have provided him a special educational opportu-
nity to meet celestial sages.
148 Again correcting the Critical Edition, which makes the speaker Vaiśampāyana; see Hilte-
beitel, “The Nārāyaṇīya and the Early Reading Communities of the Mahābhārata.”
149 The Critical Edition omits a long section here that should probably be restored.
Not without Subtales 63
Untraditional and unprecedented, the secret of all the Vedas, this treatise
[śāstra], of which everyone can convince himself, is further instruction
for my son [putrānuśāsanam]. By churning the wealth that is contained
in all the narratives [ākhyānas] about dharma and all the narratives
about truth, as also the ten thousand Ṛcs,152 this nectar has been raised
[dharmākhyāneṣu sarveṣu satyākhyāneṣu yad vasu/daśedam ṛsahasrāṇi
nirmathyāmṛtam uddhṛtam]—like butter from curds and fire from wood,
as also the knowledge of the wise, even has this been raised for the sake
of my son [putrahetoḥ samuddhṛtam]. (Mahābhārata 12.238.13–15)
The churning metaphor thus finds Vyāsa at its bottom, since he would be the
first to use it—before Bhīṣma or Ugraśravas.153 Indeed, Śuka is born when
Vyāsa sees a nymph and ejaculates his semen onto his churning firesticks
150 As the Critical Edition registers and actually prefers, some mss. attribute these words not
to Sauti but to Vaiśaṁpāyana.
151 This point is developed in Hiltebeitel, Rethinking the Mahābhārata, chapter 8, especially
pp. 284–85, 316. Note that Fitzgerald concedes that the Śuka story, like “Nala,” is “sugges-
tive of ‘fiction’.” James L. Fitzgerald, “The Many Voices of the Mahābhārata,” review of
Rethinking the Mahābhārata: A Reader’s Guide to the Education of the Dharma King, by Alf
Hiltebeitel, Journal of the American Oriental Society 123, no. 4 (2003): 807. Fitzgerald is on
uncertain ground when he says that Vaiśaṁpāyana’s response to Janamejaya’s questions
“cannot be a understood as verbatim repetitions of Vyāsa’s composition.” Fitzgerald, “The
Rama Jāmadagnya ‘Thread’ of the Mahābhārata,” 99, n. 23. Logically he is right, but in fic-
tion, not to mention futurist fiction, strange things are possible. The Mahābhārata poets
finesse this by having both Vaiśaṁpāyana and Ugraśravas relate Vyāsa’s “entire thought.”
152 Presumably Ṛgvedic mantras.
153 Though perhaps not before the ten thousand Ṛes or the “sages” (vipras) mentioned at the
end of the passage just quoted from Bhīṣma at Mahābhārata 12.326.15.
64 Hiltebeitel
References
Primary Sources
The Mahābhārata for the First Time Critically Edited. Edited by Vishnu S. Sukthankar,
Shripad K. Belvalkar, Parashuram L. Vaidya, et al. 19 vols. Poona: Bhandarkar Oriental
Research Institute, 1933–66.
The Mahābhārata, vol. 1. 1. The Book of the Beginning. Translated by J.A.B. Buitenen.
Chicago: University of Chicago Press, 1973.
The Mahābhārata, vol. 2. 2. The Book of the Assembly Hall; 3. The Book of the Forest.
Translated by J.A.B. Buitenen. Chicago: University of Chicago Press, 1975.
The Mahābhārata, vol. 3. 4. The Book of Virāṭa; 5. The Book of the Effort. Translated by
J.A.B. Buitenen. Chicago: University of Chicago Press, 1978.
The Mahābhārata, vol. 7. 11. The Book of Women; 12. The Book of Peace, Part 1. Translated
by James L. Fitzgerald. Chicago: University of Chicago Press, 2004.
The Mahabharata of Krishna-Dwaipayana Vyasa Translated into English from the Original
Sanskrit Text. Translated by Kisari Mohan Ganguli. 5th ed. New Delhi: Munshiram
Manoharlal Publishers, 1990–91.
Secondary Sources
Bedekar, V.M. “The Legend of Cirakārin in the Skanda Mahāpurāṇa and the Mahābhā
rata.” Purāṇa 11, no. 2 (1962): 210–28.
———. “The Place of Japa in the Mokṣadharma Parvan (Mahābhārata 12.189–93) and
the Yogasūtras.” Annals of the Bhandarkar Oriental Research Institute 44 (1963):
63–74.
Biardeau, Madeleine. “Un certain Kīcaka.” In Lex et Litterae: Studies in Honour of Professor
Oscar Botto, edited by Siegfried Lienhard and Irma Piovano, 35–52. Alessandria:
Edizioni dell’Orso, 1997.
———. “Compte rendu of seminar on the Mahābhārata.” Annuaire de l’École Pratique
des Hautes Études, Section 5, no. 79 (1971–72): 139–47.
———. “Études de mythologie hindoue IV: II. Bhakti et avatāra.” Bulletin de l’École
Française d’Extrême-Orient 63 (1976): 111–263.
———. “Études de mythologie hindoue V: II. Bhakti et avatāra (suite).” Bulletin de l’École
Française d’Extrême Orient 65 (1978): 87–238.
———. Le Mahābhārata: Un récit fondateur du brahmanisme et son interprétation,
2 vols. Paris: Éditions du Seuil, 2002.
———. “Nala et Damayantī. Héros épiques. Part 1.” Indo-Iranian Journal 27 (1984):
247–74.
———. “Nala et Damayantī. Héros épiques. Part 2.” Indo-Iranian Journal 28 (1985): 1–34.
———. Le Rāmāyaṇa de Vālmīki. Paris: Gallimard, 1999.
———. “Some Remarks on the Links between the Epics, the Purāṇas and their Vedic
Sources.” In Studies in Hinduism: Vedism and Hinduism, edited by Gerhard Ober
hammer, 77–119. Vienna: Verlag der Österreichischen Akademie der Wissenschaften,
1997.
Brockington, John L. Epic Threads: John Brockington on the Sanskrit Epics, edited by
Gregory M. Bailey and Mary Brockington. Delhi: Oxford University Press, 2000.
———. The Sanskrit Epics. Leiden: E.J. Brill, 1998.
Conze, Edward. The Prajñāpāramitā Literature. The Hague: Mouton, 1960.
Custodi, Andrea. “Dharma and Desire: Lacan and the Left Half of the Mahābhārata.”
PhD diss., George Washington University, 2004.
Dandekar, R.N. “Introduction.” In The Anuśāsanaparvan For the First Time Critically
Edited, vol. 1. Pune: Bhandarkar Oriental Research Institute, 1966.
Dange, Sadashiv Ambadas. Legends in the Mahābhārata. Delhi: Motilal Banarsidass,
1969.
Elliot, Sir Walter. Aboriginal Caste Book. 3 Handwritten vols. London: India Office Library,
1821–60.
———. “Mahābhārata.” In The Hindu World, edited by Sushil Mittal and Gene Thursby,
52-74. New York: Routledge, 2004.
66 Hiltebeitel
———. “The Many Voices of the Mahābhārata.” Review of Rethinking the Mahābhārata:
A Reader’s Guide to the Education of the Dharma King, by Alf Hiltebeitel, Journal of
the American Oriental Society 123, no. 4 (2003): 803–18.
———. “Negotiating the Shape of ‘Scripture’: New Perspectives on the Development
and Growth of the Mahābhārata between the Empires.” In Between the Empires:
Society in India 300 bce to 400 CE, edited by Patrick Olivelle, 257–86. Oxford: Oxford
University Press, 2006.
———. “The Rama Jāmadagnya ‘Thread’ of the Mahābhārata: A New Survey of Rāma
Jāmadagnya in the Pune Text.” In Stages and Transitions: Temporal and Historical
Frameworks in Epic and Purāṇic Literature. Proceedings of the Second Dubrovnik
International Conference on the Sanskrit Epics and Purāṇas, Aug. 1999, edited by
Mary Brockington, 89–132. Zagreb: Croatian Academy of Sciences and Arts, 2002.
Franco, Eli. The Spitzer Manuscript: The Earliest Philosophical Manuscript in Sanskrit. 2
vols. Vienna: Verlag der Österreichischen Akademie der Wissenschaften, 2004.
Ganguli, Kisari Mohan, trans. The Mahabharata, 12 vols. 4th ed. Delhi: Munshiram
Manoharlal, 1981.
Ghosh, Aurobindo. “Notes on the Mahābhārata.” In On the Mahabharata, 1–83.
Pondicherry: Sri Aurobindo Ashram, 1991.
Goldman, Robert P., trans. Sundarakāṇḍa. Vol. 5 of The Rāmāyaṇa of Vālmīki: An Epic of
Ancient India. Edited by Robert P. Goldman and Sally Sutherland Goldman. Prince
ton, NJ: Princeton University Press, 1996.
Goldman, Robert P. and Sally Sutherland, trans. Bālakāṇḍa. Vol. 1 of The Rāmāyaṇa of
Vālmīki: An Epic of Ancient India. Edited by Robert P. Goldman and Sally Sutherland
Goldman. Princeton, NJ: Princeton University Press, 1984.
Gombach, Barbara. “Ancillary Stories in the Sanskrit Mahābhārata.” PhD diss., Columbia
University, 2000.
Hiltebeitel, Alf. “Authorial Paths through the Two Sanskrit Epics, Via the Rāmopākhyāna.”
In Epic Undertakings: Papers of the 12th World Sanskrit Conference, vol. 2, edited by
Robert P. Goldman and Muneo Tokunaga, 169–214. Delhi: Motilal Banarsidass, 2009.
———. “Bhīṣma’s Sources.” In Vidyārṇavavandanam: Essays in Honor of Asko Parpola,
edited by Klaus Karttunen and Petteri Koskikallio, 261–78. Studia Orientalia, no. 94.
Helsinki: Finnish Oriental Society, 2001.
———. The Cult of Draupadī, vol. 2: On Hindu Ritual and the Goddess. Chicago: University
of Chicago Press, 1991.
———. “The Emergence of Kāvya out of the Two Sanskrit Epics.” Paper presented at
the International Seminar on the Origins of Mahākāvya, Università degli Studi di
Milano, June 4–5, 2004.
———. “More Rethinking the Mahābhārata: Toward a Politics of Bhakti.” Indo-Iranian
Journal 47 (2004): 203–27.
Not without Subtales 67
———. “The Nārāyaṇīya and the Early Reading Communities of the Mahābhārata.” In
Between the Empires: Society in India, 300 bce to 400 CE, edited by Patrick Olivelle,
227–56. New York: Oxford University Press, 2006.
———. “On Reading Fitzgerald’s Vyāsa.” Journal of the American Oriental Society 125,
no. 2 (2005): 241–61.
———. “Rāma and Gilgamesh: The Sacrifices of the Water Buffalo and the Bull of
Heaven.” History of Religions 19 (1980): 187–223.
———. Rethinking India’s Oral and Classical Epics: Draupadī among Rajputs, Muslims,
and Dalits. Chicago: University of Chicago Press, 1999.
———. Rethinking the Mahābhārata: A Reader’s Guide to the Education of the Dharma
King. Chicago: University of Chicago Press, 2001.
———. The Ritual of Battle: Krishna in the Mahābhārata. Albany: State University New
York Press, 1990 [1976].
———. “Sexuality and Sacrifice: Convergent Subcurrents in the Fire-walking Cult of
Draupadī. In Images of Man: Religion and Historical Process in South Asia, edited by
Fred W. Clothey, 72–111. Madras: New Era Publications, 1982.
———. “Sītā Vibhūṣitā: The Jewels for Her Journey.” Indologica Taurinensia 8–9, Ludwik
Sternbach Commemoration Volume (1980–81): 193–200.
———. “Śiva, the Goddess and the Disguises of the Pāṇḍavas and Draupadī.” History of
Religions 20 (1985): 147–74.
———. “The Two Kṛṣṇas on One Chariot: Upaniṣadic Imagery and Epic Mythology.”
History of Religions 24 (1984): 1–26.
———. “Weighting Orality and Writing in the Sanskrit Epics.” In Epics, Khilas, and
Purāṇas: Continuities and Ruptures, Proceedings of the Third Dubrovnik International
Conference on the Sanskrit Epics and Puranas, September 2002, edited by Petteri
Koskikallio, 81–111. Zagreb: Croatian Academy of Sciences and the Arts, 2005.
Hopkins, Edward Washburn. The Great Epic of India: Its Character and Origin. Calcutta:
Punthi Pustak, 1969. First published 1902 by Charles Scribner’s Sons.
Jamison, Stephanie. The Rig Veda between Two Worlds. Le Ṛg Veda entre deux mondes.
Paris: Boccard, 2007.
———. Sacrificed Wife/Sacrificer’s Wife: Women, Ritual and Hospitality in Ancient India.
New York: Oxford University Press, 1996.
Kantawala, S.G. “The Legend of Aṇī Māṇḍavya.” In Modern Evaluation of the Mahābhā
rata. Prof. P.K. Sharma Felicitation Volume, edited by S.P. Narang, 101–9. Delhi: Nag
Publishers, 1995.
Lefeber, Rosalind, trans. Kiṣkindhā-Kāṇḍa. Vol. 4 of The Vālmīki Rāmāyaṇa. Princeton,
NJ: Princeton University Press, 1994.
Leslie, Julia. “A Bird Bereaved: The Identity and Significance of Vālmīki’s Krauñca.”
Journal of Indian Philosophy 26 (1998): 455–87.
68 Hiltebeitel
Lévi, Sylvain. “Tato Jayam Udirayet.” Translated L.G. Khare. Annals of the Bhandarkar
Oriental Research Institute 1 (1917): 13–20.
Lubotsky, Alexander. A Ṛgvedic Word Concordance. 2 Parts. American Oriental Series,
Vols. 82–83. New Haven, CT: American Oriental Society, 1997.
Minkowski, Christopher, Z. “Janamejaya’s Sattra and Ritual Structure.” Journal of the
American Oriental Society 109, no. 3 (1989): 401–20.
Monier-Williams, Monier. A Sanskrit-English Dictionary. Oxford: Clarendon Press, 1964.
Parida, Sarat Chandra. Hospitality in Changing Indian Society: Vedic Age to Puranic Age.
Delhi: Bharatiya Vidya Prakashan, 2004.
Pathak, Shubha. “The Things Kings Sing: The Religious Ideals of Poetic Rulers in Greek
and Sanskrit Epics.” PhD diss., University of Chicago, 2005.
Patton, Laurie L. Myth as Argument: The Bṛhaddevatā as Canonical Commentary.
Religionsgeschichtliche Versuche und Vorarbeiten, vol. 41. Berlin and New York:
Walter de Gruyter, 1996.
Pollock, Sheldon I. “Ātmānam mānuṣaṃ manye: Dharmākūtam on the Divinity of
Rāma.” Journal of the Oriental Institute of Baroda 33 (1984): 505–28.
———, trans. Araṇyakāṇḍa. Vol. 3 of The Rāmāyaṇa of Vālmīki: An Epic of Ancient India.
Edited by Robert P. Goldman. Princeton, NJ: Princeton University Press, 1991.
Sastri, P.P.S. ed. The Mahābhārata (Southern Recension) Critically Edited by P.P.S. Sastri.
Madras: Ramaswamy Sastrulu & Sons, Madras, 1931.
Schlingloff, Dieter. “The Oldest Extant Parvan-list of the Mahābhārata.” Journal of the
American Oriental Society 89 (1969): 334–38.
Shah, U.P., ed. The Uttarakāṇḍa. Vol. 7 of The Vālmīki Rāmāyaṇa: Critical edition. Edited
by G.H. Bhatt and Shah. Baroda: University of Baroda, 1975.
Shulman, David. “Toward a Historical Poetics of the Sanskrit Epics.” In The Wisdom of
Poets: Studies in Tamil, Telugu, and Sanskrit. Delhi: Oxford University Press, 2001. First
published in International Folklore Review 11 (1991): 9–17.
Sukthankar, Vishnu S. “Epic Studies VI: The Bhṛgus and the Bhārata: A Text- Historical
Study.” Annals of the Bhandarkar Oriental Research Institute 18, no. 1 (1936): 1–76.
Sutherland, Goldman, Sally. “Gendered Narratives: Gender, Space and Narrative
Structures in Vālmīki’s Bālakāṇḍa.” In The Rāmāyaṇa Revisited, edited by Mandakranta
Bose, 47–85. Oxford: Oxford University Press, 2004.
Van Nooten, Barend A. The Mahābhārata. Twayne’s World Authors Series. New York:
Twayne Publishers, 1971.
Wolpert, Stanley, ed. Encyclopedia of India, 4 vols. New York: Scribners/Thomson Gale,
2006.
Chapter 2 69
Chapter 2
My first inclination was to entitle this presentation, “The Ritual of Battle: Rāma
in the Uttarakāṇḍa of the Vālmīki Rāmāyaṇa.” But somehow, in the back of my
mind, I had a haunting feeling that I had read something with a similar title
some time back. In any case, that title would have been as equally apposite to
the matter at hand as the somehow less evocative one I finally settled upon.
Before attempting to justify my second choice for a title I would like to
reflect, for a moment, on just what we understand the upasarga upa to mean
in the term upākhyāna. This may not be as simple as it would first appear.
According to the lexical work the Gaṇaratnamahodadhi of Vardhamāna,1
upa has no fewer than thirteen distinct meanings not including the standard
sense of “lesser, secondary” which is the one we normally and perhaps unre-
flectingly assign to it in the term upākhyāna. Vardhamāna says:
sāmīpyasāmarthyavyāptyācāryakṛtimṛtidoṣadānakriyāvīpsārambhā-
dhyayanapūjaneṣu (“upa can be taken in the senses of proximity, capacity,
pervasion, a teacher, fraud, death, fault, giving, action, repetition, begin-
ning, studying, and worshiping”).
How, then precisely, are we to understand the sense of upa in the term
upākhyāna? The sense of “subordinate or lesser,” as, for example, in terms like
upapurāṇa is surely an apposite one here as the upākhyānas of the Mahābhā
rata—even the longer ones—are certainly but mere drops in the vast ocean of
the monumental poem. But this sense does not, to my mind, convey much that
is not perfectly obvious. Perhaps it is more helpful if we understand the prefix
as it is used in upākhyāna to refer instead to the thematic proximity in terms of
the instructive quality of episodes (adhyayana) such as the Rāmopākhyāna
when juxtaposed with the events in the career of the Pāṇḍavas. If this
interpretation has merit then, in at least some sections, especially its closing
episodes, it would appear that Uttarakāṇḍa of Vālmīki’s Rāmāyaṇa may well
have been crafted as a kind of repetition of the ritual-political thematic of the
larger epic. I will argue that the author or authors of several of the closing sar-
gas of the Uttarakāṇḍa, in an effort to construct ritual, when associated with
battle, as a necessary legitimating force for the territorial conquest, real, meta-
phorical, spiritual, or fanciful, in fact, borrow from the longer epic. In doing so,
I believe, these sections work, albeit clumsily, against the grain of the first six
kāṇḍas of the poem in an attempt to revalorize Rāma’s famous utopian king-
dom of God on earth to make it more like the imagined universal empire
conquered through the kind of massive, ritually sanctioned violence idealized
in the Mahābhārata, Sanskrit kāvya and royal praśastis. In so doing, however,
they struggle to depict Rāma as an ideal, Mahābhārata-style cakravartin, “uni-
versal emperor,” for whom killing and conquest constitute the supreme
dharma, while still maintaining his unique character, as Pollock has pointed
out, as a ruler for whom the path to heaven is not “conquest in battle” as in the
Mahābhārata but “truth, righteousness, and strenuous effort, compassion for
creatures and kindly words, reverence for brahmans, gods, and guests.”2
So, then, with this as a background and in fond recollection of reading
Vālmīki with Alf in the vānara-haunted woods of the Western Ghats back in
the 1970s, let me turn to the rationale for viewing the Uttarakāṇḍa in part as a
kind of nachdichtung or even, dare I say, an upākhyāna of critical elements of
the central narrative of the Mahābhārata.
2 For this and the rest of Pollock’s illuminating discussion of the character of Rāma, which
combines both the nature of the kṣatriya and that of the brahman, see Sheldon I. Pollock,
trans. Ayodhyākāṇḍa, vol. 2 of The Rāmāyaṇa of Vālmīki: An Epic of Ancient India, ed. Robert
P. Goldman (Princeton, NJ: Princeton University Press, 1986), 64–73.
On the Upatva of Upākhyān as 71
3 For a thorough and learned exposition of the concept see Alf Hiltebeitel’s magisterial study
Rethinking the Mahābhārata: A Reader’s Guide to the Education of the Dharma King (Chicago:
University of Chicago Press, 2001).
4 John L. Brockington, The Sanskrit Epics (Leiden: E.J. Brill, 1998), 391–97.
5 For a discussion of this peculiar episode (VR 7.37–38) see Sally J. Sutherland Goldman,
“Appendectomies: Textual Surgeries in the Construction of ‘The National Epic of India,’” paper
72 Goldman
Compare this with the imagined military and political reach of Yudiṣṭḥira
and his Lunar Dynasty as it is described in the Sabhāparvan of the Mahābhārata
(2.23–29) with its marvelously hyperbolic account of Yudhiṣṭhira’s digvijaya
during which his four heroic brothers lead their victorious armies on fantastic
campaigns of conquest and make vassals of every king in India and far beyond
it’s borders. Indeed, one might argue that it is precisely this conquest of the
four directions on behalf of the dharmarāja, “dharma king,” that accounts for
the specific number of Pāṇḍu’s sons, five; one to rule at the center and four to
conquer the cardinal directions. The four brothers’ military expeditions leave
the dharmarāja Yudhiṣṭhira, or so the poet imagines for us, in command of
an empire encompassing not only the entire Indian subcontinent but extend-
ing all the way from Antioch in the West to China in the East. This is entirely
consonant with the Indic concept of the world-conquering cakravartin and is
indeed perhaps the original textual model for the concept, which in turn may
have been derived from early contact with the Achaemenid Empire in emula-
tion of which this fantasized imperium may well have been envisioned.6
As noted above, the first six kāṇḍas of the Rāmāyaṇa, although they do not
hesitate to indulge in the most astonishing hyperbole when it comes to the
numbers of the vānaras in Sugrīva’s host, the vast size of Kumbhakarṇa and
(on occasion) Hanumān, or the length and virtues of Rāma’s rājya, have, in
contrast, virtually nothing to say about Kosala’s military forces or about the
extent or expansion of its rulers’ territorial control.
On the other hand the authors of the Uttarakāṇḍa seem suddenly to find it
awkward that Rāma’s long-awaited and enduring utopian realm should, in the
end, be confined to the narrow territory between the Sarayū and the Gaṅgā.
And yet, as I hope to demonstrate, although the authors of the Uttarakāṇḍa
wish to construct Rāma as a true cakravartin and dharmarāja, ruling over at
least the entire Indian subcontinent as we see Yudhiṣṭhira doing in the Mahāb
hārata and many poets and kings claiming in the later kāvya and inscriptional
praśasti literature, and so forth, they are also cognizant of and influenced—no
doubt—by Vālmīki’s construction of Rāma as an ideal and uniquely merciful
presented at the conference on New Directions in the Study of the Epics of South and
Southeast Asia, October 26–27, 2012, University of California at Berkeley. See also “Historicity
and Sovereignty” in the Introduction to Robert P. Goldman and Sally J. Sutherland Goldman,
trans. Uttarakāṇḍa, vol. 7 of The Rāmāyaṇa of Vālmīki: An Epic of Ancient India, ed. Robert P.
Goldman (Princeton, NJ: Princeton University Press, 2016).
6 Note how this notion of universal sovereignty comes to be associated with the Sūryavaṃśa as
well in later poetic works such as the Raghuvaṃśa.
On the Upatva of Upākhyān as 73
monarch who conquers only when his righteous anger is provoked and never
simply to acquire tribute or territory.7
In what I would suggest is an attempt to negotiate these two opposing ideals
of Hindu kingship, the Uttarakāṇḍa authors indulge in an awkward dance
around the critical issue of the agonistic nature of the two great śrauta rites
associated with the political power of the king, the rājasūya, “the royal conse-
cration,” and the aśvamedha, “the horse sacrifice,” which ritually sanctify the
subjugation of other kingdoms. These are, of course, the very two rites that
enable Yudhiṣṭhira first to conquer the known world and later, after the resto-
ration of his lost kingship, to extend his sovereignty once again over all of
Bharatavarṣa.
The connection between ritual and battle in the Mahābhārata is very inti-
mate as Alf has shown.8 Yudhiṣṭhira’s first act as king after he has sent forth
his brothers to subjugate and lay waste to all rival kingdoms is to perform a
grand rājasūya, the purpose of which is to ritually confirm his new status as
the cakravartin (Mahābhārata 2.30–36). Subsequently, after his victory over
his enemies in the civil war of the Bhāratas, the despondent king is urged to
purify himself and once again assert his suzerainty over all other kings and
their territories by performing an aśvamedha. In this rite a war party follows
a consecrated stallion in its wanderings for a year and demands surrender
or battle from every king into whose territory the horse wanders.9 Arjuna,
appointed to lead the armed party, duly circumambulates the earth following
the hoof prints of the horse, defeating and subjugating the world’s kings once
again in the name of Yudhiṣṭhira, the newly reestablished universal monarch
(14.68–91).
In Vālmīki’s Uttarakāṇḍa as well, Rāma, like Yudhiṣṭhira, having newly
established himself as king and having demonstrated his adherence to the
brahmanical code of social and ritual order (varṇāśramadharma) and the
7 Cf. Pollock, Ayodhyākāṇḍa, etc., on Rāma. On the hyperbole of imperial control see Jesse R.
Knutson, “The Political Poetic of the Sena Court,” Journal of Asian Studies 69, no. 2 (2010):
371–401 and Jesse R. Knutson, Into the Twilight of Sanskrit Court Poetry: The Sena Salon of
Bengal and Beyond (Berkeley, London, and Los Angeles: University of California Press, 2014).
8 Alf Hiltebeitel, The Ritual of Battle: Krishna in the Mahābhārata (Albany: State University of
New York Press, 1990 [1976]).
9 On the aśvamedha see Stephanie Jamison, Sacrificed Wife/Sacrificer’s Wife: Women, Ritual,
and Hospitality in Ancient India (New York: Oxford University Press, 1996) and Robert P.
Goldman and Sally J. Sutherland, trans., Bālakāṇḍa, vol. 1 of The Rāmāyaṇa of Vālmīki: An Epic
of Ancient India, ed. Robert P. Goldman (Princeton, NJ: Princeton University Press, 1984),
145–53, 298–309.
74 Goldman
exemplary conduct of the perfectly dharmic king through his summary execu-
tion of a śūdra ascetic,10 and his banishment of his beloved and devoted wife,11
decides that he too must ritually validate his accession to the throne through
the performance of a rājasūya. He tells Bharata and Lakṣmaṇa:
punitive campaign against the demonic king who had abducted his beloved
wife. Even after his total victory in this campaign, he shows no interest what-
ever in occupying or annexing the conquered territory. Instead he simply
consecrates his slain enemy’s brother on the Laṅkan throne, and, taking noth-
ing from that fabulously wealthy kingdom but the loan of a flying palace, he
returns home to rule his own kingdom in peace. Thus there is neither conquest
nor tribute but merely, to use the contemporary idiom, regime change.
But let us return to the question of Rāma’s proposed rājasūya. The very
nature of this rite, with its inevitable slaughter of rival kings and their armies,
presents, when situated in the context of Rāma’s ushering in his millenarian
age of universal sovereignty, peace, and harmony, an irreconcilable dilemma.
Bharata, who serves here as the spokesman for the Uttarakāṇḍa authors, imme
diately understands the utter incompatibility of the two concepts. Accordingly
he sets himself immediately to the project of dissuading Rāma from his poorly
thought out plan. He says to Rāma:
Therefore, your majesty, how can you undertake such a sacrifice, which
would witness the destruction of the royal lineages on earth? For, in the
performance of that, your majesty, the annihilation—like the universal
destruction—of all men who have attained manly valor, will come to
pass on earth. Tiger among men, you who are of unequalled valor through
your virtues, should not destroy the earth for it is already under your
sway. (VR 7.74.12–14)
When truly valorous Rāma had heard that speech of Bharata, which
seemed to consist of the nectar of immortality, he experienced unequalled
delight. And he said these splendid words to that increaser of Kaikeyī’s
joy: “I am indeed pleased and gratified by your words this day. These
words in conformity with righteousness, which you have boldly uttered,
tiger among men, will be the salvation of the earth. Because of your excel-
lent speech, knower of righteousness, I will surely desist from that
intention of mine, which was to perform that foremost of rites the
rājasūya.” (VR 7.74.15–18)
Thus, in effect, Rāma, who has proposed the supposedly required rite for
the installation of a new king, a rite that is most exemplary of the true
76 Goldman
Given all of this, one would think that the authors of the Uttarakāṇḍa would
be content to portray King Rāma as a pacific and righteous monarch, more of a
cakravartin in the spiritual and moral sense of the term than in its political
sense.14 Besides, Rāma, in Bharata’s rather vague terms, is already a ruler who
holds the whole earth under his sway—a sway, remember, for which the text
has given neither history nor explanation.
Nonetheless the Uttarakāṇḍa, in its closing half, does provide an interesting
revisionist history of conquest and territorial expansion on the part of Rāma
and his ruling Ikṣvāku dynasty of Kosala. Although, as I have indicated, Rāma
is never, after the defeat of Rāvaṇa, shown to engage personally in any form of
warfare or to employ either the rājasūya or the aśvamedha rituals to legitimate
wars of conquest or territorial expansion, the later sargas of the Uttarakāṇḍa do
present some rather interesting cases of military campaigns led by his kinsmen
which serve to vastly expand the reach of an imagined Ikṣvākuid imperium.
As noted above, the first example of such a move is narrated, with several
digressions, in sargas 53–63 of the kāṇḍa. Here, responding to the petition of
the sages who live along the banks of the Yamunā and of their spokesman, the
Bhārgava ṛṣi Cyavana, concerning the depredations of a terrible and immensely
powerful demon named Lavaṇa, Rāma deputes his brother Śatrughna, who
hitherto in the long epic has had a quite minor role at best, to destroy the mon-
ster. After a heroic battle, he does so. The episode is, in a way, a minor reprise of
Rāma’s own great martial feat, the destruction of the fearsome Rāvaṇa who,
with his rākṣasa hosts, had similarly preyed on brahman sages. The critical dif-
ference here, however, is that Rāma now instructs his brother to found a city
and a kingdom in the region of the battle and consecrates him as king of the
new political formation of Madhupurī or Madhurā. In this way the Doab heart-
land of the Śūrasenas at Mathurā is represented as becoming a satrapy or a
client state of Kosala. After twelve years of ruling his new kingdom, Śatrughna
returns to Ayodhyā to see his beloved brother once more. But Rāma tells him,
rather sharply, that this abandonment of his kingdom is inappropriate behav-
ior for a ruler and that he must return—after only five days—to Mathurā
where he is to remain and continue to rule.
Then, many years later, according to the Uttarakāṇḍa, just as the whole book
and indeed the entire epic are within a few brief sargas of their end, the authors
suddenly seem to feel the need to quickly cram in a few more major territorial
acquisitions for the Solar Dynasty. Three short chapters, totaling a mere fifty-
14 Pollock, Ayodhyākāṇḍa, 64–73. Compare the ambivalence of the term in the accounts of
the birth of the Buddha.
78 Goldman
eight ślokas in all, chronicle not one but two unheralded and rather astonishing
wars of conquest and annexation.
In sargas 90 and 91 we learn that Bharata’s maternal uncle Aśvapati Yudhājit
of the Kekeyas sends a messenger to Ayodhyā to request him to invade, subdue,
and occupy the rich and beautiful country of the Gandharvas, which lies along
the banks of the Indus. Once he has done so, Bharata is to found two grand
capital cities in that region.
Delighted at the prospect of the military campaign against thirty million
powerful gandharvas, Rāma immediately resolves to send out an army led by
Bharata and the latter’s two sons, Takṣa and Puṣkala, to annihilate the natives
and occupy their land. Bharata, Takṣa, Puṣkala, and Yudhājit duly proceed to
the country of the gandharvas where, after a tremendous but inconclusive
week-long battle, described in but a single verse, Bharata manages finally to
annihilate all thirty million of the gandharva warriors in single instant with
the terrifying weapon of mass destruction of Death himself, the dreaded
saṃvartāstra. Following this stunning act of ethnic cleansing, Takṣa and
Puṣkara occupy the land, now known by its familiar name Gāndhāra (evidently
in memory of the slaughtered millions), founding, respectively, two splendid
eponymous capitals, Takṣaśīlā and Puṣkarāvatī. These cities are well-known
from the historical record in both Indic and Greek sources.
Clearly this sketchy and far-fetched account, which attempts to provide the
etymologies of the well-known geographical names Gāndhāra, Takṣaśīlā, and
Puṣkarāvatī, has been designed largely to further enhance the imperial preten-
sions of the Sūryavaṃśa by extending the reach of the Kosalan monarchy to far
flung northwestern territories in what are today Pakistan and Afghanistan.
When Rāma learns of the success of this campaign of conquest, he is
delighted but, perhaps a bit disturbed by its massive and unprecedented death
toll, he decides to expand his dynasty’s territorial reach still further but more
nonviolently. In sarga 92 he tells Lakṣmaṇa to seek out yet another large and
well-endowed country over which he, Rāma, will set his devoted brother’s two
sons, Aṅgada and Candraketu, as rulers. But this time in a kind of Aśoka
moment, he insists that this be done in a kinder, gentler fashion. He instructs
Lakṣmaṇa:
But please, gentle brother, seek out a country in which there will be no
oppression of kings and no destruction of ashrams, so that we may give
no offense. (VR 7.92.4)
15 Cf. the medieval and later efforts to render Vālmīki’s characterization of Rāma more like
that of Kṛṣṇa as represented in such works as the Harivaṃśa, the Bhāgavata Purāṇa, the
Gītagovinda, etc., in such works as the Bhuśuṇḍi Rāmāyaṇa, the modern graphic and
80 Goldman
Gandaris16 and which was under the control of a ruler known to the Greeks as
Astes (Sanskrit Hastin), seems to have been established as a military strong-
hold by Alexander’s generals Hephaistion and Perdiccas in 327 bce.17 But it
also was known to have become a significant capital, largely reconstructed in
the Greek manner, only during the reign of Menander in the middle of the
second century bce.18 Takṣaśīla (Greek Taxila) too had a somewhat similar his-
tory, having also been rebuilt by Menander and perhaps served as his capital.19
It would of course be purely speculative at this point to draw any firm
chronological conclusions from the information we have. Nonetheless, the
foregrounding of these two cities which became so politically prominent
under Greek rule in the middle of the second century bce and the Uttarakāṇḍa’s
determined effort to represent them as having been founded by scions of the
royal family of Ayodhyā, Rāma’s nephews, may suggest that it was the increased
importance of these towns in that period that inspired the work’s author or
authors to claim these towns and their surrounding provinces as part of a pan
Indian empire for the Kosalan monarchy.
References
16 Rafi U. Samad, The Greeks in Ancient Pakistan (Karachi: Indus Publications, 2002), 23. On
the names of the city and the region see W.W. Tarn, The Greeks in Bactria and India (Cam-
bridge: Cambridge University Press, 1951), 237–38.
17 Samad, The Greeks in Ancient Pakistan, 9.
18 Ibid., 99.
19 Ibid., 99–101 and Tarn, The Greeks in Bactria and India, 135.
82 Goldman
in the Study of the Epics of South and Southeast Asia, October 26–27, 2012, University
of California at Berkeley.
Hiltebeitel, Alf. Rethinking the Mahābhārata: A Reader’s Guide to the Education of the
Dharma King. Chicago: University of Chicago Press, 2001.
———. The Ritual of Battle: Krishna in the Mahābhārata. Albany: State University of
New York Press, 1990 [1976].
Jamison, Stephanie. Sacrificed Wife/Sacrificer’s Wife: Women, Ritual and Hospitality in
Ancient India. New York: Oxford University Press, 1996.
Keislar, Allan. “Searching for the Bhuśuṇḍi-Rāmāyaṇa; One Text or Many? The Ādi-
rāmāyaṇa, the Bhuśuṇḍi-rāmāyaṇa, and the Rāmāyaṇa-mahā-mālā.” PhD diss.,
University of California, Berkeley, 1998.
Knutson, Jesse R. Into the Twilight of Sanskrit Court Poetry: The Sena Salon of Bengal and
Beyond. Berkeley, London, and Los Angeles: University of California Press, 2014.
———. “The Political Poetic of the Sena Court.” Journal of Asian Studies 69, no. 2 (2010):
371–401.
Pollock, Sheldon I., trans. Araṇyakāṇḍa. Vol. 3 of The Rāmāyaṇa of Vālmīki: An Epic of
Ancient India. Edited by Robert P. Goldman. Princeton, NJ: Princeton University
Press, 1991.
———. Ayodhyākāṇḍa. Vol. 2 of The Rāmāyaṇa of Vālmīki: An Epic of Ancient India.
Edited by Robert P. Goldman. Princeton, NJ: Princeton University Press, 1986.
Samad, Rafi U. The Greeks in Ancient Pakistan. Karachi: Indus Publications, 2002.
Tarn, W.W. The Greeks in Bactria and India. Cambridge: Cambridge University Press,
1951.
Chapter 3 83
Chapter 3
⸪
Introduction
In his article “Not Without Subtales: Telling Laws and Truths in the Sanskrit
Epics,”1 Hiltebeitel counts sixty-seven upākhyānas in the Mahābhārata.
Eleven of these occur in the first parvan or major book alone. Among these, the
Śakuntalā narrative stands out in a number of ways: it is the first upākhyāna in
the text; one of ten addressed by Vaiśaṁpāyana to Janamejaya (1–4, 11, 32,
39–40, 66–67 in Hiltebeitel’s list); one of only four “about early kings of the
main dynasty”2; as well as being the upākhyāna that goes over from an
account of the partial incarnations of the gods and other beings to the first of
the accounts of the descendants of the Lunar Dynasty. Most significant, this
upākhyāna introduces “the mother of Bharata, one of the line’s eponyms.”3
Hiltebeitel also draws attention to the fact that Śakuntalā’s period of gestation
of three years4 corresponds exactly to the time Kṛṣṇa Dvaipāyana Vyāsa, the
1 Alf Hiltebeitel, “Not Without Subtales: Telling Laws and Truths in the Sanskrit Epics,” Journal
of Indian Philosophy 33, no. 4 (2005): 455–511.
2 Ibid., 475.
3 Ibid., 479.
4 pratijñāya tu duḥṣante pratiyāte śakuntalā |
garbhaṁ suṣāva vāmoruḥ kumāram amitaujasam ||
triṣu varṣeṣu pūrṇeṣu diptānalasamadyutim |
rūpaudāryaguṇopetaṁ dauḥṣantiṁ janamejaya || (Mahābhārata 1.68.1–2)
traditional author of the epic, is said to have taken for its composition.5
Śakuntalā’s son not only bestows his name upon the Kuru line, his descen-
dants, but also gives his name to the text, the Mahābhārata, and to a land called
Bhārata.6
The placement of the Śakuntalā narrative has other noteworthy aspects:
it precedes the story of Yayāti and the discussion of Yayāti’s line through his
youngest son Pūru, the Paurava dynasty, even though its main protagonist,
Duḥṣanta Paurava, belongs to this dynasty (see Mahābhārata 1.62.3; cf. also
1.67.15c, 33c, 68.59a, 69.32c, and 69.34a).7 In this inversion of the chronological
order, we see one of the epic’s dominant architectonic motifs: the epic begins
at the end of time and works its way backward in time to the beginning.8
In this chapter, I focus on the Śakuntalā-Upākhyāna (Mahābhārata 1.62–
69), perhaps the most popular of the epic’s subtales among early European
5 See his “The First Reading of ‘Śakuntalā’: A Window on the Original and the Second Read-
ing by Kālidāsa,” in Revisiting Kālidāsa’s Abhijñāna Śakuntalam: Land, Love, Languages:
Forms of Exchange in Ancient India, ed. Deepika Tandon and Saswati Sengupta (Delhi:
Orient BlackSwan, 2011), 17–37.
6 Vishwa Adluri points out (personal communication) that this suggests that Bharata
should be understood not just as a character in the epic, one of the remote ancestors in
the Kuru genealogy, but as a cipher for the coming into being of the text. Like the epic’s
composer, the seer Vyāsa, Bharata embodies the twin functions of creator (of the text)
and genitor (of the Kuru race). Also like the seer, he is a Brahmā-like figure. He is described
as a cakravartin, a “Turner of the Wheel” (Mahābhārata 1.67.29; cf. also 1.69.45–46 and
1.68.4), and as a Dakṣa-like figure (yājayām āsa taṁ kaṇvo dakṣavad bhūridakṣiṇam;
1.69.48). On Vyāsa’s relation to Brahmā, see Bruce K. Sullivan, Seer of the Fifth Veda: Kṛṣṇa
Dvaipāyana Vyāsa in the Mahābhārata (New Delhi: Motilal Banarsidass, 1999).
7 At least in the Mahābhārata’s Critical Edition. The southern manuscripts transpose these
two sections, see V.S. Sukthankar’s comments in the “Prolegomena” and in still greater
detail in his “Editorial Note (3)” to the fourth fascicule of the Ādiparvan (V.S. Sukthankar,
“Prolegomena,” in The Mahābhārata for the First Time Critically Edited, vol. 1. The Ādipar
van [Pune: Bhandarkar Oriental Research Institute, 1933], i–cx and V.S. Sukthankar, “Edi-
torial Note (3),” in The Mahābhārata for the First Time Critically Edited, vol. 1. The
Ādiparvan: Fascicule 4 [Pune: Bhandarkar Oriental Research Institute, 1930], i–iv). Biar
deau makes “disorganization” and “composition spread out in time” responsible for the
change in order (Madeleine Biardeau, “Śakuntalā dans l’Épopee,” Indologica Taurinensia
7 [1979]: 115) but I shall instead argue that an interest in restoring the text was behind the
changes in the southern recension.
8 Its first narration is to King Janamejaya, two generations after its main heroes have passed
away. Likewise, in the Ādiparvan, when the bard begins narrating, the first detail he tells
us pertaining to the Kurus is Dhṛtarāṣṭra’s post-war lament, in which the old king looks
back at the events leading up to the destruction of his hundred sons.
The Epic’s Singularization to Come 85
9 For the history of the western reception of the Śakuntalā, see Dorothy Matilda Figueira,
Translating the Orient: The Reception of Śākuntala in Nineteenth-Century Europe (Albany:
State University of New York Press, 1991). Also useful are: Montgomery Schuyler, Jr., “The
Editions and Translations of Çakuntalā,” Journal of the American Oriental Society 22 (1901):
237–48; Peter H. Salus, “Šakuntalā in Europe: The First Thirty Years,” Journal of the Ameri-
can Oriental Society 84, no. 4 (1964): 417; and Robert A. Hueckstedt, “The Plays of Kalidasa
and their Major Twentieth-Century English Translations,” Journal of South Asian Litera-
ture 22, no. 1 (1987): 215–29.
10 Note that when I say “southern recension,” I mean precisely this (i.e., a a specific state of
the text whose existence can be inferred on the basis of the manuscript evidence, specifi-
cally the common features between the T, G, and M manuscripts, and which must have
come into being at least at the time these three traditions separated if not before). I do not
mean a “southern redaction” (which, allegedly, would have been responsible for the emer-
gence of the southern recension) and even less “the work of southern redactors” (who
would have been the people behind the “southern redaction”). Since all we have are man-
uscripts, we must restrict ourselves to speaking only about manuscripts.
11 The Critical Edition refers to this upaparvan as the Ādivaṁśāvataraṇaparvan. In the
colophons of the manuscripts collated for the Critical Edition, the book is referred to
as the Aṁśāvataraṇaparvan (with the usual variation). The Parvasaṁgrahaparvan
list (1.2.72–94) also refers to it as Aṁśāvataraṇaparvan. The editor’s decision is likely
based on the sole reference to the book as the Ādivaṁśāvataraṇaparvan at 1.2.34 (cf.
parvānukramaṇī pūrvaṁ dvitīyaṁ parvasaṁgrahaḥ | pauṣyaṁ paulomam āstīkam ādi
vaṁśāvatāraṇam ||).
86 Bagchee
1.90—and the main narrative of the Mahābhārata resumes from 1.91 onward
with the story of the birth of Bhīṣma from Vasu. The Critical Edition follows the
northern sequence of narratives; the southern manuscripts are unanimous in
reversing the sequence of the Śakuntalā and Yayāti episodes as well as making
several other smaller changes. As the textual history of this section of the
Ādiparvan is quite complex, I first give the sequence of episodes in the two
recensions. The northern sequence is essentially that of the Critical Edition;
for the southern sequence, I have relied on the critical apparatus of the Critical
Edition to reconstruct the putative text of the southern recension, though of
course there are sometimes divergences from this text even within the manu-
scripts of the southern tradition.12 I have listed all such insertions as are made
in all of the manuscripts of the southern recension; those found only in a selec-
tion of them have been placed within braces ({ … }). Thus, the sequence below
indicates the essential outline of the southern tradition as represented by TG
(M differs, but is not reconstructed here).13 Continuations of the narrative are
noted by means of ellipses; the hiatus noted by Sukthankar between 1.69.51
and 1.70.1 is indicated by means of a short line.
12 These differences have been noted in the footnotes. I have incorporated the southern
variants into my text, displacing the reading of the northern recension, where this reading
was supported by all (or, at least, a clear majority) of the manuscripts (all such cases have
been explicitly highlighted in the footnotes). However, note that the text below should
not be taken as the text of the southern recension, which can only be reconstructed from
the critical apparatus with a great deal more of difficulty (if at all, since in many cases it
will be impossible to find a unanimous reading among the southern manuscripts) and
would also require a finer evaluation of the evidence (when TG have one reading against
M, do we reprint the reading of TG or M? The latter problem has been obviated in this
reconstruction by choosing to give only the text of TG). On the problems of identifying a
true critical text of the southern recension, see the conclusion.
13 Note that the changes listed here cannot be considered a continuous text of the southern
recension when read with the intervening passages from the Critical Edition, for those
passages represent the editor’s best guess of the text of the common archetype of the two
recensions. For a complete text of these sections as they are found in the southern recen-
sion, one would have to also reconstruct the text of the southern recension for the inter-
vening passages, a task that can only be achieved by carefully looking at the variants of
the southern manuscripts in the critical apparatus, as I have done here. My aim, however,
was only to reconstruct the sequence and not the content of these sections.
The Epic’s Singularization to Come 87
TG (southern sequence)14
61.99 iti devāsurāṇāṁ te gandharvāpsarasāṁ tathā | 15
–102 aṁśāvataraṇaṁ rājan rākṣasānāṁ ca kīrtitam ||16
ye pṛthivyāṁ samudbhūtā rājāno yuddhadurmadāḥ | 17
mahātmāno yadūnāṁ ca ye jātā vipule kule ||
569* ete tu mukhyāḥ kathitā mayā te rājasattama |
dhanyaṁ yaśasyaṁ āyuṣyaṁ putrīyam vijayāvaham | 18
idam aṁśāvataraṇaṁ śrotavyam anasūyayā ||19
aṁśāvataraṇaṁ śrutvā devagandharvarakṣasām |
prabhavāpyayavit prājño na kṛcchreṣv avasīdati || (Aṁśāvataraṇap. ends)
62.1 janamejaya |
–2 tvattaḥ śrutam idaṁ brahman devadānavarakṣasām |
aṁśāvataraṇaṁ samyag gandharvāpsarasāṁ tathā ||20
imaṁ tu bhūya icchāmi kurūṇāṁ vaṁśam āditaḥ |
kathyamānaṁ tvayā vipra viprarṣigaṇasaṁnidhau ||21
62.322/
70.123 vaiśaṁpāyana |
14 The sequence of M differs: M omits 62.1–2 and inserts, immediately after 61.102, the S ver-
sion of 89.1–16. This “short genealogical adhyāya” (Sukthankar’s words) occurs twice in M:
once immediately before the Yayāti episode, which is inserted after 89.16, and once again
after it, both times followed by a colophon. The second time it (i.e., 89.1–16) is followed in
all S manuscripts by the formal introductory stanza 571*, which precedes the Śakuntalā
episode.
15 99a: T1 G1–3 asurāṇāṁ surā˚.
16 99c: T puṇyaṁ (for rājan).
17 100a: T G1.2 M6–3: samutpannā; 100c: T2 G6: mahātmanāṁ.
18 101b: note the transposition of āyuṣyaṁ putrīyam from the āyuṣyaṁ putrīyam of the con-
stituted text—this is a feature of all S manuscripts.
19 101d: note the reading sūyayā (for the sūyatā of the constituted text)—this is a feature of
all TG manuscripts.
20 G5 M om. 1–2; 1a: G6 sarvaṁ; 1b: G3 devagandharva˚; after 1, G1 ins. (for the first time, the
S version of) 1.89.1–19; the passage is repeated, in its proper place, after the Yayāti episode.
21 G5 M om. 2; 2a: T1 G1–4 idaṁ; 2c: T2 G4.6 ˚yā brahman; 2d: G1–3 devarṣi˚.
22 After vaiśaṁpāyana uvāca, TG (for G1 see n. 20) insert a formal introductory stanza, fol-
lowing it up with the Yayāti-Up. and the Uttarayāyāta. The insertions of these two sections
have been ignored for the purposes of this reconstruction.
23 Sukthankar’s comments are a little confusing here, but the vaiśaṁpāyana uvāca of 62.3
evidently doubles as the vaiśaṁpāyana uvāca of 70.1 (i.e., it is not repeated). According to
the critical apparatus for 70.1, the uvāca is omitted by all S manuscripts.
90 Bagchee
24 1c: G1 ˚sya kule; 1d: note the reading ājamī˚ (for the ajamī of the constituted text)—this is
a feature of all TG manuscripts (and M5).
25 2b: G4 kauravā˚, T2 G4.5 M6–8 nityaśaḥ; 2c: T1 G2 M (except M5) bharatā˚; 2d: G3 om.
2e–3f.
26 26a: G1.3 putrāḥ (for jātāḥ), T G2.4–6 M yadorjātā yādavāstu (M3.5 ˚śca).
27 27d: T1 kāraya naḥ prabho, T2 G kārayitā prabho, M3.5 kaliyugebhavat, M6–8 prabho (for
vaśī).
28 S om. uvāca; 1b: G1.3 purum (for putram).
29 G2 om. vaiśaṁ˚u˚; S om. uvāca; 26a: after evaṁ, T1 G5 M6–8 ins. sa, T1 sumahā˚, M3.5
yayātiḥ (for hyātī˚); 26b: G (except G2.3) mātṛvākyāt (for ’mitra˚); 26d: G4–6 naptrupetaḥ
(for vyāpya pṛthvīṁ), G1.3 vṛddhiṁ (for pṛthvīṁ).
30 After the Uttarayāyāta, S inserts the first sixteen lines of chapter 89 (Mahābhārata 1.89.1–
16) with some substitutions and insertions as a separate chapter (with its own colophon).
The remainder of this chapter (1.89.17–55) is then inserted between 1.69.48 and 1.69.49 of
the Śakuntalā-Upākhyāna, which is itself inserted following S’s short chapter of sixteen
lines (i.e., after 1.89.16). Line 1.89.16 is then repeated after 1.69.48 of the Śakuntalā-Up. to
restore the context with the preceding half of this chapter (i.e., 1.89), which had been lost
due to the insertion of the Śakuntalā-Upākhyāna.
31 Hereafter, the words “of the constituted text” are understood and will not be repeated.
The Epic’s Singularization to Come 91
32 Line 1: T2 G3.4.6 puruṁ, G1.2 punar (for pūrur), G1–3 pitevāsya yathā nṛpaḥ.
33 Note the change from the pravīreśvararaudrāśvās trayaḥ putrā mahārathāḥ of the consti-
tuted text. M6–8 (first time) ˚ryaḥ (for pravīraḥ). T1 ˚gdhyauś; T2 G4(sup. lin.).6 ˚gdyauś;
G4.5 ˚dhrau; M3 ˚jyā (for śatarucyau).
34 5d: G3 ˚vān (for bhāk of S; kṛt of constituted text).
35 6a: M (except M7) namasyur; G1.2 ˚manu; G3 nabha˚; 6b: note the change from the chūraḥ
śyenīsutaḥ prabhuḥ of the constituted text; G3 ˚vo; M8 ˚byāḥ; M8 tataḥ (for prabhuḥ of
constituted text).
36 Note the change from the pṛthivyāś caturantāyā goptā rājīvalocanaḥ of the constituted
text.
37 7ab: G2 suhṛt (for subhrūḥ); T2 subhruścābhayado vā˚; T1 G1.2.4–6 M6.8 sauvīr˚; G3
saṁbhramaḥ saṁvano vāgmī sauvīraḥ (sic) tana˚; M3.5 subhruścābhayado rājā vāgmī
sauvīrajā (M5 ˚kā) strayaḥ.
38 7c: Note the change from the manasyor of the constituted text (supported by T1 G1.2.4–6
M6–8 [G4 sup. lin. ˚nyor]; G3 nabha˚; M3.5 namanyor).
39 Line 1: T1 G6 M3.5 gargaramyau; line 2: T1 G1–3 M śūrānubhayato rājā (G3 ˚taḥ subhrūḥ);
line 4: G1.2 M ˚ṣmāṁśca, G3 vasumāṁśca, T M3 śūraṁ ca dṛḍhadhanvānaṁ vapuṣmaṁtaṁ
nṛpottamaṁ; line 5: G4–6 M6–8 rathadaścaṁ (for rathāśvaṁ ca); line 6: G3 gāndhāryāṁ,
G6 gandhavān.
40 G2.4–6 hyantinārād, T1 G1.3 sarasvatī (for yaśa˚).
92 Bagchee
47 After 26, T2 G (except G6) ins.; G3 api (for tu) and dharmarājo yathaiva ca.
48 After 27ab, T1 G2.4.5 ins.
49 T2 G2.4.5 vīram (for śūram).
50 G1.2.4 ins. after 32cd: T1, after 33ab.
51 G3 parityajya, G6 parājitya, M jitā bhītāḥ (for parā˚), M prayātā (for ˚pannā).
52 After 46ab, S (except T1 G6) ins.
53 Line 1: T G2.4.5 abhi(T1 ˚ti)rājā ca na˚, G3.6 adhi(G3 ˚ti)rājātmajānāhuḥ; Line 2: T1 satya˚.
54 T2 G2 bhīmaseno mahe˚.
55 55a: M6–8 om. ca, T2 G (except G3.6) evaṁvidhā (T2 ˚rūpā) mahābhāgā; 55b: T1 M3.5
mahābalāḥ (T1 sup. lin. ˚rathāḥ), T2 G2.4.5 devarūpāh prahāriṇaḥ; 55c: note the reading
anvavāye mano rājan (for the jātā manor anvavāye of the constituted text)—this is a fea-
ture of all S manuscripts (T2 G1.4.5 M6–8: mahārāja [for mano rājan?]); 55d: G2 eṣa, G3
evaṁ (for aila˚).
94 Bagchee
56 Line 1: G3 M6–8 gangādvāraṁ, M6–8 samāsādya; after line 2, T1 G1.2 M3.5 ins. an addi-
tional line (895*): kṛtvā paitāmahe loke vāsaṁ cakre mahārathaḥ (v. l. for this line are
ignored here).
57 49c: M vai (for the second ca); 49d: note the reading te bhavan (for the viśrutāḥ of the
constituted text)—this is a feature of all TG manuscripts (M: te smṛtāḥ).
58 50a: note the reading ˚syānvaye jātā (for the ˚syānvavāye of the constituted text); 50b: note
the reading mahārathāḥ (for the mahaujasaḥ of the constituted text); 50c: note the read-
ing bahavo (for the babhūvur of the constituted text)—all three are features of all S man-
uscripts; 50d: T2 G3.6 M3 babhūvaḥ (for bahavo), T2 G (except G1) kṣatrasa˚ (for rājasa˚).
59 51a: note the reading teṣāṁ (for the yeṣāṁ of the constituted text)—this is a feature of all
S manuscripts; 51b: T2 G (except G1) saṁtyuta (for sarva˚); 51c: note the reading
yathāmukhyān (for the yathāmukhyaṁ of the constituted text)—this is a feature of all S
manuscripts (except G1: ˚nyāyaṁ).
60 S reads 1.90 after 1.56; 1.91 follows directly on the end of the Śakuntalā-Upākhyāna.
61 S om. uvāca (M5 om. the ref.); 1a: note the reading ikṣvākūṇāmanyatamo (for the
ikṣvākuvaṁśaprabhavo of the constituted text)—this is a feature of all S manuscripts
The Epic’s Singularization to Come 95
I have ignored for the purposes of this demonstration, the additions to the
Śakuntalā and Yayāti narratives in the southern recension, which are exten-
sive.62 My main purpose, rather, is to understand the problem that arises
concerning the relation of the two episodes to each other and to illustrate the
different ways in which the two recensions respond to this problem. The north-
ern recension places the Śakuntalā-Upākhyāna first in its sequence; it, in a
sense, pulls the eponymous Bharata out of his expected place in the Kuru
genealogy and foregrounds him. The southern recension, in contrast, clearly
regards this sequence as a problem. It makes obvious efforts to restore what it
perceives to be the correct order: it moves the entire Yayāti episode (i.e., both
the Yayāti-Upākhyāna and the Uttarayāyāta section) to before the Śakuntalā-
Upākhyāna; it splits the Śakuntalā-Upākhyāna itself into two, inserting the
narrative of the descendants of Bharata (Mahābhārata 1.89.17 onward) after
1.69.48, where it indeed seems to be a better fit, but keeping the first half of this
chapter (1.89.1–16) (with some changes) at the end of the Uttarayāyāta, where
this narrative of the descendants of Pūru appears a logical continuation of the
stories of Yayāti and Pūru.
(except G2: ekṣvāk˚); 1c: note the reading mahābhiṣaj (for the mahābhiṣa of the consti-
tuted text)—S reads this (uniformly) here and below; 1d: T1 anasūyakaḥ.
62 Between 1.62 and 1.69, the southern manuscripts make five longer insertions (App. 1, no.
45 after 1.67.23; no. 46 after 1.68.9; no. 47 after 1.68.11; no. 48 after 1.68.13; and no. 51 after
1.68.69) and eighty-one shorter ones (571*, 577*, 579*, 581*, 583*, 585*, 586*, 587*, 588*,
589*, 592*, 594*, 596*, 598*, 599*, 602* [G3 om. lines 3–6, M5 om. 1–3], 604*, 605* [G3 om.
lines 3–5], 607*, 608*, 609*, 610*, 611*, 612*, 613*, 614*, 615*, 616*, 617*, 618*, 619*, 620*,
621*, 623*, 624*, 627*, 628*, 629*, 630*, 631*, 633*, 634*, 635*, 637*, 638*, 639*, 641*, 642*,
643* [G6 om. lines 6–7], 644*, 645*, 646*, 647*, 649*, 651*, 652*, 654*, 657*, 658*, 659*,
660*, 662* [G2.4.5 M7 om. line 2], 663*, 666*, 668*, 669*, 670*, 671*, 672*, 673* [G3 om.
lines 2–3], 674*, 675*, 676*, 677*, 679*, 680*, 681*, 682* [T1 om. lines 1–2], 683*, 684*,
686*). For a discussion of these passages, see Hiltebeitel, “The First Reading of ‘Śakuntalā’,”
though he is dismissive of the southern innovations. Hiltebeitel also arrives at a slightly
higher count (eighty-four instead of eighty-one), perhaps because he includes in his list
some insertions only found in a few southern manuscripts (these are: 570* [TG (except
G2)], 584* [G (except G3.6)], 590* [T1 G1.2.5 M3.5], 595* [except M3.5], 597* [T2 G (except
G3.6) M (except M5)], 603* [except G6], 606* [M3.5], 622* [M3.5], 625* [except T2], 626*
[T2 G (except G3)], 636* [G1.2.4.5], 648* [except G6], 650* [G3], 653* [T G1.3–6 M6–8],
656* [G2], 661* [except G3], 664* [except G3], 665* [except M6–8]).
96 Bagchee
In the preceding section, we saw that the southern recension makes extensive
changes to the sequence of the Śakuntalā and Yayāti episodes. Even disregard-
ing additions to these episodes, the changes present the Śakuntalā-Upākhyāna
in a completely new light: it is now no longer the first upākhyāna in the Mahā
bhārata; it has been restored to its proper place in the chronological sequence;
and it is just one of the narratives integrated within a continuous genealogy
that now runs from 1.62 to just before 1.91. In fact, if we look at the preceding
sections (the accounts of the descent of mortals from divine and demonic pro-
totypes) as genealogies of a kind, the southern text now creates a continuous
genealogical narrative that runs from the end of 1.56 (where it is immediately
followed by the prose genealogy of 1.91, now displaced to just after the eulogy
of the Bhāratas) through 1.57 (birth of Vyāsa from Vasu, birth of the elders
and the heroes of the Mahābhārata),63 1.58 (birth of the demons upon earth,
the gods decide to descend with a portion of themselves to uplift the earth),
1.59 and 1.60 (cosmology proceeding downward from Brahmā to genealogy),
and 1.61 (the partial incarnations), to the beginning of the Yayāti-Upākhyāna
(1.70).
63 This narrative, the first of all the genealogical narratives Vaiśaṁpāyana narrates in the
epic, is significant. As two recent studies by Adluri have shown (Vishwa Adluri, “The Vasu
Narratives of the Mahabharata: Some Lexical and Textual Issues,” in Proceedings of the
Sixth Dubrovnik International Conference on the Sanskrit Epics and Puranas [Zagreb: Croa-
tian Academy of Arts and Sciences, forthcoming] and Vishwa Adluri, “ The Divine Andro-
gyne: Crossing Gender and Breaking Hegemonies in the Ambā-Upākhyāna of the
Mahābhārata” in the present volume) the story of King Vasu contains important clues to
the interpretation of the epic: the king’s name relates semantically to the Vasus, a group
of gods responsible for keeping pravṛtti in motion. Further, as the “indweller” or the
“indwelling one,” Vasu perfectly signifies the central theme of an epic that is concerned
with the problem of what it means to be or to exist in time. Finally, Vasu’s turn away from
nivṛtti and to pravṛtti right at the beginning of the Mahābhārata sets up one of the domi-
nant themes of the epic. Vasu’s fall into Becoming, explored paradigmatically in terms of
a return to the sacrificial order dominated by Indra, is the appropriate beginning for an
epic that claims to explicate not only Being but also Becoming. Like Bharata, Vasu’s narra-
tive is pulled out of its expected place and foregrounded in the Mahābhārata. His narra-
tive is the first of the narratives to be related as part of Vaiśaṁpāyana’s narration of the
Kuru conflict (cf. Mahābhārata 1.54.19, 22, and 23–24.) and thus stands in some way at the
head of the entire Mahābhārata.
The Epic’s Singularization to Come 97
More remarkable still is the fact that the transitions between the various
chapters (e.g., Janamejaya’s request to hear how the dynasty of the Kurus
came into being [imaṁ tu bhūya icchāmi kurūṇāṁ vaṁśam āditaḥ; Mahā
bhārata 1.62.2ab] or Vaiśaṁpāyana’s statement that he will recount the story
of Duḥṣanta, a hero of the Paurava line [pauravāṇāṁ vaṁśakaro duḥṣanto
nāma vīryavān; Mahābhārata 1.62.3ab]) are now much more appropriate in
context—and this is so even before the scribes compose additional verses to
bridge the gap between chapters! It is almost as though, realizing that the origi-
nal sequence of the narratives, implied or actually extant, has been disrupted,
the scribes are now working to repair the text. In the process, they restore not
only what is a better order but also specifically heal the breaches in the text. For
instance, if we look at just these transitions in context, we see how the south-
ern recension brings to light an order that is already implicit in the text. As
they exist in the northern recension, the transitions from Janamejaya’s ques-
tions to Vaiśaṁpāyana’s responses are quite awkward:
Śakuntalā-Up. (1.62–69)
Janamejaya: I have heard fully from you, O brahmin, how the Gods,
Dānavas, and Rākṣasas, and also the Gandharvas and Apsarās, descended
to earth with a portion of themselves. Now I wish you to tell me from the
beginning in the presence of these brahmins and seers, O brahmin, how
the dynasty of the Kurus came into being.
Vaiśaṁpāyana: A dynast of the Pauravas was a mighty hero called
Duḥṣanta, herdsman of all the earth to her four horizons.
…
From Bharata springs the Bhārata frame, from the Bhārata race and those
other ancient men who are famed as Bhāratas. In the continuing lineage
of Bharata there arose great and puissant kings, the likes of Gods, the
likes of Brahmā, whose names are famous beyond measure everywhere.
I shall celebrate those among them who were their chiefs, O Bhārata, the
fortunate and godlike ones, given to truth and honesty.
[Lines 29–33 repeat Bharata’s genealogy; this time the story of Śakuntalā
is clearly presumed:
Ilina begot on Rathantarī five sons, the first being Duḥṣanta. Duḥṣanta
married Viśvāmitra’s daughter Śakuntalā, who bore him Bharata. On that
there are two verses: “The mother is the father’s water sack. He is the
father by whom the son is born. Support your son, Duḥṣanta, do not
reject Śakuntalā. A son who has seed, O king of men, saves from Yama’s
realm. You are the planter of this child, Śakuntalā has spoken the truth.”
Hence his name Bharata.]
…
Thus has the lineage of Pūru and the Pāṇḍavas been described: he who
hears this genealogy of Pūru is freed from all evil. (Mahābhārata
1.62.1–90.96)
In contrast, the southern recension, working almost exclusively with the mate-
rial of the archetypal text, reinstates not only the broad contours of the
narrative (i.e., reversing the sequence of the Śakuntalā—Yayāti episodes) but
also specific linkages (e.g., Janamejaya’s requests to Vaiśaṁpāyana or the lat-
ter’s responses to the king). For instance, at Mahābhārata 1.61.1–2, the southern
recension has Janamejaya utter his initial request,64 but then, instead of having
Vaiśaṁpāyana narrate the story of Śakuntalā, has him declare, “Now I shall
celebrate to you, king sans blame, the lineages of all the Yādavas and Pauravas
as well as the Bhāratas, in the genealogy of Dakṣa Prajāpati, Manu Vaivasvata,
Bhārata, Kuru, Pūru, and Ajamīḍha, and their holy and grand progress in bliss,
which brings wealth and fame and a long life.” Thereafter, the southern recen-
sion has Vaiśaṁpāyana begin with the story of Yayāti, a story that answers
much better to Janamejaya’s request to hear how his dynasty came into being,
since it begins with a remote ancestor Nahuṣa and then continues on to Yayāti
64 “I have heard fully from you, O brahmin, how the Gods, Dānavas, and Rākṣasas, and also
the Gandharvas and Apsarās, descended to earth with a portion of themselves. Now
I wish you to tell me from the beginning in the presence of these brahmins and seers,
O brahmin, how the dynasty of the Kurus came into being.”
100 Bagchee
and Yayāti’s son, Pūru.65 The narration of the Yayāti-Upākhyāna and the
Uttarayāyāta, are perhaps somewhat longer than the king likely anticipated,
but they have the advantage of bringing the narration two generations for-
ward. At the end, his interest whetted by these fantastic accounts of Yayāti and
Pūru, Janamejaya asks: “My lord, I wish to learn who the kings were that
became the dynasts in the line of Pūru, how many they were and what manner
of men, how mighty they were and puissant. For in this dynasty no king ever
lacked in character or prowess or offspring. Of these kings of famous feats and
wisdom I wish to hear the exploits in their fullness, O man of austerities!” and
Vaiśaṁpāyana obliges with a simple list that has the advantage of breaking off
just at the point where Bharata is introduced and the birth of Bharata from
Śakuntalā (1.89.16). Following a formal introductory stanza, Vaiśaṁpāyana can
now recount the story of Bharata’s birth in full and we are better placed to hear
it, for we now know the history and descent of the dynasty and know, also, who
Bharata is and where he fits into that genealogy. When Vaiśaṁpāyana pauses,
he will pause at a most opportune place. It is now time to reintroduce, via a
single-line insertion, the remainder of the chapter, which has not been forgot-
ten, and because the scribes are not trying to hide anything from anyone, they
have no qualms about repeating line 16, which restores the context with the
preceding half of the chapter. The transition works beautifully: from a descrip-
tion of the virtues of the deceased king (“He was a king, a Turner of the Wheel,
a majestic world monarch. He sacrificed many sacrifices, he was an Indra, lord
of the winds. Like Dakṣa, he had Kaṇva officiate at a richly rewarded sacrifice,
and, an illustrious king, Bharata offered a Horse Sacrifice that was styled Vast-
in-Cows, at which he gave a thousand lotus counts of kine to Kaṇva.”) to a
discussion of the merits or lack of merits of the next generation (“It is from him
that the great fame of the line of the Bhāratas began. Bharata begot nine sons
on his three wives, but the king did not approve of any of them, for they were
not of his stature.”) and, thereafter, a continuation of the narrative that brings
the genealogy almost all the way up to the present generation of heroes.66 All
65 It is always a rather arbitrary matter to decide where to cut off genealogies in the Mahāb
hārata. I have chosen Nahuṣa here as someone whose story figures significantly in the
Mahābhārata. Of his ancestors Āyus and Purūravas, little is mentioned. Perhaps the
human genealogy can be traced all the way back to Manu, even though this genealogy
does not appear fully human, since Pūruravas is born only from Īla (she is said to be have
been both his mother and his father; sexual dimorphism or sexual generation appears not
to have emerged as yet).
66 Present, that is, in the sense bringing the genealogy down to a generation of warriors one
or two generations prior to the heroes involved in the great war; not present in the sense
The Epic’s Singularization to Come 101
that remains now is the last two lines of the Śakuntalā-Upākhyāna that had
been suspended while this insertion (1.89.17–55) was being made.
Even though the southern recension changes the order of the narratives, it is
important to note that it does not simply create a new sequence. Rather, work-
ing with a palette provided to it by the archetypal text, the southern recension
reconstructs what is intuitively the better sequence. It is almost as though its
authors were aware of the fact that the transmission has been interrupted
in the northern sources. Their work, though an act of revision, is almost cer-
tainly an act of preservation and restoration rather than a new composition.67
In contrast, the northern sequence appears to be a falling away, a deviation
from the intended sequence. To understand the nature of the southern revi-
sions, consider the fact that, setting aside interpolations to the narrative, the
southern recension makes minimal additions to the text. With just a single
constituted text follows): first, the Śākuntala (or the story of Bharata);
then the genealogy of Yayāti; and finally, the genealogy (in one stretch)
from Yayāti’s son Pūru to Śaṁtanu. Logically, therefore, the Southern
arrangement of the whole of this section is much superior to that of the
rival reason; only it looks, in comparison with the other, a trifle artificial,
as though it were an afterthought, conceived and carried out by a
diaskeuast.”68
Sukthankar thus sees that the scribes responsible for the revisions are not
working with new narratives, but re-sequencing a set of narratives following
an order that we too perceive to be intuitive. The scribes are not modifying the
epic; rather, they appear to be picking up a text that they realize has been put
together wrongly, identifying its breaks or joints or fault lines, and, with a mini-
mum of insertions,69 taking the text as it has been recompiled apart and
bringing it together again. To invoke once again the metaphor of an archaeo-
logical site (and to now bring it to its correct and fullest expression), these
scribes are like curators in a museum who, noticing that a particular artefact
(e.g., an urn) has been put together incorrectly, take it apart along its seams
and reassemble it. They do not make new breaks in the pottery; their entire
concern is curatorial, restorative, and the only intervention they permit them-
selves is the use of some glue (i.e., the new lines, which they compose for the
occasion) and this too is something they do with extreme care.
The Śakuntalā-Upākhyāna in the southern recension likewise begins with
a minimum of adaptation. A single two-line insertion suffices to bridge the
transition and to announce the new chapter (Mahābhārata 571*.1–2: bhaga-
van vistareṇeha bharatasya mahātmanaḥ | janma karma ca śuśrūṣus tan me
śaṁsitum arhasi |) and thereafter almost the whole of the Śakuntalā narra-
tive can come in. Yet, the remaining half of chapter 89 is neither lost from
sight nor tacked on randomly to some other bit of the narrative. At 1.62.48,
the scribes identify a natural joint (the concluding portion of the chapter
is a retrospective summary and a prospective declaration70) and choose to
insert the remainder of 1.89 here. The repetition of line 16, even though used
earlier, is a clear sign that the scribes were aware that they were revising the
text.71 It should not be read as evidence of ignorance or carelessness, since
verses in fact repeat quite often verbatim in the Mahābhārata. The verse
provides a fitting conclusion to the substantive portion of the Bharata nar-
rative (which ends by narrating that Bharata offered up a horse sacrifice at
which he gave a thousand head of cattle to Kaṇva); it also serves to bracket
off the Śākuntala by recalling its opening line. The authors have achieved the
near-impossible: they have managed to both fit the Śakuntalā-Upākhyāna
into its proper place in the Pūru genealogy and yet to highlight its nature
as a separate, self-contained unit within that genealogy. All they need do is
compose a three-line insertion referring to Bharata’s horse sacrifice (878*.1–
3: so ’śvamedhaśatair īje yamunām anu tīragaḥ | triṁśatā ca sarasvatyāṁ
gaṅgām anu catuḥśataiḥ | dauḥṣantir bharato yajñair īje śākuntalo nṛpaḥ |)
to bring the narration up to its end (the repeated line 16) and they can now
insert the Pūru genealogy from Bharata onward, culminating in the refer-
ence to the generation of warriors but two before the present generation of
epic heroes—Bāhlīka, Devāpi, and Śaṁtanu.
The conclusion of the Śākuntala (Mahābhārata 1.62.49–55) follows as
though automatically after the end of chapter 89 (I ignore here the longer
insertion 894*, which seems to me to be one of the substantive additions to the
narrative and not made for the purpose of smoothing over the transitions), and
thereafter the main narrative (the Kuru narrative) resumes from 1.91 onward.
Chapter 90 is moved up entirely (to before the description of the incarnation
of the Mahābhārata narrative in the genealogy of Vasu at 1.57) and functions
now not as it does after 1.89 as a genealogical narrative (where it is obviously
wrong because Janamejaya asks to hear again in “greater detail” [cf. vistareṇa;
1.90.2c] but the succeeding narrative at 96 lines is much shorter and more sum-
marizing than the preceding genealogy of Yayāti/Pūru) but as a prospectus of
the entire epic. Thus, it begins with the incarnation of the genealogy from
Dakṣa; proceeds via summaries of the genealogies of Purūravas, Āyus, and
Nahuṣa; mentions the names of Devayānī and Śarmiṣṭha (who will soon play
a major role in the Yayāti narrative); already indicates the partitioning of the
Nāhuṣa genealogy into two branches, a divine one headed by Devayānī, which
will ultimately give rise to Kṛṣṇa of the Yādavas, and a demonic one headed by
Śarmiṣṭha, which will give rise to the evil-minded and lawless Kauravas; and
provides, in two pithy verses, a kind of synopsis of the story of Bharata.72
Although in this case it is less easy to determine whether the scribes merely
restored an implicit order or have actually modified the narrative, the displace-
ment of 1.90 to before 1.56 is of a piece with the other changes in the southern
recension. As with the inversion of the Śakuntalā and Yayāti Upākhyānas, the
actual extent of changes made to fit the chapter into its new place are minimal.
For instance, the scribes delete the entire first five lines (and substitute the
sixth line), which would have made no sense in this new context. In the consti-
tuted text, the opening of chapter 90 reads:
I have now heard from you, O brahmin, the vast origins of our ancestors,
and the noble kings in our lineage have been recounted. But this account,
which I hold dear, has been retold too briefly to please me. Therefore, tell
it to me again with greater detail, the same divine account from Manu
Prajāpati onward. For to whom should their holy genealogy not bring,
joy, sublime and increased with the glorification of the virtues of the Law
of the good? Abundant fame stands firmly rooted in all three worlds, of
these men who were gifted with virtues and might and prowess and puis-
sance, character, and enterprise. I cannot listen enough to this history
that has the taste of the Elixir of immortality. (Mahābhārata 1.90.1–5)
This passage makes no sense in the chapter’s new position (after Mahābhārata
1.56), for Janamejaya can hardly ask to hear again in greater detail the geneal-
ogy of his ancestors if this is the first genealogy he is told in the epic. The scribes
therefore delete this entire segment; the speaker is now, as before (i.e., in the
preceding chapter), Vaiśaṁpāyana and, with a two-line transition (896*.1–2:
pūror vaṁśam ahaṁ dhanyaṁ rājñām amitatejasām | pravakṣyāmi pitṝṇāṁ te
teṣāṁ nāmāni me śṛṇu |) he continues the narration of the Mahābhārata. Yet,
72 These verses need not be oversight, indicating that the story of Bharata is already pre-
sumed at this point in the narrative, because there are many other instances in the chapter
of pithy one-verse or two-verse sayings indicating at the meaning or the role of particular
characters or narratives (e.g., at 1.90.47–49: atrānuvaṁśo bhavati | yaṁ yaṁ karābhyāṁ
spṛśati jīrṇaṁ sa sukham aśnute | punar yuvā ca bhavati tasmāt taṁ śaṁtanuṁ viduḥ
| tad asya śaṁtanutvam | or at 1.90.91–92: tam utsaṅgena pratijagrāha pṛthā niyogāt
puruṣottamasya vāsudevasya | ṣāṇmāsikaṁ garbham aham enaṁ jīvayiṣyāmīti ||
saṁjīvayitvā cainam uvāca | parikṣīṇe kule jāto bhavatv ayaṁ parikṣin nāmeti ||).
106 Bagchee
efforts have clearly been made to integrate this prose genealogy into its new
place. First, there is the repetition of pravakṣyāmi at line 2, echoing that at
1.55.2c (pravakṣyāmi mataṁ kṛtsnaṁ vyāsasyāmitatejasaḥ) and at 1.56.12c
(pravakṣyāmi mataṁ kṛtsnaṁ vyāsasyāmitatejasaḥ; note also the evocation of
the amitatejasaḥ of both lines in amitatejasām). Second, the concluding
phalaśruti of this chapter (i.e., 1.90) also echoes the preceding eulogy of the
Bhāratas. Once again, there appears to be a conscious effort to recall the intent
of the preceding section, especially in the near identical concluding line. In
juxtaposition, the parallel between the two phalaśrutis is unmistakable:
The Mahābhārata, they say, is the great Birth of the Bhāratas; he who
knows this etymology is rid of all his sins [niruktam asya yo veda
sarvapāpaiḥ pramucyate]. (Mahābhārata 1.56.28–31)
Thus has this lineage of Pūru and the Pāṇḍavas been described: he who
hears this genealogy of Pūru is freed from all evil [pūror vaṁśam imaṁ
śrutvā sarvapāpaiḥ pramucyate]. (Mahābhārata 1.90.96)73
This leaves us with just one final circumstance to account for and this is the
“palpable hiatus” Sukthankar refers to between the end of chapter 69 and the
beginning of chapter 70.74 Sukthankar notes that this hiatus opens up between
these two chapters because “the thread of the narrative dropped at the end of
adhyāya 69 seems to be resumed at adhyāya 89 (or, strictly speaking, at stanza
17 0f that adhyāya), after skipping the entire Yayāti episode.”75 In translation,
the two passages are as follows:
From Bharata springs the Bhārata frame, from the Bhārata race and those
other ancient men who are famed as Bhāratas. In the continuing lineage
of Bharata there arose great and puissant kings, the likes of Gods, the
likes of Brahmā, whose names are famous beyond measure everywhere.
73 The latter phalaśruti is, in fact, the southern one. It replaces the northern version, of
which Sukthankar notes that it is a “rambling phalaśruti [and] contains some repetition
and bears other signs of inflation.” V.S. Sukthankar, ed., The Mahābhārata for the First
Time Critically Edited, vol. 1. The Ādiparvan (Poona: Bhandarkar Oriental Research Insti-
tute, 1933), 397. In a deviation from his normal editorial practice, Sukthankar therefore
opted to set the southern benediction above the line, relegating the inflated northern ver-
sion to the critical apparatus.
74 Sukthankar, “Editorial Note (3),” i.
75 Ibid.
The Epic’s Singularization to Come 107
I shall celebrate those among them who were their chiefs, O Bhārata, the
fortunate and godlike ones, given to truth and honesty. (Mahābhārata
1.69.50–51)
Now I shall celebrate to you, king sans blame, the lineages of all the
Yādavas and Pauravas as well as the Bhāratas, in the genealogy of Dakṣa
Prajāpati, Manu Vaivasvata, Bhārata, Kuru, Pūru, and Ajamīḍha, and their
holy and grand progress in bliss, which brings wealth and fame and a long
life. (Mahābhārata 1.70.1–2)
From Bharata springs the Bhārata frame, from the Bhārata race and those
other ancient men who are famed as Bhāratas. In the continuing lineage
of Bharata there arose great and puissant kings, the likes of Gods, the
likes of Brahmā, whose names are famous beyond measure everywhere. I
shall celebrate those among them who were their chiefs, O Bhārata, the
fortunate and godlike ones, given to truth and honesty. (Mahābhārata
1.69.50–51)
There was a king of the name of Mahābhiṣa, a lord of the earth who
sprang from the dynasty of Ikṣvāku, true in his promises and of proven
prowess. With a thousand Horse Sacrifices and a hundred Horse Race fes-
tivals he satisfied the lord of the Gods, and so the king attained to heaven.
(Mahābhārata 1.91.1–2)
that the southern arrangement of the narratives was a revision, and I think this
is substantially correct. The southern arrangement does presume knowledge
of the archetypal text. As I have shown, it does not create a new sequence or a
new text but works with a palette of materials that have been provided it by the
archetypal text. The repetition of 89.16cd after 878*.3 and a whole host of other
such details show this to be the case. Yet, what this reconstruction does suggest
is that the relation of the two recensions to each other must be rethought: it is
not a simple matter of an original arrangement present in the northern recen-
sion that is violated or remade by the southern recension.
T.P. Mahadevan has argued in all his articles that the Pūrvaśikhā Brahmans
brought to the south of India a northern recension text and that this text served
as the basis of their (inflated) southern version.77 This cannot be true for the
simple reason that the northern recension is by definition not the text from
which the southern recension descended. On the contrary, the northern and
southern recensions are both descended from a common ancestor, the
78 Again, note that I do not think this was carried out all at once. It is a short—and seem-
ingly obvious—step from speaking about a recension to speaking about a “redaction,” but
it is a temptation we must avoid.
79 This statement is also not without its problems, but for now I have chosen to address only
the main confusion (i.e., between the northern recension [N] and a text of the northern
type). See, however, n. 94 for the problems with where to place the southern recension.
80 Incidentally, I cannot agree with Mahadevan’s assertion that “Sukthankar’s European
training orients him to a special type of diaskeuasis, to produce a critical edition of the
text that is consonant with the two well-known desiderata of the praxis of critical editing,
as developed in the nineteenth century from the editing of the great classical texts in the
West: the shortest version, attested in all manuscript traditions, is likely to be the arche-
type; and the texts of a given script hang together.” Mahadevan, “The Śakuntalā-Yayāti
Transposition,” 60 (the same error also occurs in Wolfgang Morgenroth, “Vishnu Sitaram
Sukthankar as a Student in Berlin, 1911–1914,” Annals of the Bhandarkar Oriental Research
Institute 58/59, Diamond Jubilee Volume [1977–78]: 193–201, but is there explained by the
German Indologists’ need to claim Sukthankar as one of their own). Sukthankar’s training
may have been European, but his work is thoroughly Indian, being carried out in the same
spirit of conservatism and concern with preservation of the tradition that he addresses
when he writes, “taking away something from the received text of the Mahābhārata and
passing it off as the original work is a thing categorically different from adding something
to it. To add small details here and there, embellishing and amplifying the original, would
be merely a gentle and lowly service ad majorem gloriam dei.” Sukthankar, “Prolegomena,”
lii. The decision to not omit “even the seemingly most irrelevant line or stanza, actually
found in a Mahābhārata manuscript collated for the edition … on any account” is
The Epic’s Singularization to Come 111
ultimately rooted in his debt to this tradition. Sukthankar explicitly acknowledges the
debt in his “Epic Studies III: Dr. Ruben on the Critical Edition of the Mahābhārata,”
Annals of the Bhandarkar Oriental Institute 11, no. 3 (1930): 259–83, when he notes: “No
orthodox Hindu work can begin without a maṅgala; and this edition of the Mahābhārata,
critical though it be, is and remains a Hindu work, which, could not dispense with a
maṅgala.” Ibid., 268.
81 Sukthankar, “Prolegomena,” xci (Sukthankar’s italics).
82 Those who treat the southern recension merely as a derivative of the northern end up
cutting off the very branch they are sitting on, because the Critical Edition’s reconstruc-
tion is based on the assumption that the agreement of N with S is the strongest argument
for their reading being the reading of the archetype. If S is treated as a derivative of N, they
are no longer independent witnesses for each other, which undermines the editor’s claim
to have been able, on the basis of their evidence, to reconstruct a stage of the transmis-
sion older than that contained in either taken by itself.
83 Contini frames the question in the same way: “A multiple innovation at the same varia-
tion place does not preclude reasoning: why have all the manuscripts […] innovated, and
in a colorless manner to boot? Was this not because there was an objective obstacle in
the original?” Gianfranco Contini, “La critica testuale come studio di strutture,” in Bre-
viario di ecdotica (Milan: Ricciardi, 1986), 140 (the translation is Trovato’s). Before Contini
(originally in a conference paper in 1967), I find that Edgerton had articulated a simi-
lar principle: “Moreover, in many cases the context may suggest easily, and so to speak
almost necessarily, a particular alteration; especially if the original had any troublesome
feature, such as irregularity in metre or saṁdhi or grammar, or if it was recondite and
difficult in interpretation. In such cases it often happens that a particular change is so
natural that we may easily suppose it was introduced by more than one person, without
mutual contact of any sort. On the other hand, it often happens in such cases that differ-
ent changes—sometimes quite a number of them—are made, apparently motivated in
the same way (namely, to remove what was felt as a troublesome feature of the original).”
Franklin Edgerton, “Introduction,” in The Sabhāparvan for the First Time Critically Edited
(Poona: Bhandarkar Oriental Research Institute, 1944), xxxix.
112 Bagchee
Such would seem, indeed, to be the case in the Śakuntalā and Yayāti
upākhyānas. The southern recension is plainly modifying an existing text but
that text, it seems to me, is not the northern recension. The divisions the south-
ern recension draws dovetail too neatly into each other for us to believe that
they were not already present, either implicitly or in some explicitly realized
form, in the text of its source. There is (or, rather, there was) a text that had a
clearly defined architecture: whether the southern recension preserves that
architecture or is re-creating it after its disruption is something we shall prob-
ably never know. What is clear is that the commonalities between the northern
and southern recensions concerning these two episodes are a case of a com-
mon inheritance and not of borrowing (i.e., of the southern from the northern).
It is as likely that this architecture was lost in the northern recension (explain-
ing the fact that it has the same episodes and the same joints as the southern
recension, only in a jumbled form) as that the northern recension accurately
and completely preserved the reading of the archetype and the southern
recension improved on it.84
T.P. Mahadevan has argued in numerous articles for the literary sophistica-
tion of the “southern redactors,”85 whom he identifies with the Pūrvaśikhā
Brahmans who migrated to Tamil country around the third century CE.86 He
considers that this period presents us with the necessary literary and scribal
culture to undertake such a large-scale re-envisioning of the epic.87 I agree
84 We must remember that in both cases, we are speaking only of hypothetical archetypes,
reconstructed on the basis of the extant witnesses. We do not have access to these texts
nor do we know that they existed exactly in the form that we think they might have
existed. One cannot hypostatize N and S and attribute, on their basis, real movements to
people, who would have been carrying these precise texts.
85 Mahadevan, “The Śakuntalā-Yayāti ‘Transposition’,” 52; ibid., 57, where they are referred to
as “SR redactors”; and see also Mahadevan, “The Southern Recension of the Mahābhārata,
the Harivaṁśa, and Āḻvār Vaiṣṇavism,” 93–95, 100–105, and 108–9.
86 See his “The Southern Recension of the Mahābhārata, the Harivaṁśa, and Āḻvār Vaiṣṇa
vism,” 63–110, see especially the section titled “Redactors, Scribes, Manuscripts and Pale-
ography” (99–110).
87 See ibid., 103–4: “As I have argued, the Pūrvaśikhā Brahmans in all probability possessed a
script at their departure from the Vedic realm, and from [Iravatam] Mahadevan’s data, it
has to be a form of the Southern Brāhmī, eventually giving rise to the Grantha script in the
Tamil country and the Ārya-eḻuttu in Malabar, both able to meet the phonology of San-
skrit. This gives us the script of the *SR, taking shape ca. 300 CE in the Tamil country, the
Pūrvaśikhā Brahmans functioning digraphically in the peninsula, during the Sangam
period, taking part in the Sangam poetic culture through the Tamil Brāhmī script and
redacting the *SR in a Southern Brāhmī script they brought with them. [Iravatam]
The Epic’s Singularization to Come 113
Mahadevan has drawn emphatic attention to the wide-spread literacy of the Tamil coun-
try in the Sangam period, far earlier in it than in the regions that surround it in the penin-
sula. Wide-spread literacy would imply an active scribal culture, with skilled scribes,
writing implements and the necessary manuscript materials. As we know, the very mode
of writing—holding an iron stylus enclosed tightly in the right fist, the fingers playing no
role in the calligraphy—shapes the evolution of the Tamil Brāhmī syllabary, becoming
what comes to be called vaṭṭeḻuttu (“round writing”) in time; it is reasonable to think that
the self-ascribing term, vaṭṭeḻuttu, driving from a specific scribal mode, is itself evidence
of a wide-spread writing culture. As a corollary, it is possible to envisage the rise of the
written text of the *SR in the Sangam period.”
88 I take “scribes” quite literally; I think the changes to the southern tradition are better and
more economically explained if one assumes the standard processes of emendation and
corruption and revision that literate scribes are known to undertake. Mahadevan, how-
ever, assumes a single, concerted “redaction.”
89 I have found only one instance where Mahadevan distinguishes between them; it is in a
long footnote, summarizing the findings of his 2008 EJVS study (Thennilapuram P.
Mahadevan, “On the Southern Recension of the Mahābhārata, Brahman Migrations, and
Brāhmī Paleography,” Electronic Journal of Vedic Studies 15, no. 2 [2008]: 1–143), in his 2013
article. In this note, Mahadevan writes: “A version of the epic, possibly the archetype or
the *Śāradā text, came to the peninsula, in what evolves later there as the Southern
Brāhmī script, with the first group (Pūrvaśikhā Brahmans) by the beginning of the Com-
mon Era, from which they create the SR in the first half of the millenium.” Mahadevan,
“The Southern Recension of the Mahābhārata, the Harivaṁśa, and Āḻvār Vaiṣṇavism.” 63,
n. 2. It is the only place I find that Mahadevan uses the word “archetype”; otherwise, he
confuses the northern recension with the archetype, imagining the former to have been
the source of the southern recension.
90 Mahadevan, “The Śakuntalā-Yayāti Transposition,” 52.
114 Bagchee
91 Ibid., 57.
92 See, for instance, his comment that “in the case of the recensions, this [i.e., the fact that
all our Mahābhārata manuscripts are descendants of a single, common archetype] forces
us to accept the presence of a Mbh1 in the peninsula to function as a template for the
Mbh2” (ibid., 62) and see also his identification of this Mbh1 text with the northern recen-
sion (an “NR Mahābhārata text,” he calls it); ibid., 47.
93 See ibid., 62.
94 The reader will have noticed that I have been careful to avoid using Mahadevan’s “south-
ern redactors” just as, throughout, I have been careful to distinguish “southern recension”
from “the recension from the south.” This is because I do not think the case has been
made for these scribes (I see no need to call them “redactors”; they are scribes and schol-
ars in the sense that Reynolds and Wilson spoke of with regard to the copyists of Greek
and Latin manuscripts) being from the south. I also do not think that the move from
“southern recension” to “the south Indian recension” (via the intermediate stage of an
ambiguous “recension from south India”) is warranted, however self-evident such an
identification may seem. The definition of the “southern recension,” it may help to recall,
is: the hypothetical archetype of all our southern manuscripts or, alternatively, the latest
ancestor that the manuscript groups T, G, and M have in common. I find nothing in Suk-
thankar’s work to suggest that the origins of this recension would have been in the south.
In fact, the earliest reference I can find identifying a location in south India for this recen-
sion is the work of T.P. Mahadevan (specifically, in his 2008 study). Mahadevan, however,
does not see that the southern recension is a hypothetical archetype, inferred on the basis
of certain common features of the southern manuscripts: he takes it to be a real text and
attributes the origin of those features to this text. Thus, according to him, not only was
Sukthankar’s S a physical reality but it was also made all at once, that is, there was one
single concerted effort at redaction (of a northern source). Based on his research on Brah-
man migration, he places this redaction in south India (the Tamil country) and attributes
it to the Pūrvaśikhā Brahmans (who he thinks brought Sukthankar’s N south). This is to
treat hypothetical ancestors as real manuscripts and, moreover, to significantly distort
The Epic’s Singularization to Come 115
fact, we can dispense with this text, except as a kind of heuristic aid, altogether:
what the scribes are really modifying is not a physical text (this is Mahadevan’s
error: he makes the entire process mechanical and does not realize that in
doing so he puts his Pūrvaśikhā Brahmans at one remove from the epic’s com-
posers) but an architecture in their heads. They are the ones who know (not
merely “correct”) what the correct sequence is, because they are the ones who
have intended or worked out the narrative. This is why they do not need to
compose new narratives or significantly modify their text to bring it into order:
the order is already there; it merely requires elucidation.
Let me be very clear: I am not arguing that the southern recension does not
represent a revision. I would not debate Sukthankar’s findings and indeed,
I think the situation is very clear: the southern recension is a further develop-
ment of the archetype. What it, however, is not is a further development of the
northern recension. Thus, what I am really arguing against is the tendency to
view the southern recension as a modification of the northern recension and
then to imagine that in cases where we are unable to reconstruct the reading of
the archetype all we have to do is substitute the reading of the northern recen-
sion and we will soon have the original.95 Seen from this perspective, the
southern recension can only represent a fall away from the northern; it must
then be a corruption of the latter. Yet, the question of the southern recension’s
changes to the Mahābhārata cannot be reduced to a question of mechanical
changes to a physical template. It is not a case of two static texts (or, rather, of
one static text and its undesirable but dynamic revision in the other); rather, it
is a case of an intellectual architecture, which has been and (for the period we
are speaking of) is being realized in two very different ways in the northern and
southern recensions. The Pūrvaśikhā Brahmans, if we are to accept Mahadevan’s
claim that they are the agents responsible for the southern recension,96 are the
very authors and architects of the epic and not just its recipients or custodians.
They are completing with the southern recension what they began doing or set
out to do with the archetype but, for various reasons could not carry out to its
fullest extent. The southern recension may be more northern than the north-
ern recension itself!97
I have shown in the first two sections of this article that, at least in the case
of their revision of the Śakuntalā and Yayāti episodes, they can be seen to be
doing one of three things:
96 I am skeptical of this claim (see n. 94), but here I wish to represent Mahadevan’s views. I
would personally have preferred the term “scribe.” I also think that we must assume a
longer period of time before the southern recension attained its final form (final not in
the sense of “intended” or “purposefully shaped,” but in the sense of “form in which we
now infer it to be”).
97 I am, obviously, highly critical of the view that there is something like an identifiably
“southern” understanding of the epic, which understanding, moreover, is taken to be
characteristic of south India (in an illegitimate move that replaces “southern” with “south
Indian”). I do not see that this case has been made either on historical or text-critical
grounds; it is mostly a prejudice deriving from “higher critical” views (which, in turn,
derive from modern, i.e., late nineteenth-century prejudices about north and south
India). At any rate, I find nothing in Sukthankar to associate the “precision, schematiza-
tion, and thoroughly practical outlook” (Sukthankar, “Prolegomena,” xxxvi; his italics) he
found to be characteristic of the southern recension with south India. The southern tradi-
tion is probably the fullest realization of the Mahābhārata architecture (to use an expres-
sion from Adluri) for the simple reason that manuscripts were being copied and read with
greater intensity in this part of India than any other.
The Epic’s Singularization to Come 117
2a. Either it was fully realized in their archetype and then was lost,
giving rise to the varying attempts to recover the sequence in the
northern and southern recensions; or
2b. it was present only implicitly in their archetype and they began the
work of completing it as the manuscripts were successively copied
and (re)edited.
Of the three scenarios (i.e., 1, 2a, and 2b), I think the first is impossible, because
it would contradict much of what we do know about the Mahābhārata tradi-
tion. It would also create insurmountable obstacles for textual criticism of the
epic, even undermining Sukthankar’s signal achievement in establishing for
the most part the reading of the common archetype of the two recensions.
I also think the second is highly unlikely, though not impossible. It would
explain the divergence between the northern and southern recension, with-
out according one tradition priority. But of the three, it is the third option
I most prefer. There are three reasons why it appears, to my mind, to be the
best solution: first, it retains the priority of the northern recension as required
by Sukthankar’s critical principles (something the second solution fails to
recognize); second, it accords well with the observed fact that, although the
solutions of the two recensions are different, the problem is clearly common,
being inherited (this is something the first solution sidesteps in making the
problem a purely northern one); and third and most important, it reconciles
the chronological priority of the northern recension with the logical superiority
of the southern recension.
If this is indeed the case, it seems to me that we must also begin to recon-
sider the relation of the two recensions to each other. There is a long-standing
prejudice in Mahābhārata scholarship that all that the southern recension
does is to provide an inflationary version of the northern. Concomitant with
this is the assumption that the southern text is therefore the less accurate text,
that it is less faithful to the text of the archetype than the northern recension,
and that therefore its variants can be neglected in all cases of doubt for the
northern readings. This has been the case in spite of Sukthankar’s cautions
that:
It should thus seem that the infidelities of the Southern recension are
confined mainly to a tendency to inflation and elaboration. In parts unaf-
fected by this tendency, it is likely to prove, on the whole, purer, more
conservative and more archaic than even the best Northern version. The
118 Bagchee
Southern variants, therefore, deserve the closest attention and most sym-
pathetic study.98
1. We do not know what the source (Vorlage) of the scribes and authors of
the southern recension was.
2. In cases of crux, we do not know which of the two recensions has
preserved the correct reading.
3. We do not know if these scribes and authors were modifying, dilating
on, or merely restoring the text.100
As I have shown in this article, at least in the case of the Śakuntalā and Yayāti
upākhyānas the presumption is strongly in favor of the scribes restoring the
text. Although it is impossible to know whether the southern recension pre-
serves an order explicitly realized in the archetype or one that was implicitly
present but never brought to completion in the archetype or one that was real-
ized in the archetype but lost in the northern recension and late reconstructed
in the southern recension, it seems clear that the scribes’ revision is not an act
of innovation but one of restitution. This corresponds perfectly with what
Sukthankar observed to be their “conservative” strain. It also explains why the
southern changes do not appear forced to us but merely capitalize on breaks
we also feel to be in the text. It is a case of a Vorlage that makes a certain
Vorgabe (a prompt or a guideline or a specification) and, in this case, the south-
ern recension seems to respond correctly to that Vorgabe, whereas in the north
the Vorgabe either went unheard or was instituted but then lost and conditions
were such that a later generation of scribes and scholars could never respond
to it.
Conclusion
This chapter has shown that the textual history of the Śakuntalā-Upākhyāna
offers a wealth of clues to the textual history of the Mahābhārata. Even though
the common assumption is that the upākhyānas are interpolations and in the
case of the Śakuntalā-Upākhyāna this assumption is further reinforced by the
fact that the manuscripts are uncertain about its placement in the epic, this
assumption risks overlooking what is most essential about the upākhyānas.
Precisely because they are relatively well-defined, independent narrative units,
the upākhyānas can sometimes be of the greatest use in interpreting the
unfolding of the Mahābhārata narrative. The places they occur (or fail to occur)
can be important indices not only for the state of development of the narrative
but also of the textual corpus itself.
This chapter also showed that the shifting placement of the Śakuntalā-
Upākhyāna ought not be taken as evidence of indecision or hesitancy about
the narrative. Rather, in both recensions and especially in the recension that
appears to not place the episode correctly there is a similar narrative logic. The
southern recension makes this logic explicit; with the expedient of a handful of
extra verses, it achieves a much more consistent unfolding of the first ninety-
one or ninety-two chapters of the Mahābhārata. This unfolding, I once again
emphasize, may or may not be later but even if it is later in time it can still be
earlier in terms of logical priority: the southern recension capitalizes on an
architecture, a plan, and a vision already latent in the archetype. Recent schol-
arship has made it fashionable to speak of the southern recension as a
“makeover” of the northern recension of the Mahābhārata, by which is often
implied that the southern recension deflects from the blueprint of the Mahā
bhārata.101 That this cannot be so is clear from three circumstances:
1. The southern recension does not modify the northern recension but a
shared ancestor, the text of the archetype.102
2. The southern recension’s changes are made in an essentially conserva-
tive spirit. Thus, even when this recension adds to the narrative, these additions
are made in the interests of restitution rather than innovation.
3. As I have shown specifically of the Śakuntalā-Upākhyāna, following a
hint of Sukthankar’s that “in parts unaffected by this tendency [i.e., the “ten-
dency to inflation and elaboration”], it [the southern recension] is likely to
prove, on the whole, purer, more conservative, and more archaic than even the
101 See, for instance, Alf Hiltebeitel, “From Ṛṣidharma to Vānaprastha: the Southern Recen-
sion Makeover of the Mahābhārata’s Umā-Maheśvara Saṃvāda,” in The Churning of the
Epics and Purāṇas at the 15th World Sanskrit Conference, ed. Simon Brodbeck, Alf Hilte
beitel, Adam Bowles (New Delhi: Rashtriya Sanskrit Sansthan and D.K. Printworld, in
press), 14–45.
102 Although this can be shown quite easily via a comparison of their readings, it would take
time to reconstruct even a few lines of the Śakuntalā-Upākhyāna. In lieu of such a dem-
onstration, let me cite Sukthankar’s discussion in the “Prolegomena”: “The presence of an
astonishingly large number of additions, some of which are undoubtedly late and spuri-
ous, should not be allowed to impair our appreciation of some real merits of the Southern
recension. It would be, in fact, a grievous error to ignore on that account the Southern
recension or underestimate its value. This recension is an indispensable aid for control-
ling the deviations of the Northern recension, both in point of readings and sequence. In
comparison with γ, it has unquestionably preserved a very large number of original read-
ings, proved by actual agreements between S and ν, as well as by their intrinsic merits. The
superiority of the Southern recension in comparison to the Vulgate may be said to be
quite evident. It may, however, quite easily happen that in a particular instance, the whole
of the Northern recension is corrupt, and the true reading is preserved only in the South-
ern recension.” Sukthankar, “Prolegomena,” xlv–xlvi (italics in original). Sukthankar goes
on to show of one verse (Mahābhārata 1.214.5) that the southern reading is incontrovert-
ibly superior.
The Epic’s Singularization to Come 121
and instead start distinguishing between two possibly concurrent but almost
certainly competing impulses within the intellectual life of the scribes and
authors of the epic: the tendency of restoration leading to an epic that is the
accurate image of the past and the tendency of realization leading to an epic
that is the accurate but only the anticipated image of a future still to come.107
Indeed, as Adluri has seen, this fact (i.e., the dual task of preservation and
interpretation) is noted by the epic itself: “Learned men elucidate the complex
erudition of this Grand Collection; there are those who are experienced in
explaining it, others in retaining it.”108 Adluri argues that this passage ought to
be read as a reference to the two agencies or two skills that played a role in the
preservation of the epic. He writes:
Those who retain it are skilled in memory (smṛti) and those who are
skilled in explaining it are skilled in hermeneutics. This is the twin task of
the introductory nature of the list of contents in particular and the
Ādiparvan in general. Smṛti goes backward into the past, while herme-
neutics goes forward, bringing the text to us in the future. The smṛti task
of memory, that is, the philological task, which looks backward, is
described in verses 51–94, and the hermeneutic task or the philosophical
task, which is not textual but eschatological, is described in verses
95–160.109
It is in this sense that I have offered a reading of the epic’s differing treatment
of the Śakuntalā and Yayāti upākhyānas in the northern and southern recen-
corruption inspired by the anti-Semitic fears of the German scholars). (On the latter, see
Vishwa Adluri and Joydeep Bagchee, The Nay Science: A History of German Indology [New
York: Oxford University Press, 2014]). Rather, addition and interpretation are part of the
living life of the text. They actually preserve the text with greater fidelity than a naïve
historicism would achieve.
107 V.S. Sukthankar had a similar intuition, when he wrote in his “Prolegomena” that “If the
epic is to continue to be a vital force in the life of a progressive people, it must be a slow-
changing book! The fact of expurgation and elaboration is only an outward indication of
its being a book of inspiration and guidance in life, and not merely a book lying unused
and forgotten on a dusty book-shelf. Those are probably just the touches that have saved
the Mahābhārata from the fate of being consigned to the limbo of oblivion, which has
befallen its sister epics like the Gilgamesh.” Sukthankar, “Prolegomena,” ci (italics in origi-
nal).
108 vividhaṁ saṁhitājñānaṁ dīpayanti manīṣiṇaḥ |
vyākhyātuṁ kuśalāḥ ke cid granthaṁ dhārayituṁ pare || (Mahābhārata 1.1.51)
109 Vishwa Adluri, “Frame Narratives and Forked Beginnings: Or, How to Read the Ādiparvan,”
Journal of Vaishnava Studies 19, no.2 (Spring 2011): 170–71.
The Epic’s Singularization to Come 123
110 I have spoken in the preceding sections of the “southern” arrangement of the Śakuntalā
and Yayāti upākhyānas, but this expression was, of course, an abbreviation and a simplifi-
cation. There is no arrangement common to the TG and M recensions; as the critical
apparatus materials included above indicate, there are variances even between the T and
G manuscripts. However, my reasons for focusing only the TG sequence here is not simply
that M diverges in key respects: rather, the question of the reading of their common
archetype (i.e., S) requires much more reflection and study than I am able to provide here.
It cannot be done merely by looking at the critical apparatus of the Mahābhārata Critical
Edition. In this respect, I am skeptical of Hiltebeitel’s view, “he [Sukthankar] did not see
well enough what his Critical Edition was unveiling …. Most volume editors of the Pune
Critical Edition subordinated their work with southern manuscripts to the reconstitution
of an all-Indian Mahābhārata. But miraculously, for the two volumes he edited, the first
(Ādiparvan) and third (Āraṇyakaparvan), Sukthankar did the side-job of critically editing
what I believe we can call a provisionally reconstituted S recension text, calling it pre-
cisely ‘S’. Sukthankar differentiated ‘S’ from the omnibus Kumbhakonam edition that is
conflated with many Northern Recension passages …. As far as I can see, Sukthankar’s
work in tracing out what a critical edition of the Southern Recension should look like has
not been noticed in scholarly discussion of the Pune Critical Edition. Sukthankar clearly
suggests that ‘S’ is a prototype for a critical edition of the Southern recension. As he says,
‘S is the ultimate source from which all versions of the Southern recension are, directly or
indirectly, derived’!” Hiltebeitel, “The First Reading of ‘Śakuntalā’,” 20. Not only does this
claim run counter to Sukthankar’s stated view, “the gulf between the Northern and the
Southern recensions is so vast, that it is extremely difficult, if not practically impossible,
to reconstruct the Southern text, completely and correctly, from the critical notes of this
edition” (Sukthankar, “Prolegomena,” cv), but it also misrepresents the status of S. When
Sukthankar writes that “S is the ultimate source from which all versions of the Southern
recension are, directly or indirectly, derived” he is no more saying that his “S” will have to
be the prototype for any future critical edition of the southern recension than that his “N”
was the “prototype” for the Mahābhārata Critical Edition when he says “N is the ultimate
source from which all versions of the Northern recension are, directly or indirectly,
derived.” Sukthankar, “Prolegomena,” xxx. Rather, he means that it is the source of all
manuscripts of the southern tradition; if anything, the text any future critical edition
would reconstruct would be his S. Hiltebeitel is misled by his view that Sukthankar pro-
vided us with a preview of what “the Southern Recension should look like,” by which he
means that Sukthankar outlined its “Brahmanic” character, that is, its “penchant for ‘sen-
tentious maxims’; for adding new characters; and its tendency to be ‘much richer in
details, leaving little or nothing to the imagination of the reader or hearer’.” Hiltebeitel,
“The First Reading of ‘Śakuntalā’,” 19. This, of course, is to enter into the domain of “higher
criticism.” I am similarly at a loss as to what to make of Mahadevan’s suggestion that “ a
124 Bagchee
in my treatment of these two narratives that the epic’s fate cannot be treated
as one of inevitable loss and a fall away from the origin: the epic itself possesses
the power to renew itself—it is the Horse Head that, at the dawn of the eon,
will enter the primeval waters to recover the Veda that has been lost.111
References
Adluri, Vishwa. “The Divine Androgyne: Crossing Gender and Breaking Hegemonies in
the Ambā-Upākhyāna of the Mahābhārata.” In Argument and Design: The Unity of
the Mahābhārata, edited by Vishwa Adluri and Joydeep Bagchee, 275–319. Leiden:
E.J. Brill, 2016.
———. “The Vasu Narratives of the Ādiparvan and the Nārāyaṇīya: Some Lexical and
Textual Issues.” In Papers of the Sixth Dubrovnik International Conference on the
Sanskrit Epics and Purāṇas. Zagreb: Croatian Academy of Arts and Sciences,
forthcoming.
Adluri, Vishwa, and Joydeep Bagchee. The Nay Science: A History of German Indology.
New York: Oxford University Press, 2014.
———. Philology and Criticism: A Guide to Mahābhārata Textual Criticism. London:
Anthem Press, 2016.
Biardeau, Madeleine. “Śakuntalā dans l’ épopée.” Indologica Taurinensia 7, Dr. Ludwig
Sternbach Felicitation Volume, Part I (1979): 115–25.
Contini, Gianfranco. “La critica testuale come studio di strutture.” In Breviario di ecdotica,
135–48. Milan: Ricciardi, 1986.
Edgerton, Franklin. “Introduction.” In The Sabhāparvan for the First Time Critically
Edited, ix–l. Poona: Bhandarkar Oriental Research Institute, 1944.
critical edition of the SR may nearly be conjured from the apparatus of the Poona Critical
Edition: it would be essentially Mbh2M with the Mbh2GT ‘excesses’ relegated to the appa-
ratus.” Mahadevan, “The Śakuntalā-Yayāti Transposition,” 61–62. We must give up the idea
that a critical text can be created simply by eliminating the “additional passages” from the
Mahābhārata—an idea that has devestating consequences for the work of Andreas Bigger
and Reinhold Grünendahl, as discussed in Philology and Criticism. The relation of M to TG
and thus to the entire Mahābhārata tradition deserves a separate study.
111 See Mahābhārata 12.355 (the Nārāyaṇīye Hayaśira-Upākhyāna). When the Veda has been
lost, Nārāyaṇa himself takes on the Horse Head (Hayaśiras) form to recover it. This is not
an empty boast: the Mahābhārata self-consciously imitates the form of the horse, with a
horse’s tail at its beginning and the horse’s head at its end. Vishwa Adluri argues (personal
communication) that this “incarnation” is intentional: the Mahābhārata intends the
notion of a transformation of the physical sacrifice into a textual sacrifice quite literally.
The Epic’s Singularization to Come 125
Figueira, Dorothy Matilda. Translating the Orient: The Reception of Śākuntala in Nine
teenth-Century Europe. Albany: State University of New York Press, 1991.
Hiltebeitel, Alf. “From Ṛṣidharma to Vānaprastha: the Southern Recension Makeover of
the Mahābhārata’s Umā-Maheśvara Saṃvāda.” In The Churning of the Ocean Epics
and Purāṇas at the 15th World Sanskrit Conference, edited by Adam Bowles, Simon
Brodbeck, Alf Hiltebeitel, 14–45. New Delhi: Rashtriya Sanskrit Sansthan, in press.
———. “Not Without Subtales: Telling Laws and Truths in the Sanskrit Epics.” Journal
of Indian Philosophy 33, no. 4 (2005): 455–511.
———. “The Southern Recension’s Śakuntalā as a First Reading: A Window on the
Original and the Second Reading by Kālidāsa.” In Revisiting Kālidāsa’s Abhijñāna
Śākuntalam: Land, Love, Languages: Forms of Exchange in Ancient India, edited by
Deepika Tandon and Saswati Sengupta, 17–37. Delhi: Orient BlackSwan Edition, 2011.
Hueckstedt, Robert A. “The Plays of Kalidasa and their Major Twentieth-Century English
Translations.” Journal of South Asian Literature 22, no. 1 (1987): 215–29.
Mahadevan, Thennilapuram. “On the Southern Recension of the Mahābhārata,
Brahman Migrations, and Brāhmī Paleography.” Electronic Journal of Vedic Studies 15,
no. 2 (2008): 1–143.
———. “The Śakuntalā-Yayāti Transposition, the Southern Recension of the Mahā
bhārata, and V.S. Sukthankar.” In The Churning of the Epics and Purāṇas at the 15th
World Sanskrit Conference, edited by Simon Brodbeck, Alf Hiltebeitel, Adam Bowles,
46–70. New Delhi: Rashtriya Sanskrit Sansthan and D.K. Printworld, in press.
———. “The Southern Recension of the Mahābhārata, the Harivaṁśa, and Āḻvār
Vaiṣṇavism.” In Ways and Reasons for Thinking about the Mahābhārata as a Whole,
edited by Vishwa Adluri, 63–117. Pune: Bhandarkar Oriental Research Institute, 2013.
———. “The Three Rails of the Mahābhārata Textual Tradition.” Journal of Vaishnava
Studies 19, no. 2 (2011): 23–69.
Morgenroth, Wolfgang. “Vishnu Sitaram Sukthankar as a Student in Berlin, 1911–1914.”
Annals of the Bhandarkar Oriental Research Institute 58/59, Diamond Jubilee Volume
(1977–78): 193–201.
Salus, Peter H. “Šakuntalā in Europe: The First Thirty Years.” Journal of the American
Oriental Society 84, no. 4 (1964): 417.
Schuyler, Montgomery Jr. “The Editions and Translations of Çakuntalā.” Journal of the
American Oriental Society 22 (1901): 237–48.
Sukthankar, V.S., ed. The Mahābhārata for the First Time Critically Edited, vol. 1. The
Ādiparvan. Poona: Bhandarkar Oriental Research Institute, 1933.
Sukthankar, V.S. “Editorial Note (3).” In The Mahābhārata for the First Time Critically
Edited, vol. 1. The Ādiparvan: Fascicule 4. i–iv. Poona: Bhandarkar Oriental Research
Institute, 1930.
———. “Epic Studies III: Dr. Ruben on the Critical Edition of the Mahābhārata.” Annals
of the Bhandarkar Oriental Institute 11, no. 3 (1930): 259–83.
126 Bagchee
———. “Prolegomena.” In The Mahābhārata for the First Time Critically Edited,
vol. 1. The Ādiparvan. i–cx. Poona: Bhandarkar Oriental Research Institute, 1933.
Sullivan, Bruce M. Seer of the Fifth Veda: Kṛṣṇa Dvaipāyana Vyāsa in the Mahābhārata.
New Delhi: Motilal Banarsidass, 1999.
Chapter 4 127
Chapter 4
1 Since most of my references in this chapter are to the Āraṇyakaparvan, hereafter the number
“3” will not be repeated. Where there is only a single number, the reference is to the adhyāya
in the Āraṇyakaparvan; where there are two numbers, the reference is to the adhyāya and the
verse in the Āraṇyakaparvan.
2 See the valuable article by James Laine, “Out of Character: Marginal Voices and Role-
Transcendence in the Mahābhārata’s Book of the Forest,” Journal of Indian Philosophy 19 (1991):
273–96, which demonstrates the striking levels of initiatory frames and role inversions that
occur throughout this book.
3 One of the heritages of a structural analysis is the tendency to analyse a text as a whole, often
arbitrarily defined, overlooking its individual parts as anything other than variants of the
whole.
Literary Structure of the Mārkaṇḍeyasamāsyāparvan 129
(186–87), are strikingly different we still try and locate common features of
interpretation. We also continually ask whether the fragmentation of content
against desire for unity is intended as another thematic take on one of the
themes running across most of the parvan. This concerns the ubiquitous prob-
lem of the interpretation of dharma and the question of who is its privileged
interpreter. My own feeling, based on repeated readings of the Msp, is that its
texts are speaking to brahmans more than any other group, implying that it is
them as a class who should know the sound interpretations of dharma, even
where in reality they differed amongst themselves on this issue.
“You yourself have seen with your own eyes the great ascetic and seer
Mārkaṇḍeya moving about, a man of boundless soul and long-lived in the
Law …” (32.10; Van Buitenen, p. 282 modified)4
Within the Msp itself there is a constant reinforcement of his virtually unique
status as an observer in the foundational events of the cosmos. His status as an
appropriate teacher for the Pāṇḍavas must be demonstrated not just assumed.
A capacity to see things before one’s very eyes is strongly recommended in
the Msp,5 but it is magnified hugely in the narratives where, and because,
Mārkaṇḍeya himself actually plays a role in the narrated action, in which he
parallels Vyāsa. He describes some of the personal experiences he has had in
order to derive his knowledge of the cosmogony and in response to the ques-
tion asking if there is anyone older than him? Already he has demonstrated his
grand antiquity in the famous myth where he appears on the primeval ocean
and enters Kṛṣṇa’s mouth. What this really means is that he has experienced
the world during the pralaya, and when he actually sees another world ready
to be created in Nārāyaṇa’s stomach. In that sense he is almost beyond time
because he is seemingly capable of describing, on the basis of eyewitness, the
transition between kalpas and the course of the yugas. Above all he is con-
cerned with what is primeval and how this evolves. But even this is not enough,
because in 191, 1 the sages and Pāṇḍavas ask him: “Is anyone older than you?”6
And his explanation in that chapter told them what he had actually experi-
enced (anubhūtam) and seen (dṛṣṭam; 191.26).7 Why is this asked, given that
they already know that he has witnessed events others have never witnessed?
Is it because he has experienced it as well as seen it, thus putting his knowl-
edge on a different epistemological register, from just having learnt about it
second-hand?
The mode of recitation is virtually the same as that employed by Bhīṣma in
the twelfth and thirteenth books of the Mahābhārata. It takes the form of a
dialogue, though it is really a monologue where the subjects treated are sub-
stantially guided by the questions Yudhiṣṭhira and the others put, and this
governs the flow of narrative direction. Whilst all of the teachings and argu-
ments—because this is what they are—are directed towards Yudhiṣṭhira, in
particular, it is surely not insignificant that the other Pāṇdavas also form an
important component of the audience, as they will potentially have to sustain
Yudhiṣṭhira as king when that possibility eventually arises. We are often
reminded that they are present as Mārkaṇḍeya frequently refers to the
Pāṇḍavas in the plural at many points in his narrative.
Equally, the presence of Kṛṣṇa is highly significant, especially, since like
Mārkaṇḍeya himself, he seems to defy/transcend time—for at one point he
is alive during the pralaya and then he is hearing Mārkaṇḍeya describe actu-
ally what happened then. His presence places special light on Mārkaṇḍeya’s
description of his own encounter with Nārāyaṇa, reinforcing the special reve-
latory nature of his own knowledge about cosmogony and cosmology and
his direct experience enlightening the Pāṇdavas about Kṛṣṇa’s true nature,
dramatically enhanced because he is presently with them in his mild appear-
ance as a man, which is quite at variance with his appearance in the more
theophanic mode in chapters 186–87. Here the devotional intervention must
be regarded as a highpoint of the first part of the Msp, containing a particular
version of the bhakti myth occurring elsewhere in the Mahābhārata and the
Purāṇas.8 Such a presentation reaches far beyond the immediate audience, as
everything in the Msp does, but it is a central feature of this myth that the
devotional relationship is always between an individual and a deity, and we
note that, as Hiltebeitel has brilliantly shown,9 the different Pāṇḍavas relate
to Kṛṣṇa in obviously different ways, and even here Mārkaṇḍeya is describing
to a group what can only be an individual experience characterized by high
emotion.
The audience is not just restricted to the Pāṇḍavas and Kṛṣṇa, but also
includes the host of brahmans and tapasvins who are said to have accompa-
nied the former into the forest. These are presented as an undifferentiated host
and so provide another backdrop against which the Pāṇḍavas can be placed.
But in the narratives that are told the brahmans are not presented as an undif-
ferentiated mass, because sometimes they are presented as being opposed to
each other (see 183), and in another text (196–205) as being taught by a woman
and then by a seller of meat, always the dispute occurring over an interpreta-
tion of dharma and the brahman’s role. This suggests to me that the Msp is
strongly directed towards a brahman audience, a view given support by the
fact that as much as forty percent (20 out of 43 chapters) of the Msp consists of
a series of brāhmaṇasya māhātmyāni, directed no doubt to the kṣatriya
Pāṇḍavas, who must rule the kingdom in conjunction with the brahmans act-
ing as consultants.
Finally, Nārada (180.44 arriving before the narration begins) is also present
in the audience, though it is not made clear why he should be so. It could be a
consequence of his status as communicator between the gods and men, and
his presence accordingly enhances the value of the content of what is being
narrated.
8 See my article, “The Semantics of Bhakti in the Vāmanapurāṇa,” Rivista degli Studi Orientali
LXII (1988): 25–57. Having read many more Purāṇas over the years, I am now wondering
whether the structural features defining this myth occur mainly in Vaiṣṇava, and not Śaiva
texts. However, Arjuna’s encounter with Śiva as a mountain man would seem to contradict
this. See Mahābhārata 3.38–42 and Laine, “Out of Character: Marginal Voices and Role-
Transcendence in the Mahābhārata’s Book of the Forest,” 286.
9 Alf Hiltebeitel, The Ritual of Battle: Krishna in the Mah ābh ārat a (Albany: State
University of New York Press, 1990 [1976]).
132 Bailey
Style
The Msp is typical of Sanskrit epic literature in using anuṣṭubh as its basic
meter, with triṣṭubh employed for emphasizing particular themes or for con-
cluding adhyāyas. There is also one chapter composed in prose (190.1–59)
interspersed with both anuṣṭubh and triṣṭubh, but there is nothing in its con-
tent suggestive of any reason why extra emphasis should be given by way of
metrical change. Use of anuṣṭubh provides a backdrop of constancy against
which other stylistic variation can be made.
These metrical forms constitute the minimal stylistic units, and the next
larger one is the adhyāya and the largest the parvan itself. Between these two
sizes of content, there is a third level, which I have called Narrative Unit10 in
a different literary context and it applies here as well. In the Msp the obvi-
ous Narrative Units are comprised of chapters 196–206 dealing with pātivratya
and the question of svadharma and the final fourteen chapters (207–21) deal-
ing with the related subjects of the lineages of fire and the birth of Skanda.
Arguably both of these two sections can be subdivided into two other sec-
tions and this is indicated clearly in the questions asked of Mārkaṇḍeya by
Yudhiṣṭhira and are bounded by appropriate question and answer. Question
and answer and introduction and conclusion do not necessarily correlate with
adhyāya beginnings and endings. But the fact that these formal and semi-for-
mal divisions exist does not mean there is no overarching semantic and literary
frame transcending them. I will return to this later.
The most basic stylistic distinction, as opposed to units of textual size, and
one which transcends bounded units of text is the well recognized difference
between didactic and narrative, utilizing texts differing in both content and
diction, where the former consists of descriptive statements in dharmaśāstric
style as opposed to the latter where a distinctive plot is essential. Both of course
can be interrelated, as one can be used to illustrate the content of the other.
Didactic teaching is sometimes used to advance an argument within the con-
text of a particular narrative plot, and/or to give emphasis to what has been
said in the narrative section by repeating it. Whilst usually quite straight
forward in its literary construction, its purported audience is not always so, and
10 See The Gaṇeśa Purāṇa, vol. 1. Upāsanākhanda (Wiesbaden: Otto Harrassowitz, 1995),
26ff. A Narrative Unit always encompasses several individual plots presented sequentially
and it can include didactic material as well. Its main distinguishing feature is an underly-
ing thematic relation between all of these. The best indigenous candidates for this narra-
tive type may well be the kathā and/or the itihāsa, both flexible enough to encompass one
or more myths and a combination of myths and didactic units.
Literary Structure of the Mārkaṇḍeyasamāsyāparvan 133
one wonders how much the obvious stylistic intertextuality of the didactic
material limits or broadens its audience. This stylistic difference is fundamen-
tal in the Mahābhārata11—less so in the Rāmāyaṇa—but it is still not well
understood and is derided as literature when it is really one of the fundamen-
tal distinguishing features of much Sanskrit literature.
Generic Forms
Sanskrit and Pāli literature by the time of the Mahābhārata already knows
of many different generic forms as outlined carefully already by Horsch.12
I do not want to enter the question of whether the Mahābhārata as a whole
is consistent with the sense of itihāsa, but will be content to say that many
old generic names commonly enough occur in the Mahābhārata and in the
Msp itself. That they occur in the Msp is not surprising, the important con-
sideration being how they combine to create a larger whole and what as
individual forms they might have contributed to the whole. A list of them illus-
trates the variety of names used to cover different forms of recitation: upamā13
(181.33), māhātmya (182.1; 196.2, 6; 206.32), gīta (184.1), caritam (185.1), purāṇam
… ākhyānam (185.53; similar is 189.14), kathā (186.11; 192.3), vicitra (188.7),
anubhūtam (191.26), ākhyānam (192.6; 195.40; 206.33), prakaraṇaṃ (196.21),
itihāsaṃ purātanam (207.6), and vaṃśaḥ (213.1). In addition, one chapter also
contains a significant number of subhāṣitas, even though that term is not
explicitly used.
Of these only māhātmya, carita, ākhyāna, kathā, prakaraṇa, itihāsa, and
vaṃśa can be regarded as terms indicative of formal genres. It is difficult to
know what to make of upamā, vicitra, and anubhūta, and gīta here simply
means what was recited by Sarasvatī—it is not a gītā in the more formal sense.
Of these four words upamā is well known in poetics as designating a simile, but
it cannot mean this here. I take the sentence where it occurs as meaning:
“There is this explanatory statement about this, best of speakers,”14 where the
context tells of whether men acquire things through fate, chance or their own
actions, and Mārkaṇḍeya goes on to illustrate this and expand it in verse form.
Van Buitenen (p. 576) translates upamā as parable, but it is really a statement
11 See also Adam Bowles, Dharma, Disorder and the Political in Ancient India. The Āpaddhar
maparvan of the Mahābhārata (Leiden: E.J. Brill, 2007).
12 P. Horsch, Die vedische Gathā-und Śloka-Literatur (Bern: Fancke, 1966).
13 This term is not mentioned in Horsch.
14 imām atropamāṃ cāpi nibodha vadatāṃ vara | (Mahābhārata 3.181.33ab)
134 Bailey
that fills out the details of an elliptical verse. Vicitrāṇi, on the other hand
seems, to refer to the tone of what Mārkaṇḍeya has been reciting and
Yudhiṣṭhira says, “Speak about this in detail as you are speaking about various
strange things.”15 Anubhūta, on the other hand, refers to what Mārkaṇḍeya
has directly experienced and is rather a qualification as to why his descriptions
carry so much weight. It occurs three times and always to past events:
1. “All that happened before occurred before your eyes, therefore we want
to hear your tale (kathā) complete with all the causes. Because it has
been experienced (anubhūtaṃ) by you alone many times, best of
brahmans. There is nothing ever unknown to you in all the
worlds”(186.11–12).16 Kathā is the generic form and anubhūta the reason
why he is able to tell the kathā.
2. “As such, these paths of transmigration in their multiplicity have been
seen and experienced (dṛṣṭāś caivānubhūtāś) by me who is long-lived
and I have told you of them”(189.15).17 Again both anubhūta and dṛṣṭa
occur together with kath as a verb, but the use of the two past passive
participles seems to be indicating that Mārkaṇḍeya has not just experi-
enced what he has seen but that his encounter with it has been personal
and possibly emotional.
3. “This is what I, long-lived, experienced and saw (mayānubhūtaṃ …
dṛṣṭam)”(196.26A),18 he said to the Pāṇḍavas, when speaking about
which creature might have been older than he.
Excluding these four, it is necessary to determine whether the others are used
randomly, and even if they are, whether they create an actual generic frame
providing guidance to a possible mode of interpretation by reader/hearers. If
this is so it must rest on the fact that these were already established historical
genres, implying a well-educated audience. Even for those who were not
sophisticated interpreters of these texts there are many clues given as to how
they should be read and in the case of the Msp these remain close to the actual
contents of the text. Yet there may still be a contrast between content of text
and generic type, even where there will be a constancy in use of meter, and
where the only real stylistic distinction to be drawn will be between didactic
and narrative. Of course, both of these do perform an educative function, and
an entertainment function, though differing considerably by scale of degree.
Much of the speech activity knitting together the individual chapters and
adhyāyas involves question and answer, and provides guidelines for how the
relevant passage should be understood. I have tried to trace them sequentially
through the Msp, and although it begins at chapter 179, Mārkaṇḍeya’s mono-
logue only really begins at the end of 180.43, where Kṛṣṇa speaks on behalf of
the brahmans and the Pāṇḍavas, Draupadī, Satyabhāmā and himself, request-
ing to hear some tales of the past. This gives rise to a more formal request
establishing the contents to be addressed: “Tell us the holy tales of the past
(purāvṛttāḥ kathāḥ puṇyāḥ)—which are about good conduct, which are
everlasting—of good kings, women, and seers, Mārkaṇḍeya” (180.43; Van
Buitenen, p. 574 modified).19 Whilst kings, sages, and women are present in all
components of the Msp, this categorization is obviously too broad really to
function as anything other than a general guideline. Moreover, what part of the
Mahābhārata, with the possible exception of the battle books, would lack
these three types of characters?
Nārada then arrives and in 181,1ff. Yudhiṣṭhira repeats the question, slightly
altering the topics: “The Pāṇḍava, king of the Kurus, observing that the great
hermit was willing to speak, urged him so that he might begin the tales: ‘Sir,
forever have you known the exploits of all the deities, Daityas, great-spirited
seers, and wise kings’”20 (181.1–2; Van Buitenen, p. 574). Two generic terms
kathā and carita are used, consistent with what one would expect of narratives
employed in telling about the categories of actors mentioned in verse 2. In
both cases Mārkaṇḍeya will be telling tales about the deeds/actions of gods,
demons, sages, and royal sages. But surprisingly, Yudhiṣṭhira narrows the nar-
rative direction by asking him to speak about karma and this is prefaced by no
formal generic category, though it does tie perfectly in with Kṛṣṇa’s own reflec-
tions (chapter 180) on the injustice of the preceding events. From verses 5–8
Yudhiṣṭhira asks in systematic but straightforward terms about the workings of
yoga, and then Mārkaṇḍeya, in a manner found often in this kind of literature
commends him for the quality of the question: “This extra question is truly fit-
ting for you, eloquent speaker. You know what must be known and you are
asking again to seek confirmation” (181.9).21 Both anupraśna and sthityartham
are not common in the Mahābhārata, but on the only other occasion where
sthityartham is found,22 it is the object of anu/pṛcch. I take it that Mārkaṇḍeya
knows that Yudhiṣṭhira has a cultivated knowledge of what he is asking for, but
is seeking for somebody with prodigious learning to confirm its validity. As
such I regard this as contributing to the discourse rather than functioning as
the introduction of a new generic category, though it would certainly be allow-
able to frame the response using a kathā or an ākhyāna, for example.
The remainder of this chapter embeds systematic homiletic material into
mythic narrative and vice versa. Verses 11–20 provide a brief, entirely descrip-
tive (there is no dialogue), narrative of a typical decline story where people
who were originally near perfect, become imperfect because of their sub-
jection to various desires. Mārkaṇḍeya ties this back to Yudhiṣṭhira’s initial
question when he says: “Kaunteya, a dead man’s course here is governed by
his own acts done here. Where does the treasure of acts stay of both the wise
and the foolish, and from whence does he recover his good or ill-done deeds?
That was what you were considering, now hear the response”(181.21–22c; Van
Buitenen, p. 575 modified).23 This is a dialogical marker also functioning as a
boundary marker between Mārkaṇḍeya’s narrative description and his more
homiletic presentation of the correct and incorrect understanding of karma
that follows (181.23–32), virtually in list form, and drawing on the opinion of
the wise and the learned to give it further justification. But this is not enough,
for Mārkaṇḍeya then proposes to give an upamā, which must be taken as an
This becomes a pivotal verse in allowing Sarasvatī to expatiate on what are the
correct forms of behavior that will lead to the acquisition of śreyas, but in some
ways it is quite conservative as it recommends performance of the agnihotra
and is utterly steeped in brāhmaṇical imagery. Yet it also foreshadows the pow-
erful theme of adherence to one’s own class dharma, something that will be
rehearsed often in the next eleven chapters. The fear of falling from one’s own
svadharma is raised again directly at 189.2025 by Yudhiṣṭhira after he has
already been given a full set of teachings on it, and it may be implied near the
end of the Msp when (209.12) Agni is said never to fall away in terms of renown,
brightness and splendor. It is all the more important because Sarasvatī, god-
dess of wisdom has given the teachings about svadharma in this chapter, thus
helping to build up a sense of authority already established by Mārkaṇḍeya
himself.
The chapter simply concludes with a statement that the worship of the gods
in the sacrifice is the padaṃ paramam (184.25), not a formal conclusion, but
fully appropriate to the context. But the next chapter appears to break the nar-
rative sequence completely, yet is still in the māhātmya since it has not yet
been formally completed. Yudhiṣthira simply asks to hear of the caritam of
Manu Vaivasvata (185.1). I take caritam as a formal genre because it gives spe-
cific guidance to the reader/hearer in respect of what might follow. Yet the
māhātmya is broken up by the caritam (also called purāṇa in verse 53) of Manu
(185) which is a creation myth detailing the recreation of the Earth after a huge
flood, where Brahmā (185.48) and Manu (49) are brought together as the agents
of the creation. Both figures imply a primacy of brāhmaṇical values such that
this myth can still form a coherent part of these chapters explicitly dealing
with the obligations kings must observe towards brahmans. But it also is
another explicit sign of Mārkaṇḍeya’s great antiquity and his authority as a
teacher. A phalaśruti concludes (185.54) this chapter, formally separating it
from what follows, but there is a clear thematic continuity, even repetition,
between 185 and 186–189 even if the content of the latter eventually shifts away
considerably from what is found in the former.
There is a formal ending here, naming (185.53) the source of the tale as relat-
ing to the fish,26 declaring it to be an ancient narrative (ākhyānam) and then
including a final verse which is a phalaśruti. In that sense this whole chapter is
bounded by a formal beginning and ending, fitting for a separate generic piece,
even one included in a larger generic space. This enables Yudhiṣṭhira to ask
once more (tataḥ sa punar evātha prapraccha; 186.1) what is a related question.
He reiterates over ten verses that Mārkaṇḍeya alone has the proper credentials
for describing the destruction and recreation of the Earth. All this is summed
up when he says: “All that happened before your eyes, therefore we want to
hear your tale covering the causes of everything (sarvahetvātmikāṃ kathām).
Because it has been experienced by you alone many times, best of brahmans.
There is nothing ever unknown to you in all the worlds”27 (186.11–12). A certain
uniqueness about Mārkaṇḍeya is stressed in Yudhiṣṭhira’s affirmation to him
when he says “You alone (tvam eva) attend upon Brahmā … (186.3),” “You alone
(tvam eva) see the beings being created … (186.4),” “You alone (tvam ekaḥ)
attend upon Brahmā” (186.10), and “Because it has been experienced by you
alone (tvayaikena) …” (186.12). Then follows a description of the decline in
behavioral standards during the yugas, a preface to the wonderful narrative of
the decline of the yugas and the pralaya, where Viṣṇu appears on the ocean as
a child. It is a long section of about 170 verses, crosses two chapters, and could
be described as autobiographical to the extent that Mārkaṇḍeya himself has
entered the narrated action, shown by the several conversations he has with
Nārāyaṇa as a child within a well-known initiatory frame. Chapter 187, lines
50–55 is a stotra and forms a concluding function to these two chapters. Though
they depict Mārkaṇḍeya in a condition beyond time, the last two verses of 187
bring him back to the present situation: “On seeing this tiger of the Vṛṣṇis this
26 I do not want to enter the vexed question of whether some kind of Matsyapurāṇa
(mātsyakaṃ nāma purāṇaṃ) is the subject of this reference.
27 etat pratyakṣataḥ sarvaṃ pūrvavṛttaṃ dvijottama |
tasmād icchāmahe śrotuṃ sarvahetvātmikāṃ kathām ||
anubhūtaṃ hi bahuśas tvayaikena dvijottama |
na te 'sty aviditaṃ kiṃ cit sarvalokeṣu nityadā ||.
140 Bailey
memory came to me about this primordial god, the unborn, Viṣṇu, the Man
who wears the yellow robe. He is the killer of Madhu, father and mother of all
beings. Bulls of the Kauravas, go to him for refuge as he is conducive to being
approached for refuge” (187.54–55).
Such a statement functions as an easily recognizable conclusion to the nar-
rated action, telling us that Māṛkaṇḍeya’s primordial encounter with Nārāyaṇa
works as an argument as to why the Pāṇḍavas must treat Kṛṣṇa as Viṣṇu. It is
both theological and devotional. It has its correlative response in 188.1–2 where
the Pāṇḍavas bow to Kṛṣṇa, who correctly acknowledges them in turn. But this
is a pretext for changing the direction of the narrative without the need for a
new generic category to be introduced. As it happens we have not formally left
the māhātmya of the brahmans, but there has been a radical shift in content
with the revelation of Kṛṣṇa’s true status. It continues in 188.3 where Yudhiṣṭhira
makes the general request of asking Mārkaṇḍeya about the future course of
the world under his sovereignty, where I take this as meaning Kṛṣṇa’s sover-
eignty. This gives rise to some very specific questions (188.4–7) about the
Kaliyuga and the transition to the Kṛtayuga, once more giving appropriate
guidance to what narrated action will be communicated. Beginning at 188.9, it
continues through until 189.15, with verses 14–15 functioning as a concluding
statement: “I have declared to you all that is past and future, as I remember the
Lore, lauded by the seers, that was promulgated by the Wind God. Long-lived
as I am, I have also witnessed and experienced the pathways of transmigration
and I have told them to you” (189.14–15, Van Buitenen, p.598).28 Van Buitenen
renders purāṇa as “Lore,” possibly implying a generic category, as it was spoken
by Vāyu. The use of ākhyātam as a verb here must also be significant given the
importance of ākhyāna as a generic type in the Mahābhārata and much earlier.
At the least we are presented with a definite conclusion to what has been a
lengthy selection of verses dealing with the four yugas. Then follows a brief
interlude dealing with dharma: “Unflappable man, you and your brothers must
hear still more of my words so that you will be free from doubts about the
Law.”29
There follows only a few verses about the benefits of observing the Law and
then Yudhiṣṭhira breaks in again, asking of his duties as a king: “In what Law
should I stand for protecting my subjects, sage? And how should I live
(vartamāno) and not fall away from my own Law?”30 (189.20). We might sug-
gest that this is a personal lesson to Yudhiṣṭḥira alone after the detailed
narrative of the yugas where dharma is dramatically overturned and confusion
reigns, but no doubt he has heard it all before. It evokes the theme of falling
away (na cyaveyaṃ) from his own Law, doubts about which pervade his whole
career. It is also a pep talk (21–27) for Yudhiṣṭhira, who duly acknowledges it
(28–29), before the discourse moves back to the primary interlocutor Vaiśaṃ
pāyana who describes the positive reaction to Mārkaṇḍeya’s speech, and then
informs us that they have heard a kathā: “Then after they had heard that auspi-
cious tale of the wise Mārkaṇḍeya, they were astonished at the communication
of the primordial tale”31 (189.31).
This sets the scene for a new subject to be introduced if so desired, and
Yudhiṣṭhira gives the lead, when he asks, “Please speak still further about the
exalted status of the brahmans (brāhmaṇamahābhāgyaṃ)” (190.1A). This
enables Mārkaṇḍeya to narrate the story of a king, Parikṣit who marries the
daughter of the king of frogs, and ends up in conflict with the king until the
latter resolves this conflict (190.42). Continuing straight on from this is a tale of
one of the descendents of Parikṣit and the frog-king’s daughter, a certain King
Śala who ends up in a conflict with a brahman sage, Vāmadeva, over some
horses he has borrowed from the brahman and refuses to return. Only after
some examples of the brahman’s psychic powers does he return the horses in
deference to this power, an example of the brahman’s mahābhāgya. Mārkaṇ
ḍeya finishes this off by simply stating that the king’s wife returned the horses
to the hermit and he, in turn, agreed to have a high regard for the king.
Now in 191.1 all the Pāṇḍavas change the subject dramatically by asking
Mārkaṇḍeya if anybody is older than him. His response involves a short
adhyāya which is really about a king Indrayumna who fell from heaven because
he had exhausted his good karma (kṣīṇapuṇyas tridivāt pracyutaḥ; 3.191.2B)
and how Mārkaṇḍeya had him sent back to heaven when a tortoise recognized
a good action he had done. Embedded in the middle are three ślokas in didac-
tic style urging people to good works, and as such taking us back to the themes
of Ch.182.
band, Lord.” (196.5cd–6ab)35 This is the pretext, which continues until verse 13,
for a lengthy telling of the famous tale about an irascible brahman named
Kauśika who ends up being given a detailed teaching about dharma from a
woman and then from a pious seller of meat who had been a brahman in a
previous life. Mārkaṇḍeya only begins telling it in chapter 197, taking another
seven verses to confirm the difficulty of the problem Yudhiṣṭhira has raised
and finishing with a verse seemingly indicating to Yudhiṣṭhira that he is deal-
ing with a formal shift in subject: “In respect of this subject (prakaraṇaṃ) King
Yudhiṣṭhira, concentrate and hear the constant law of wives who are loyal to
their husbands” (196.21). Prakaraṇam comes to mean “chapter, treatise” in later
Sanskrit literature, but I would not want to take it in this mode here, rather it is
simply a precise re-affirmation of the subject that is to be addressed.
This lengthy narrative finishes at the end of 206 and each of the adhyāyas
simply breaks up in the middle of a series of sequential actions or where a
particular didactic item has come to its logical conclusion. In all of the chap-
ters following 197 the brahman and the seller of meat engage in a dialogue
where the brahman asks questions pushing along the narrative. Yet another
interlocutory level is created, but there are constant reminders throughout
these chapters that Yudhiṣṭhira and Mārkaṇḍeya are still the primary episodic
interlocutors. Each chapter concludes with an explanation, leading onto the
development of a new subject in the next chapter, but all relate to the problem
of the interpretation of svadharma. Mārkaṇḍeya summarizes the whole thing
at the end in six verses, of which the first two bring to a close the dialogue
between the brahman and the seller of meat and the final three close off this
section of the Msp:
“All this has been narrated to you in its entirety, Yudhiṣṭhira, that which
you asked concerning the Law, son, best of the upholders of the Law, and
the greatness of the loyal wife and of the brahman, excellent man, and
of the obedience towards the mother and father and the Law as pro-
claimed in the figure of the hunter.”
Yudhiṣṭhira said,
“O brahman, this extraordinary, superb, narrative about the Law
(dharmākhyānam), was narrated by you, best of the upholders of all the
laws, excellent brahman.” (206.31–33)36
And he goes on to say that “I am still not satiated, illustrious man, hearing
about the supreme Law” (206.34cd).37 Which may imply that the following,
and final, long narrative will be about the Law as well.
This it may be, yet it is not explicitly so. The first verse (207.1ab)38 simply
reiterates what is said at the end of 206, alternating kathā for ākhyāna. Perhaps
implying a connecting link to the previous chapter, Yudhiṣṭhira then goes on to
ask his questions. These cover two elements: why did Agni go the forest and
when he had gone why did Aṅgiras carry the sacrificial offerings to the gods,
with the subquestion of why, when Agni is one, does he become many in dif-
ferent rites? The second question relates to the birth of Skanda from the
Kṛttikās. Both function in the manner of an anukramaṇikā, not dissimilar to
3.182.2, the verses foreshadowing the broad topics of the Msp, but much more
specific. In the remainder of chapter 207 Mārkaṇḍeya relates why Agni disap-
peared into the forest and how Aṅgiras was required to take over his role, and
in the last verse spells out his intention precisely: “I am going to declare the
various kinds of very splendid fires as they are named in the Brāḥmaṇas
according to the many sacrifices and in their multiplicity” (207.20).39 The sub-
sequent six chapters describe these fires in terms of vaṃśa as a generic form, a
series of lists with specific elaborations. There is a formal conclusion at the end
of chapter 212 to these vaṃśa chapters: “Such is the very great lineage (vaṃśaḥ)
of fires which I have recited (kīrtito). When purified by different mantras it car-
ries the offering of people.” Mārkaṇḍeya said, “The various lineages of fires was
recited by me, faultless man. But hear, descendant of Kuru, about the birth of
the wise Kārttikeya” (212.30–213.1; Van Buitenen, p. 646).40 Though this kind of
repetition occurs elsewhere in the Msp, here it seems designed to precisely
separate the two components of these fifteen chapters. There is also a system-
atic structuring of the six chapters of 207–212 where specific chapters are
organized in terms of particular lineages deriving from particular men and two
women.
For the next three chapters the birth of Skanda is narrated with the plot
sequence proceeding smoothly across these chapters. And finally at the end of
221 there occurs a phalaśruti signalling a formal ending to this section: “Who,
when fully concentrated, cites this birth of Skanda, attains prosperity here and
then goes to Skanda’s world” (221.80). This story of Skanda extends for nine
chapters and uses Skanda’s birth, appointment to the position of general of the
gods and his defeat of Mahiṣa as the armature crossing the entire nine chap-
ters, providing them with thematic cohesion. But within the individual
chapters it is arguable that there are also some more instances of vaṃśa to be
found, associated with Skanda but also potentially independent of him as well.
Yet they are not vaṃśas in the vertical sense, rather they are descriptions of
Skanda’s companions (217) and his six mothers who become snatchers of foe-
tuses (219) and such are classificatory in terms of present time rather than of
temporal sequentiality.
and answers. Biardeau, I suspect is correct in saying, “Il ne semble pas possible
d’aligner ces chapitres (depuis III, 182) comme s’ils étaient simplement juxta-
poses au hazard des questions suscitées par la curiosité de Yudhiṣṭhira.”42 But
what are the thematic principles of alignment?
To begin with it is necessary to return to the anukramaṇikā-type listing at
180.43: “Tell us the holy tales of the past—which are about good conduct,
which are everlasting—of good kings, women and seers, Mārkaṇḍeya.” As so
often in early Sanskrit literature the past is definitely a model for the present
and Kṛṣṇa is asking that Mārkaṇḍeya tell them of good conduct authenticated
because it is in some measure beyond time in the sense that it is eternal, and
classified by means of three fundamental categories of actors: kings, women,
and sages. This good conduct is directed towards the Pāṇḍavas and especially
Yudhiṣṭhira, the main recipient of the teachings, in part because he is such a
fragmented character. Besides the three categories of characters listed here,
the kathās have three qualities attached to them: they refer to what has hap-
pened in the past (purāvṛttāḥ), they are concerned with communicating good
conduct (sadācārāḥ)43 and they are everlasting (sanātanāḥ). I take “good con-
duct” as being the dominant descriptor here because so much that follows in
the Msp is about ācāra, a concept relating to dharma, and though denoting a
very broad sense of conduct, correlates directly with the dilemma in which
Yudhiṣṭhira always finds himself: to act or not to act. The word ācāra occurs
thirty-two times in the Msp, overwhelmingly in chapter 198 in the discussion
between the seller of meat and the irascible brahman, yet it is definitely a
pivotal sub-text of the Msp.
What is significant here because of its absence is the category of the brah-
man. The status of the brahman and his responsibility in regard to dharma is
the preoccupation of about two thirds of the Msp, yet it is not mentioned in
Yudhiṣṭhira’s question, unless the category ṛṣis is meant to include brahmans.
I am uncertain why it should be omitted from his question, unless the brah-
man is regarded as a marked category in the Mahābhārata and so requires no
extra mention in its different parvans. If we accept as a valid measure words
designating brahmans a statistical count shows that they are by far the most
is wrong, and eventually Gautama insults him when he accuses him of being
deluded. It is the eternal sage Sanatkumāra who finally breaks the deadlock
and the king, who is several times associated with dharma in no specific sense,
accepts Atri’s opinion, bolstered by Sanatkumāra’s as to how a brahman and a
king should relate. That is, though he is a rājarṣi and dharmārthasaṃyuktaḥ
(183.6), and someone of whom the sages say: tvad anyo nāsti dharmavit (183.11),
he accepts the judgements of the brahmans in spite of these sterling qua-
lifications.
This chapter is pivotal in asserting three themes that will be thoroughly
explored over the next twenty-three chapters: firstly, the brahmans’ disagree-
ments over the nature of dharma and, more specifically, of svadharma;
secondly, the role of the brahman as interpreter of dharma and the king as its
enforcer; thirdly, the role of low status people such as women in making con-
tributions to the proper understanding of dharma. Collectively they question
how sound are the received understandings of dharma, especially of svad-
harma, in the manner these have been taken up in the Mahābhārata, which
attempts in so many places to throw light on the real difficulties involved in
giving dharma a clear epistemological and pragmatic foundation. In this sense
the Msp is being directed at both royalty and the brahman class, the latter of
which had most to gain from a universalistic notion of dharma, but most
responsibility in disseminating it.
Chapter 184 consolidates the view that the brahmans’ status is based on the
Vedas and the sacrifice, perhaps skirting over the vexed issue of interpreting
the more complex understanding of dharma that only emerges in the Mahāb
hārata and post-Vedic literature. But then the narrative takes us into the myth
(185) of Manu and the fish and and onto chapters 186–189, the famous yugakṣaya
narratives where twice in 186 and 188 there is given a detailed description of
the Kaliyuga and of its transformation into the Kṛtayuga. These have been
well studied already by González-Reimann45 so I need not go into them in any
detail. It is the themes of the pralaya and Mārkaṇḍeya’s transformation into a
devotee of Viṣṇu/Nārāyaṇa that dominate these chapters, yet, given that the
45 L. González-Reimann, The Mahābhārata and the Yugas: India’s Great Epic Poem and the
Hindu System of World Ages (New York: Peter Lang, 2002). Writing of inconsistencies he
finds in the yugakṣaya chapters he says, “Biardeau seems willing to pass over these
and other contradictions for the sake of an elusive unity in the text, a unity that requires
considering all chronological references to a larger mythological plot based on the
need for the periodic appearance of Viṣṇu’s avatāras.” Ibid., 96. But the point for this
article is that even if there are inconsistencies, as there seem to be, how do they fit
together?
Literary Structure of the Mārkaṇḍeyasamāsyāparvan 151
former is a brahman sage and the latter represents kṣatriya values, they still
fall within the range of the larger context of the māhātmya of the brāhmaṇa.
The lengthy narratives that follow are as much about Mārkaṇḍeya himself as
about the cosmogony and the transformation of the world during the yugas.
When he goes into detail there is much directly impinging on the status of
brahmans in the “new” society, but told in a negative sense as an overthrow of
brāhmaṇical values (186.25–28, 31, 33) and the mixing up of the classes in such
a manner as to prevent the distinctiveness of each from being apparent. Kings
(38) will refuse to uphold the status of brahmans, and brahmans will become
false beggars (39), and so it goes on with a consistent statement of the reversal
of normative relations, demonstrating a relationship between kings and brah-
mans quite the opposite of what is laid down in chapter 183 and a complete
reversal of dharma, but not anarchy.
From 186.55–92 we are given a description of the pralaya and of Mārkaṇḍeya’s
vision of the amazing child sitting on the banyan tree, and his subsequent
entry into his mouth. In 99–100 a description is given of the four classes
engaged in their appropriate duties, a deliberate and sharp contrast with what
is seen in the earlier verses where their status is completely ignored. Then for
another fifty-one verses (186.100–129 and 187.1–22) brahmans are not men-
tioned, the content being taken up with a devotional dialogue. Then the
yugakṣaya theme resumes from 188.14 and continues in a famous section,
where brahmans are specifically referred to in verses 26, 41, 57–58, 60–61, 63,
64, 69–70, suggesting the declining socio-economic conditions in the world are
essentially a reversal of the brāhmaṇical order held to be normative if not
achievable. And this is reinforced in verses 85ff. describing the regeneration of
the world from the brahmans onward (dvijātipūrvako lokaḥ), and goes onto
describe how a brahman named Kalki Viṣṇuyaśas:
He will establish the good limits of behavior as laid down by the Self-
born, and renowned and meritorious in his deeds he will take refuge in
the forest in his old age. (189.16ce–18).
Brahmans are at the center of the cultural reconstruction that occurs at the
beginning of the Kṛtayuga and this is the normative model, which is the
obverse of the greater social and cultural complexity to which the composers
and disseminators of the texts must have been constantly exposed. It is against
a background of local power based in part on possession of economic resources,
where class status was not necessarily an essential factor.
If the yugakṣaya chapters stress again the tight cooperation between the
kṣatriyas and the brahmans, the devotional relationship developed between
Mārkaṇḍeya and Nārāyana could also symbolize this if it is construed as an
allusion to the avatāra doctrine. Biardeau suggests that when Mārkaṇḍeya sees
the world inside of Nārāyaṇa’s body, “Bref, tout est en ordre, tout est prêt pour
une nouvelle naissance. Le monde est guéri de son adharma par l’habitation
bien réglée en Nārāyaṇa.”46 And she adds later that “Nārāyaṇa s’identifie suc-
cessivement aux différentes formes qu’il a prises dans les myths où il est
question du sort de la terre …”47 This is most obvious in the brief narrative of
Kalkin, but Biardeau also finds avatāric activity in the tale of Utaṅka and
Dhundhumāra (192–95).48
But once it is accepted that limits of behavior49 have to be established, the
details of these limits must be defined in some detail and the mode by which
they can be interpreted. Because in spite of, or perhaps because of, the vague-
ness of the normative role of the brahmans it was felt necessary to expose the
limits of behavior and the difficulties in observing it in detail. The point is that
the didacticism of the homiletic passages is too abstract to make sense to any-
one who would continually witness the fluidity of social relations at village
levels, even more complex in large urban conurbations. This difference
between the homiletic abstractions and the much greater complexity of actual
social life is the key to understanding what will follow in the remaining chap-
ters dealing with brahmans. But there is also another important thematic
development coming into play from chapters 190–206. That is the inviolability
46 Biardeau, Le Mahābhārata, vol. 1, 603. See also Alf Hiltebeitel, Dharma: Its Early History in
Law, Religion, and Narrative (New York: Oxford University Press, 2011), 590–94.
47 Biardeau, Le Mahābhārata, vol. 1, 605.
48 Biardeau, Le Mahābhārata, vol. 1, 618–20.
49 Ācāra being the practical manifestation of behavior, dharma its theoretical underpin-
nings.
Literary Structure of the Mārkaṇḍeyasamāsyāparvan 153
of dharma as the foundation for the abstract system of behavior and its appli-
cation to real life situations.
This shift is somewhat suggested by Mārkaṇḍeya in 189 after he has described
the re-establishment of a world where normative boundaries are found. After
stating the sources for his just completed narration, he goes on to say: “Now
listen to these other words of mine, … which will set you free from doubts
concerning the Law. Always embrace the Law, chief upholder of the Law, for
a king who embodies the Law always enjoys happiness here and hereafter.
Listen to the auspicious message that I shall declare to you, blameless prince:
never insult a brahman, for an offended brahman can destroy the worlds”
(189.16ce–18; Van Buitenen, p. 598 modified).50 And this is where Yudhiṣṭhira
generalizes the question, if it can be further generalized, in saying: “In what Law
should I stand for protecting my subjects, sage? And how should I live and not
fall away from my own Law?” (189.20). The point is that even where the ques-
tions and potential answers can be general to the point of abstraction, they
will always be personalized because of the presence of the audience to whom
the answers are directed. This is especially highly profiled for Yudhiṣṭhira who
has already several times been exposed as being deeply troubled about the
whole question of dharma and of his own svadharma. And this returns us to
the theme of chapter 183 again where not even brahmans can agree upon the
correct dharma. But it must also be noted that given the intervening chapters
186–89, the whole thesis of the cooperation of the brahmans and the warriors
has been presented in reverse, where kings refuse to support brahmans as a
class in the correct manner. Hence Mārkaṇḍeya’s instructions to Yudhiṣṭhira,
a potential king.
Mārkaṇḍeya immediately gives an answer to this, but one characterized by
an annoying generality:
have conquered the entire Earth be joyful and happy. This is the Law of
the past and the future that has been proclaimed to you.51 (Van Buitenen,
p. 598)
But though dialogical in form this is mainly homiletic in style with minimum
entertainment value or illustration through actual case studies. It makes more
sense to present these arguments through actual examples and this conforms
to a long established principal that a precept will be illustrated by an example.
Accordingly it is a straightforward process to move directly from an “historical”
narrative illustrating the origins of dharma and the necessity for its application
at the hands of brahmans to several situations demonstrating how it might be
used in situations paralleling real world crises and possibilities. Thus follows a
series of illustrative examples, but they too are furnished with a semi-formal
introduction such as in the lengthy prose chapter 190 where Yudhiṣṭhira asks to
hear more about the exalted position (brāhmaṇamahābhāgyam) of the brah-
mans, but which leads to the story of the king who marries the frog princess
where no brahmans play any role at all. This leads into another myth where an
illustration is given of King Śala insulting a brahman sage named Vāmadeva by
stealing some of his magical horses and the negative consequences flowing
from this. The two narratives that occur here and the following killing of the
demon Dhundhu each represents a shifting attitude of a king towards a brah-
man. King Śala behaves initially how a king should not behave towards a
brahman, by stealing his property, whereas Kuvalāśva protects Utaṅka in the
manner appropriate for a king towards a brāhmaṇical sage.
But now the subject appears to change, though it really just deepens one
aspect of what has been raised earlier. Initially Yudhiṣṭhira asks about the
supreme greatness of women, and the perennial question of the subtlety of the
law.52 Both are foreshadowed at the beginning (180.43) of the Msp and they are
given more specificity when Yudhiṣṭhira asks: “To me the obedience of a
woman faithful to her husband seems extremely difficult, so you should tell me
It is the third part of the Msp that should attract our attention because it con-
tains almost no reference to Viṣṇu, yet is strongly theological, and appears not
What was the reason that the Fire once departed for the forest and that
the great seer Aṅgiras, in the absence of the Fire God, became fire and
carried the oblations? Though the Fire God is one he appears to be many
in the various rites, reverend sir. I wish to know all this. And also, how
Kumāra was born, how he became the son of the Fire God, how he was
born from Rudra by the Ganges and the Pleiades. (207.2–4; Van Buitenen,
p. 640 modified)56
55 Biardeau’s point at Le Mahābhārata, vol. 1, 659 is worthy of further exploration which will
the form the topic of another paper. “Une première réponse, la seule peut-être que nous
puissions donner avec assurance, c’est qu’en introduisant Skanda cet épisode introduit
aussi au premier plan Maheśvara et un foisonnement de déesses dont beaucoup sont
sanguinaires.”
56 katham agnir vanaṃ yātaḥ kathaṃ cāpy aṅgirāḥ purā |
naṣṭe ’gnau havyam avahad agnir bhūtvā mahān ṛṣiḥ ||
agnir yadā tv eka eva bahutvaṃ cāsya karmasu |
dṛśyate bhagavan sarvam etad icchāmi veditum ||
kumāraś ca yathotpanno yathā cāgneḥ suto 'bhavat |
yathā rudrāc ca saṃbhūto gaṅgāyāṃ kṛttikāsu ca ||.
Literary Structure of the Mārkaṇḍeyasamāsyāparvan 157
This implies prior knowledge of the events surrounding Agni, Aṅgiras, and
Kumāra, all of whom are associated in some way with fire and destruction. It is
behind this association with both of these motifs that Śiva himself is present in
his role as destroyer in the trimūrti formulation. Besides the theological/devo-
tional aspect of these chapters, especially chapters 221–22, it is rather a mystery
why Yudhiṣṭhira should be interested in these mythologemes, unless the birth
of Kārttikeya and his defeat of Mahiṣa is to be regarded as a framework within
which the conflict against the Kauravas might be carried through.
Initially (207) Agni goes, apparently without reason, to the forest to perform
tapas and Aṅgiras takes over the role of fire as dispeller of darkness and carrier
of oblations. Maybe here there is a conflict between fire used for others (Agni’s
traditional role) and fire used for oneself in the form of the performance of
tapas. Agni complains to Aṅgiras, who tells Agni to become fire again and that
he should make Aṅgiras his first-born son. The latter gives birth to Bṛhaspati
(207.17), Agni tells them the reason for this, leading Mārkaṇḍeya to list a gene-
alogy of fires and the functions to be performed by specific fires.
The principal myth in these final chapters of the Msp narrates the birth of
Kārttikeya and here it directly brings Agni into association with Śiva, thus
developing an ontological and functional picture of these deities. At the very
least the circumstances of Kārttikeya’s birth are complicated, but lead in chap-
ters 220–21 to a clear veneration of Śiva. Even before he is born the time when
his conception is foreshadowed by Indra is one in which the hour of Rudra is
present in the constellations (213.27–30). Eventually Svāhā, Dakṣa’s daughter,
who is in love with Agni, assumes the body of Śivā, Aṅgiras’s wife, in order to
attract Agni (214.1). It is hardly coincidental that her name is Śivā, even if she is
not Śiva’s wife. She and Agni make love and she assumes the appearance of six
of the wives of the seven great sages and six times57 casts down Agni’s seed.
Eventually Kārttikeya/Kumāra is born, but the ensuing chapters narrate
how the sages are concerned by the portents associated with his birth and how
Indra initially engages in physical conflict with him. Extensive description is
given of malevolent mothers, and the so-called Graspers who harass men after
their sixteenth year. Only in the two penultimate chapters (220–21) does the
text revert to Śiva and reflect the development of his image as a deity of devo-
tion. Even 219.45 tells us that “Now after an obeisance to Maheśvara, I shall
proclaim the Graspers that afflict men after their sixteenth year” and at 58cd:
“no Graspers touch those who are devoted to God Maheśvara.” In the next
chapter the tortuous details of Kārttikeya’s birth are firmly sheeted back to Śiva
and Umā.
Invincible are you in battle to any foe, like the lord who is Umā’s hus-
band. This deed shall be your first claim to fame and your glory will
be imperishable in the three worlds …. Rudra went on to Bhadravaṭa
and the celestials returned. The Gods had been told by Rudra to look
upon Skanda as himself … (221.75cd–76cd, 77cd–78ab; Van Buitenen,
p. 664)
Conclusion
It is almost a truism to say that like the Mahābhārata as a whole, the narrative
of the Msp is dominated by these themes: devotional theologies, cosmogony,
the difficulty in interpreting dharma, the excesses of kings and kṣatriya power,
the irascibility of sages, and brahman power. Unlike the entire Mahābhārata
there appears to exist no overarching narrative plot tying the whole lot togeth-
er.59 Yet the Msp does contain the usual combinatoric devices found in itihāsa
and Puṛāṇa: primary and secondary interlocutory systems, questions guiding
the direction of the immediate narrative and phalaśrutis. The obvious con-
necting point of the whole is established by the questions Yudhiṣṭhira puts to
Mārkaṇḍeya, suggesting that this parvan is a fundamental component in the
former’s education, to use Alf Hiltebeitel’s term.
Equally it is an education to the other Pāṇḍavas about Kṛṣṇa, Śiva and
the role of the brahman. Why the latter? A simple answer would be that it is
directed towards the hosts of brahmans who have accompanied the Pāṇḍavas
into the Kāmyaka forest. However, this by itself is an inadequate explanation.
In the figure of the cursing brahman named Kauśika (197.1) we are given the
image of a figure who is unsure of his own svadharma and it is only following
the intervention of a woman and a seller of meat that he is able to be given
the fundamental message that absolute adherence to one’s dharma, deter-
mined by birth, is the foundation for status in this life and the next. Clearly
Yudhiṣṭhira, with his constant vacillation is in the sights here, rehearsing a
theme found everywhere in the Mahābhārata, and virtually from beginning
to end. Repetition is used to make a point, a common rhetorical device in all
literature. But beneath this brahmans are speaking to each other in an attempt
to create a unity in dharma interpretation, an impossible task the Mahābhā
rata never stops trying to achieve.
References
Bailey, Gregory M., ed. The Gaṇeśa Purāṇa, vol. 1. Upāsanākhanda. Wiesbaden: Otto
Harrassowitz, 1995.
———. “The Semantics of Bhakti in the Vāmanapurāṇa.” Rivista degli Studi Orientali
LXII (1988): 25–57.
59 Though could anybody ever have held the entire epic in their head along the lines of the
Ṛgveda or another text of similar size? No doubt there were early attempts made to sum-
marize its contents in the manner found in some Purāṇas.
160 Bailey
Chapter 5
Among the more intriguing and yet largely unexamined points of differentia-
tion between the Rāmakathā as it is found in the Mahābhārata’s Rāmopākhyāna
and the Vālmīki Rāmāyaṇa is the history and representation of the rakṣovaṃśa,
or the lineage of the rākṣasas. The relative prominence that both epics give this
lineage is an indicator of its importance to the narrative. The differences
between the two histories of the rākṣasas are remarkable in a number of sig-
nificant ways, and an examination of them can shed new insights into the
relationship between the two epics and the socio-cultural worlds that they
reflect. This paper will compare these superficially parallel narratives of the
rākṣasa race with an eye toward identifying underlying internal narrative
structures that can aid us in understanding the narrative logics of the two pas-
sages in terms of their loci within their respective epics, their relationships to
their corresponding sister epics, as well as the needs and concerns of the epic
audiences.
A full one-third of the sargas of the Uttarakāṇḍa of the critical edition of
Vālmīki’s epic tale concerns itself with the history of the rākṣasa race.1 While
this section of the epic has been largely ignored, there have been, nevertheless,
a few, somewhat convincing, rationales put forth to explain and contextual-
ize this seemingly tangential narrative. The foremost among these argues
that this section of the Uttarakāṇḍa was originally an independent epic tale.2
Yet, this opening section of the kāṇḍa is more than a haphazard afterthought
on the part of redactors. The critical edition of the Uttarakāṇḍa devotes the
first thirty-four of its one hundred sargas, some one thousand two hundred
seventy-seven ślokas, “verses,” to the history of the rākṣasa race. This might
be compared, for example, with epic’s treatment of the vānara race found in
the Kiṣkindhākāṇḍa, the epic’s shortest kāṇḍa (sixty-six sargas) of which the
first thirty-eight sargas tell of the events in Kiṣkindhā and contain approxi·
1 All references are to the Critical Edition of the Vālmīki Rāmāyaṇa (The Vālmīki Rāmāyaṇa:
Critical Edition, 7 vols., ed. G.H. Bhatt and U.P. Shah [Baroda: Oriental Institute, 1960–75]).
2 Robert Antoine, Rama and the Bards: Epic Memory in the Ramayana (Calcutta: A Writers
Workshop Publication, 1975), 46–54; also cited in John L. Brockington, The Sanskrit Epics
(Leiden: E.J. Brill, 1998), 391–92.
mately twelve hundred ślokas. Moreover, the length of what Jacobi called “die
Rāvaṇeïs”3 is even more striking, when we understand that the average length
of the individual sargas that make up this narrative is approximately twice that
of the remaining sargas of the book, making this first section actually account
for over one-half of the entire Uttarakāṇḍa. Virtually all earlier studies of this
section of the kāṇḍa,4 whatever be their methodology, are primarily con-
cerned with the relative dating of the passage and what it can tell us about the
Uttarakāṇḍa’s relationship to the larger Rāmāyaṇa narrative.
Of all the scholarship on the Uttarakāṇḍa, in general, and the section that
concerns the history of the rākṣasa lineage, in particular, only Antoine’s5 looks
to the narrative structure as a mechanism through which embedded cul-
tural concerns might be understood historically and contextually. He focuses
on two specific instances in the epic, one in the Bālakāṇḍa and the other in
Uttarakāṇda, where the narrative seemingly pauses to relate extensive histo-
ries of two important characters, Viśvāmitra and Rāvaṇa, respectively. While
admitting that both of these narratives might have originally been “indepen-
dent epic songs,” he argues that their role in Vālmīki’s epic is not only concerned
“with the faithful transmission of the inner spirit of the epic” but additionally
offers “a very rich study in contrast which enables the modern reader to share
something of the vision which inspired the ancient bards and fascinated their
listeners.”6 Antoine is silent on whom he understands the “modern reader” to
be.
Antoine further understands that the two passages share a parallel con-
struction of genealogy, for both figures trace their origins to Prajāpati,7 and a
Antoine concludes that the two narratives set “before us two types of attitudes
to life, two different sets of values,”9 and he ultimately understands these atti-
tudes to reflect important aspects of larger narrative concerns. While in some
places he is perhaps a bit overzealous in his analysis of formulaic parallels that
are found throughout the epic and purāṇic narratives and in others he seems
to conveniently overlook narrative parallels (such as Rāvaṇa’s own austerities,
which gain for him the power through which he can challenge and defeat the
gods), Antoine has presented some intriguing arguments and should be given
credit for trying to understand the narrative in ways that allow a more careful
reading of cultural concerns both on the part of author and audience.
While Antoine focuses primarily on the digvijayas of Rāvaṇa, he expends
much of his initial energy suggesting parallels between the genealogies of both
Viśvāmitra and Rāvaṇa, primarily based on their common descent from
Prajāpati. However, while the narrative of Viśvāmitra includes some informa-
tion about his paternal line—Brahmā Prajāpati, Kuśa, Kuśanābha, and Gādhi
(1.50.18–19)—it occupies only two ślokas of the Bālakāṇḍa’s sarga 50.10 The
narrative in the Bālakāṇḍa is completely silent on Viśvāmitra’s maternal ances-
try. In stark contrast with this, sargas 2–9 of the Uttarakāṇḍa detail the history
of the lineage of Rāvaṇa, with sarga 2 primarily detailing his paternal lineage
and sarga 3 narrating the birth of his paternal half-brother, Kubera. Most strik-
ing, though, are the contents of sargas 4–9, which provide a detailed history of
the maternal line of Rāvaṇa.
8 Antoine, Rama and the Bards: Epic Memory in the Ramayana, 53.
9 Ibid., 70.
10 The Bālakāṇḍa, however, is not silent on these figures, but they are not included in what
Antoine understands as the “Viśvāmitra narrative.” See sargas 31–32, which relate the nar-
rative of the one hundred daughters of Kuśanābha. Although narrated by Viśvāmitra,
these sargas do not directly mention Viśvāmitra’s relationship to the lineage. See Sally J.
Sutherland Goldman, “Gendered Narratives: Gender, Space, and Narrative Structures in
Vālmīki’s Bālakāṇḍa,” in The Ramayana Revisited, ed. Mandakranta Bose (New York:
Oxford University Press, 2004), 47–85.
164 Sutherland Goldman
Is there a man in the world today who is truly virtuous? Who is there who
is mighty and yet knows both what is right and how to act upon it? Who
always speaks the truth and holds firmly to his vows?12
His name is Rāma and he was born in the House of Ikṣvāku. All men know
of him, for he is self-controlled, mighty, radiant, steadfast, and masterful.13
11 Ibid.
12 ko nv asmin sāmprataṃ loke guṇavān kaś ca vīryavān |
dharmajñaś ca kṛtajñaś ca satyavākyo dṛḍhavrataḥ || (Rāmāyaṇa 1.1.2)
13 ikṣvākuvaṃśaprabhavo rāmo nāma janaiḥ śrutaḥ |
niyatātmā mahāvīryo dyutimān dhṛtimān vaśī || (Rāmāyaṇa 1.1.8)
Of Daddies and Demons 165
them all to a pile of ashes. The story continues, telling of Aṃśumant, the son
of the wicked Asamañja,14 and his search for holy water in order to perform
the funerary libations for his half brothers. Aṃśumant is told by Garuḍa that
in order to perform them he needs water from the heavenly Ganges, which
must be brought to earth. We are then told of the descendants of Aṃśumant,
Dilīpa, and Bhagīratha, each of whom performs austerities in order to bring
the heavenly Ganges to earth. Eventually it is Bhagīratha who brings the holy
river to earth through his austerities (sarga 43). While the narrative is told in a
frame that provides a brief history of some of Rāma’s ancestors, it is primarily
a story that is concerned with the descent of the Ganges as is emphasized as
Viśvāmitra concludes his narrative:
The tale I have just told you, “The Descent of the Ganges,” brings one
wealth, fame, long life, heaven, and even sons.15
Only this selective genealogy of the epic’s hero, Rāma, told in the Bālakāṇḍa,
can be said to vie in length with that of Rāvaṇa’s narrated in the Uttarakāṇḍa.
Yet it takes up merely seven, relatively short sargas, containing but one hun-
dred seventy-two ślokas, and its focus is largely on the descent of the Ganges
rather than the lineage of Rāma.
It is only at sarga 69.15–32 that we find an actual listing of Rāma’s ancestors.
Rāma, upon breaking Śiva’s bow (1.66), has been given Sītā in marriage (1.67).
Daśaratha is then summoned to Mithilā to participate in the wedding cere-
mony. The ceremony is preceded by the recitation of the lineages of each
family, first Rāma’s, and then Sītā’s (1.70). Rāma’s lineage is recited by Vasiṣṭha,
while Sītā’s, is narrated by her father, Janaka. The names given for the Ikṣvāku
lineage are the nearly the same as those found later in the epic (2.102) and are
generally in line with those found in other contexts, for example, in Kālidāsa’s
Raghuvaṃśa, which, as the title indicates, supplies us with a detailed narrative
of the Sūryavaṃśa.16
14 Asamañja is introduced again into the narrative in the Ayodhyākāṇḍa, where Kaikeyī, in
her anger, likens Rāma to Asamañja (Rāmāyaṇa 2.32.12–16). He is also included in both
accounts of the lineage (1.69.25 and 2.102.20–21).
15 dhanyaṃ yaśasyam āyuṣyaṃ svargyaṃ putryam athāpi ca |
idam ākhyānam ākhyātaṃ gaṅgāvataraṇaṃ mayā || (Rāmāyaṇa 1.43.20)
16 The first nine sargas of the Raghuvaṃśa tell of Rāma’s four immediate ancestors: Dilīpa,
Raghu, Aja, and Daśaratha. Sargas 16 through 19 tell of the descendants of Rāma, starting
with his own sons, Kuśa and Lava, and ending with the dissolute Agnivarṇa, who, suffer-
ing from consumption (19.51), is secretly placed on the funeral pyre by his ministers
(19.54) and cremated. In the last verse of the poem we are told that his queen, pregnant,
166 Sutherland Goldman
is consecrated and rules awaiting the birth of her unnamed son (19.57). The Harivaṃśa,
too, knows this lineage (Harivāṃśa 8–10). See Simon Brodbeck, “Solar and Lunar Lines in
the Mahābhārata,” Religions of South Asia 5 no. 1 (2011): 128–30.
17 The story of Nimi is told at Uttarakāṇḍa Appendix I, No. 8, lines 83–212. However, the
lines in question have universal manuscript support and should have been admitted to
the critical text. See Sally J. Sutherland Goldman, “Critical Evidence: Vālmīki’s Uttarakāṇḍa
and the Critical Edition,” in The Churning of the Epics and Purāṇas at the 15th World San-
skrit Conference, ed. Simon Brodbeck, Alf Hiltebeitel, Adam Bowles (New Delhi: Rashtriya
Sanskrit Sansthan and D.K. Printworld, in press), 298–324 and Sally J. Sutherland Gold-
man, “Appendectomies: Textual Surgeries in the Construction of ‘The National Epic of
India,’” paper presented at the conference on New Directions in the Study of the Epics of
South and Southeast Asia, October 26–27, 2012, University of California at Berkeley. See
note 19.
18 The one possible exception to this might be the history of Śāntā, who becomes the wife of
Ṛśyaśṛṅga (1.8.11–23, especially verse 16). See, too, Asoke Chatterjee, “The Problem of
Śānta’s Parentage as Affecting the Text of the Rāmāyaṇa,” Our Heritage 2, no. 2 (1953):
353–74 and Asoke Chatterjee, “Śāntā’s Parentage,” Indian Historical Quarterly 33 (1957):
146–51. See note 19.
19 Yudhājit is mentioned some ten times in the Vālmīki Rāmāyaṇa (1.72.1; 2.1.2; 2.1.5; 2.64.22;
3.66.6; 3.66.8; 7.90.9; 7.91.1,3), while the name Aśvapati is mentioned six times (2.1.6;
2.9.16;2.38.4; 2.65.23; 2.68.9; 7.90.4). As suggested by M.R. Parameswaran (email to Robert
P. Goldman dated September 21, 2012), it appears that the only named kings invited to
Daśaratha’s sacrifice in the Bālakāṇḍa are those that have close family ties. Parameswaran
notes that at Rāmāyaṇa 1.12.19–20, the kings invited are:
(1) Janaka of Mithilā (1.12.18cd–19):
Of Daddies and Demons 167
are also told of the marriage contract between Kaikeyī’s father and Daśaratha
(2.99.3). Other family histories that include the mother’s lineage20 in the epic
corpus are rare for the most part and never detailed21 Yet, in the Uttarakāṇḍa,
the narrative of Rāvaṇa’s lineage takes up approximately twenty-seven percent
of the sargas making up the Rāvaṇa-carita, and it includes a significant num-
ber of stories that relate the family histories of Rāvaṇa’s mother’s family and
even includes some stories about the women specifically. Clearly such detail
must be understood as an indicator that both author and audience considered
Rāvaṇa’s maternal and paternal lineages to be important.
A further index of how important the lineage of Rāvaṇa is to these narra-
tives as well as how really remarkable this description of the histories of the
women’s families found in the Uttarakāṇḍa can be gleaned from the Mahābhā
rata’s much shorter version of the Rāmakathā, where we also find a genealogy
of the lineage of Rāvaṇa. The very existence of this lineage in the Rāmopākhyāna
has been used as evidence, I might note, to argue that the Mahābhārata was
aware of the Uttarakāṇḍa.22 There, however, the entire narrative of Rāvaṇa’s
lineage is pushed to the front of the episode, where we are initially given a brief
introduction to the Rāmakathā in four ślokas (Mahābhārata 3.258.1–5), fol-
terms patrilineal and matrilineal are understood as social structures and are defined as
relating to, based on, or tracing ancestral descent through the paternal and maternal
lines, respectively. The terms are also used of traditions wherein the father’s or mother’s
surnames are inherited from father to son, or mother to daughter and or wherein heredi-
tary succession or other inheritance is handed down through the father’s or mother’s line.
Brodbeck distinguishes the terms patriarchy and matriarchy from patriliny/patrilocy and
matriliny/matrilocy; the –archy suffix is associated with power; while the –liny associated
with a “lineal” descent, and the –locy with location of residence and understands the
Mahābhārata to be patrilineal and patrilocal, at least in theory (Simon Brodbeck, The
Mahābhārata Patriline: Gender, Culture, and the Royal Hereditary [Aldershot: Ashgate
Publishing, 2009], 14).
21 An exception would be, for example, the history of Śakuntalā’s family. Note that a simi-
larly disproportionate amount of attention is evident in the Rāmopākhyāna (259–75). The
narrative there is particularly interesting because the lineage and adventures of Rāvaṇa
are the only sections of the Uttarakāṇḍa narrated as part of the Rāmakathā in the
Āraṇyakaparvan (258.10–16; 259.1–13). Like many other aspects of the story, the events
found the Mahābhārata are much less detailed. See Brodbeck, The Mahābhārata Patriline,
57–63, 133–38. The epic does, however, provide a brief lineage of Sītā’s family at Bālakāṇḍa
70 and at Ayodhyākāṇḍa 110.26–52. The former is narrated by Janaka during the wedding
ceremony of Rāma and Sītā and the latter by Sītā, who tells Anasūya how she came to
have a svayaṃvara and wed Rāma. See the discussion below. The Uttarakāṇḍa’s App. I,
No. 8, lines 89–212 tell the history of Nimi, who is an ancestor of Janaka, Sītā’s father. See
note 17.
22 Brockington, The Sanskrit Epics, 398.
Of Daddies and Demons 169
lowed by an equally brief, and somewhat uninspired, lineage of Rāma and his
brothers in two plus ślokas:
There was a great king named Aja, born of the Ikṣvāku lineage.
His son, Daśaratha was virtuous and ever dedicated to the study of the
vedas.
He had four sons, learned in dharma and artha. They were Rāma, Lakṣ
maṇa, Śatrughna, and the mighty Bharata.
Rāma’s mother was Kausalyā, and Bharata’s Kaikeyī. Lakṣmaṇa and
Śatrughna, those scorchers of their enemies, were the sons of Sumitrā.23
Janaka was the king of Videha, and his daughter was Sītā, O lord, whom
Tvaṣṭṛ himself had created to be the beloved queen of Rāma.24
Mārkaṇḍeya, finished with the lineages of the hero and the heroine, turns to
the antagonist’s lineage.
This origin of Rāma has been declared to you, and of Sītā as well. I will
now relate to you the origin/birth of Rāvaṇa, O lord of men.25
Mārkaṇḍeya then narrates Rāvaṇa’s lineage, which occupies the next twenty
ślokas (Mahābhārata 3.258.10cd–16, and 3.259.1–13 [up until Rāvaṇa’s birth]).
The contrast is extraordinary. One, of course, might argue that, despite the
ignorance reflected in Yudhiṣṭhira’s question to Mārkaṇḍeya at 3.258.5 (kasmin
rāmaḥ kule jātaḥ kiṃvīryaḥ kimparākramaḥ, “In what lineage was Rāma born?
What was his valor? What was his strength?”) that both Rāma and his lineage
are well-known and need, as it were, no introduction, while Rāvaṇa does.
Nevertheless the almost excessive attention given to Rāvaṇa must still be seen
as unusual. Yet, even with all of this focus, the story in the Mahābhārata pro-
vides only a somewhat confused version of the patrilineal descent of Rāvaṇa
and does not concern itself even minimally with the maternal side of his fam-
ily. Mārkaṇḍeya now tells of the birth of Rāvaṇa:
The grandfather of Rāvaṇa was the god Prajāpati, himself, the self-exis-
tent, ascetic lord, and creator of the entire world.
His beloved, mind-born [mānasaḥ] son was named Pulastya. On a cow he
had an excellent son, named Vaiśravaṇa.
Having left his father, he approached his grandfather. And, O king, his
father from anger created [a second] self by means of himself.
Then a twice born, embodying that anger,26 was created named Viśravas,
in retaliation against Vaiśravaṇa.
But the grandfather [Prajāpati], pleased, gave to Vaiśravaṇa immortality,
lordship of wealth, and the status of being a protector of the world,
as well as friendship with Īśana, a son, Nalakūbara, and a capital city,
Laṅkā, inhabited by the rākṣasas.27
The lineage so far is notable in that only the males in the family are men-
tioned—the one exception being Pulastya’s somewhat unclear relationship
with a cow in verse 12, marked by the word gavi, “in or on a cow.” Prajāpati’s
begetting of Pulastya is indicated by the word mānasaḥ, “mental,” and in this
context “mind-born” is clearly the intent, while Pulastya’s creation of his sec-
ond self, named Viśravas, is said to be self-generated from anger (tasya kopāt
pitā rājan sasarjātmānam ātmanā, “His father created a [second] self by him-
self from anger” [13]). Finally, Vaiśravaṇa’s son Nalakūbara is given to him by
his grandfather Prajāpati along with his other rewards (15–16).
The relationship between Pulastya and Viśravas is not that of father and
son, but literally a second or half-self (ardhadehaḥ), although one hesitates to
use the term split personality, this seems to be the intent: Viśravas is Pulastya’s
“angry half.” This is made clear at 3.259.1–2ab:
And that muni, named Viśravas, who was a half-body [of Pulastya’s] from
the wrath of Pulastya, looked upon Vaiśravaṇa with rage.
That lord of the rākṣasas realized that his father [pitaram] was angry at
him.28
Thus, the term father (pitaram) here can only refer to Viśravas, the second-self
of Pulastya, while rākṣaseśvaraḥ must refer to Vaiśravaṇa, his son, who is also
known as Kubera.
The narrative in the Rāmopākhyāna reduces the lineage to a somewhat
incestuous situation between the two, Pulastya/Viśravas and Vaiśravaṇa. At
3.259.3cd, Kubera Vaiśravaṇa gives his father—again the word used is pitur
(“of/to the father”)—three rākṣasīs to wait upon him (rākṣasīḥ pradadau tisraḥ
pitur vai paricārikāḥ). And these women set about to soothe the angry sage
Viśravas:
Those women then undertook to please that great ṛṣi, O tiger among the
Bharatas, skilled as they were in singing and dancing.29
The three rākṣasī women, named Puṣpotkaṭā, Mālinī, and Rākā (259.5), com-
pete with one another, each desiring happiness (puṣpotkaṭā ca rākā ca mālinī
ca viśāṃ pate | anyonyaspardhayā rājañ śreyaskāmāḥ sumadhyamāḥ || 5).
Viśravas, pleased, grants them boons, that is, children (tāsāṃ sa bhagavāṃs
tuṣṭo mahātmā pradadau varān | 3.259.6ab) Puṣpotkaṭā gives birth to Rāvaṇa
and Kumbhakarṇa; Mālinī to Vibhīṣaṇa; and Rākā to Śūrpaṇakhā and Khara.30
The names of the women are significant and differ in both number and
identity from those found in the Uttarakāṇḍa of the Vālmīki Rāmāyaṇa. In
31 The name Rākā is marked as doubtful by the editors of the Critical Edition. Variants
include vākaḥ, bālā, and bakā. Rākā could well be the woman to whom Rāvaṇa is referring
here.
32 The term mālinī, the feminine of mālin, “wearing a garland or a florist,” ususally refers to
a woman who makes garlands or the wife of a man so employed. It is also the used as the
name of a meter, and is the name of various women. The –ini suffix is normally possessive
and is not typically used in connection with apatyavācaka, “genealogical,” derivations.
33 rājan sṛṣṭakaro brahmā pulastyas tatsuto ’bhavat |
tatas tu viśravā jajñe vedavidyāviśāradaḥ ||
tasya patnīdvayaṃ jātaṃ pativratyacaritrabhṛt |
ekā mandākinīnāmnī dvitīyā kaikasī smṛtā ||
pūrvasyāṃ dhanado jajñe lokapālavilāsadhṛk |
yo ’sau śivaprasādena lan̄ kāvāsam acīkarat ||
vidyunmālīsutāyāṃ tu putratrayam abhūn mahat |
Of Daddies and Demons 173
birth. Not insignificantly, the first narrative we have in the Uttarakāṇḍa con-
cerns Rāvaṇa’s patrilineal descent. That both the kāṇḍa and the history of
Rāvaṇa begin with such genealogical stories again suggests their importance to
both author and audience.
The narrative begins in the second sarga and tells of the birth of Viśravas,
the father of Rāvaṇa (7.2.4–29). Pulastya, a brahmarṣi and the mind-born son
of Brahmā Prajāpati, lives in an aśrāma near the ashram of the royal-seer
Tṛṇabindu on the slopes of Mt. Meru. But young girls—the daughters of the
gods and great serpents as well as the apsarases and the daughters of the royal-
seers—repeatedly come there to play, and in doing so distract Pulastya from
his austerities (7.2.4–7.2.9). Pulastya becomes furious and curses them (atha
ruṣṭo mahātejā vyājahāra mahāmuniḥ; 7.2.11ab): “Any young girl that comes
within my sight shall become pregnant.”41 Needless to say the girls stop com-
ing. But Tṛṇabindu’s daughter (who remains unnamed throughout the passage)
does not hear the sage’s pronouncement, and goes there, absolutely fearlessly
(tṛṇabindos tu rājarṣes tanayā na śṛṇoti tat | gatvāśramapadaṃ tasya vicacāra
sunirbhayā ||; 7.2.11). As soon as she hears the sound of his vedic recitations and
catches sight of the sage, she manifests the signs of pregnancy (14–15). Upon
her return home, Tṛṇabindu discovers her condition and asks: “How could you
have come to have so unseemly an appearance?”42 Tṛṇabindu’s daughter, still
ignorant of what has happened, tells her father that she went alone to the
heavenly ashram of Paulastya to look for her friends. Through meditation,
Tṛṇabindu discerns what had happened. He takes his daughter to the ashram
of Paulastya and offers her to him, to which Pulastya, who all along has desired
her, agrees (24). Pulastya is pleased with the already pregnant young girl, and
blesses her with a son named Viśravas. Viśravas will eventually become the
father of both Kubera and Rāvaṇa.
This story, like so many others, links the power of asceticism with that of
fertility and calls to mind the sexual potency of the male gaze. But the narra-
tive is not so straightforward. The phrase yā me darśanam āgacchet (7.2.10a) in
this context is somewhat ambiguous. The term me, “mine, of me” can be under-
stood as either the object or subject of the darśanam, “seeing.” Thus, one is
unable to tell whether the curse means: “[any young girl] who would see me
(lit., who would come to a seeing of me) or would come within my sight,” or
both. The term darśanam is significant as well, as it is commonly used in a
religious sense of seeing or perceiving the sacred. But the term can be under-
stood both actively and passively. Thus the devotee can gaze upon the deity, or
the deity might gaze upon the worshipper, or both actions can occur recipro-
cally.43 The context here is clearly sacral. The region is sacred, as it is here that
the religious activity of Pulastya is undertaken. We are told that the sage is
practicing austerities, engaged in vedic recitation, and of restrained sense
organs (tapas tepe … svādhyāyaniyatendriyaḥ; 7.2.7). Additionally, Tṛṇabindu’s
daughter is said to hear the vedic chanting (vedadhvaniṃ śrutvā; 7.2.14). While
the agent of the action at 7.2.10 is ambiguous, at 7.2.14 it is made clear. The
young girl is the visualizer of the sage (sā … dṛṣṭvā … tapodhanam [7.2.14]), and
her transgression is the very activity of gazing. This agency makes her both an
aggressor and the transgressor.44
Not unexpectedly, the narrative reflects the deeply ingrained cultural atti-
tudes toward young women and the ease with which they become threatening
sexual and sexualized beings.45 The narrative also highlights the virtual obses-
sion of the epics to avoid depicting its sages and other males as having physical
sexual intercourse as a means of procreation. Immediately outside of the pro-
tective world of her father and friends, Tṛṇabindu’s daughter, seeming unaware
of her sexual viability, nevertheless is marked as the transgressor and “suffers”
43 See Diana Eck, Darśan: Seeing the Devine Image in India, 3rd ed. (New York: Columbia
University Press, 1998), 3–7. The history and significance of the gaze in the early Indian
tradition is discussed extensively by Jan Gonda (The Eye and Gaze in the Veda. Verhan-
delingen der Koninklijke Nederlandse Akademie van Wetenschappen, Afd. Letterkunde,
Nieuwe reeks, deel 75, no. 1 [Amsterdam: North-Holland Publishing Company, 1969]).
44 Compare Mahābhārata 1.104.6–10, where Kuntī, having pleased Durvāsas, is given a man-
tra to the effect that whichever god she might summon would give her a son. Curious,
Kuntī, still a virgin, employs the mantra and, gazing upon the Sūrya, the sun god, whom
she has summoned, is impregnated. Note particularly that Kuntī is the agent of the seeing
(dadarśa sā; dṛṣṭvā, 9) and that the immediate result of that visualization is pregnancy
(tasyāṃ garbhaṃ dadhau tataḥ, 10ab).
45 Perhaps most influential and best known of the expressions of a woman’s position in
society is found in Manusmṛti, a text on dharmaśāstra, dating from around 200 CE, which
has codified attitudes that predate it by many centuries. Manu succinctly situates a wom-
an’s place in its patriarchal world:
bālye pitur vaśe tiṣṭhet pāṇigrāhasya yauvane |
putrāṇāṃ bhartari prete na bhajet strī svatantratām || 5.148.
In childhood she should be under her father’s control, in youth, in the control of her hus-
band, who has taken her hand. But when her husband has passed away, a woman should
be under her sons’ control. A woman should never enjoy independence.
178 Sutherland Goldman
For all those concerned about their honor, being a father of an unmarried
girl is a great trouble. For, dear daughter, one does not know who might
marry the girl.
A young girl will always remain a source of suspicion for three families—
that of the mother, that of the father, and that of the one to which she is
given.49
Kaikasī is hesitant, but, fearful of disobeying her father, she approaches the
sage even though it is at a bad time:
At that very time, the brahman, Pulastya’s son, who was like a fourth sac-
rificial fire, was performing an Agnihotra rite.
Ignoring that fearsome hour out of deference to her father, she approached
and stood before him, her gaze lowered to his feet.51
Viśravas, gazing upon her, questions her, although he, as an omniscient sage,
already knows the answers to his questions:
Looking at that fair-hipped woman, her face like the full moon, the highly
illustrious sage, blazing, as it were, with his vital energy, spoke:
“Whose daughter are you, my good woman? Where have you come from
and for what reason? What can I do for you? Tell me truthfully, lovely
lady.”52
Addressed in this fashion the young girl cupped her hands in reverence
and then said: “Through your own power, sage, please determine my
intention.
However, brahman, you must know that I have come here on my father’s
instructions and that I am known by the name Kaikasī. Please determine
the rest for yourself.”53
Having done so, Viśravas agrees to grant Kaikasī’s wish and gives her offspring,
but on one condition:
After having entered into meditation, the sage said these words: “My
good woman, I have discovered the purpose that you have in mind.
Viśravas tells her that, since she has come at a fearful hour (dāruṇāyāṃ tu
velāyām), her sons would be rākṣasas of cruel deeds. Kaikasī then implores
him not to give her only cruel sons, and he grants a small reprieve, the youngest
shall be righteous (7.9.17–20).
The last part of the story of the birth of Rāvaṇa and his siblings is fairly well-
known, but how Kaikasī is forced to find her own husband and how she comes
to conceive her children is less so. The theme of a young woman being forced
to find her own husband is found elsewhere,55 and the narrative parallels in a
number of ways the story of the birth of Viśravas: a young woman, who is a
potential [sexual] danger to her family, is sent or comes into the presence of a
powerful sage, who, engaged in ritual practices, is distracted from those prac-
tices by the very presence of the young woman. The result is that the young girl
is impregnated. In the story of the birth of Viśravas, the power of the male gaze
is explicit. In this story it is implicit, as the gaze is mediated by the mental con-
centration or focus—a mental gaze, as it were. Both stories tell of female
intrusion into a masculine space that is ritually sacred (marked by mental con-
centration and religious observations) and both lead to impregnation. We are
left with little doubt that the mere gaze or sight of a male is potent enough to
impregnate a female, and we are also reminded that no young woman is safe
outside of the immediate supervision of her family or husband. Once again,
the narrative clearly establishes the woman as the sexual aggressor and trans-
gressor, a theme repeated from the opening story.56
What is unique here is that these two clearly parallel narratives frame the
lineage of the rākṣasa Rāvaṇa, while providing both sides of the family with a
similar story of female ancestry. Rāvaṇa descends from noble, but clearly
flawed lineages. And at the heart of that flaw is sexual transgression, which is
located in the feminine. Here at Rāvaṇa’s birth we have the perfect storm, as it
were—sexual transgression, misogyny, and the demonic—wherein the demo
nic is located in and emerges from the sexualized female body. This alignment
is one repeated earlier in the epic, particularly in the Sundarakāṇḍa, and forms
a core element of the imagination of the Rāmāyaṇa poet[s].57
This understanding is further reinforced in the Uttarakāṇḍa by the history
of Rāvaṇa’s half brother Kubera, whose own history is also given at this point,
and serves as a foil, for he and Rāvaṇa share a father, Viśravas—whose own
mother, Tṛṇabindu’s unnamed daughter, it will be recalled, is sexually trans-
gressive, but not a rākṣasī. Kubera’s mother is Devavarṇinī, the daughter of the
sage Bharadvāja. We know little of her history except that she is given in mar-
riage and then conceives and bears a child (7.3.3–4). She is nowhere in the
Uttarakāṇḍa associated with transgressive behavior and as the daughter of a
sage, clearly lacks any rākṣasa blood. It comes as no surprise that her offspring
does not suffer a similar fate to Rāvaṇa’s.
Elsewhere in Vālmīki’s Rāmāyaṇa, sexual transgression and the demonic are
located on/in the body of the female.58 To see how this is an explicit concern
of Vālmīki, we can turn again to the Rāmopākhyāna, which, as I noted earlier,
removes the narrative from its gendered frame, and, although once mention-
ing the rākṣasa association of the three women given to Viśravas, virtually
silences any other information concerning them. We are only told what is
absolutely necessary to explain the birth of Rāvaṇa and his siblings:
Pleased with them, that blessed and magnanimous [sage] gave them
boons.59
The Rāmopākhyāna’s narrative does not concern itself with any sexually trans-
gressive behavior of the women nor does it construct the rākṣasa women
inherently as negative, at least any more so than other women (although, like
57 See note 46 above and Sally J. Sutherland Goldman, “Re-siting Sītā: Gender and Narrative
in Vālmīki's Sundarakāṇḍa,” Purāṇa 45, no. 2 (2003): 115–35; “Nikumbhilā’s Grove: Rākṣasa
Rites in Vālmīki’s Rāmāyaṇa,” in Proceedings of the XIIIth World Sanskrit Conference, Edin-
burgh: Epic Studies, ed. J. Brockington (Delhi: Motilal Banarsidas, 2009), 255–75; “Illusory
Evidence: The Construction of Māyā in Vālmīki’s Rāmāyaṇa,” in Epic and Argument in
Sanskrit Literary History: Essays in Honor of R.P. Goldman, ed. Sheldon Pollock (New Delhi:
Manohar, 2010), 209–33.
58 See Goldman, “Re-siting Sītā: Gender and Narrative in Vālmīki's Sundarakāṇḍa”; “Nikum
bhilā’s Grove: Rākṣasa Rites in Vālmīki’s Rāmāyaṇa.”
59 tāsāṃ sa bhagavaṃs tuṣṭo mahātmā pradadau varān | (Mahābhārata 3.259.6ab)
Of Daddies and Demons 183
References
Primary Sources
Agni-Purāṇa. (Electronic Text). Göttingen Register of Electronic Texts in Indian
Languages (GRETIL), Archive of E-Texts in Unicode (UTF-8). Based on the edition
by Rajendralal Mitra Calcutta: Asiatic Society of Bengal, 1870-79, 3 vols. (Bibliotheca
Indica, 65,1–3). Input by Jun TAKASHIMA Database copyright (C) Jun TAKASHIMA
2001. <http://gretil.sub.uni-goettingen.de/gretil/1_sanskr/3_purana/agp_bi_c.txt>.
Agnipurāṇa. 1879. Bibliotheca Indica Edition (65.1–3).
Brahmāṇḍa-Purāṇam, Part 1. (Electronic Text). Göttingen Register of Electronic Texts
in Indian Languages (GRETIL), Archive of E-Texts in Unicode (UTF-8). Based on the
edition Bombay: Venkatesvara Steam Press (or a reprint thereof). Input by members
of the SANSKNET-project (www.sansknet.org). <http://gretil.sub.uni-goettingen.de/
gretil/1_sanskr/3_purana/brndp1_c.txt>.
184 Sutherland Goldman
Secondary Sources
Antoine, Robert. Rama and the Bards: Epic Memory in the Ramayana. Calcutta: A Writers
Workshop Publication, 1975.
Brockington, John L. The Sanskrit Epics. Leiden: E.J. Brill, 1998.
Brodbeck, Simon. “Solar and Lunar Lines in the Mahābhārata.” Religions of South Asia
5, no. 1 (2011): 127–52.
Chatterjee, A.K. “A Note on the Uttarakāṇḍa of the Rāmāyaṇa.” Journal of the Oriental
Institute, Baroda 22, no. 3 (1972–73): 304–15.
Chatterjee, Asoke. “The Problem of Śānta’s Parentage as Affecting the Text of the
Rāmāyaṇa.” Our Heritage 2, no. 2 (1953): 353–74.
———. “Śāntā’s Parentage.” Indian Historical Quarterly 33 (1957): 146–51.
Eck, Diana. Darśan: Seeing the Divine Image in India. 3rd ed. New York: Columbia
University Press, 1998.
Goldman, Sally J. Sutherland. “Appendectomies: Textual Surgeries in the Construction
of ‘The National Epic of India.’” Paper presented at the conference on New Directions
in the Study of the Epics of South and Southeast Asia, October 26–27, 2012, University
of California at Berkeley.
———. “Critical Evidence: Vālmīki’s Uttarakāṇḍa and the Critical Edition.” In Proceed
ings of the 15th World Sanskrit Conference, edited by Adam Bowles, Simon Brodbeck,
and Alfred Hiltebeitel. New Delhi: Rashtriya Sanskrit Sansthan, forthcoming.
———. “Gendered Narratives: Gender, Space, and Narrative Structures in Vālmīki’s
Bālakāṇḍa.” In The Ramayana Revisited, edited by Mandakranta Bose, 47–85. New
York: Oxford University Press, 2004.
———. “Illusory Evidence: The Construction of Māyā in Vālmīki’s Rāmāyaṇa.” In Epic
and Argument in Sanskrit Literary History: Essays in Honor of R.P. Goldman, edited by
Sheldon Pollock, 209–33. New Delhi: Manorhar, 2010.
186 Sutherland Goldman
Chapter 6
1 I would like to thank those who attended the session at which I presented this paper during
the 41st Annual Conference on South Asia at the University of Wisconsin, Madison, in October,
2012. Particular thanks go to Vishwa Adluri, Greg Bailey, Simon Brodbeck, Robert Goldman,
James Laine, and Philip Lutgendorf for comments after the presentation, to Vishwa Adluri for
organizing the event, and to Alf Hiltebeitel both for comments on an earlier version of this
paper, and for inspiring me to approach the Mahābhārata in a new way. All citations of the
Mahābhārata are to the Critical Edition: Sukthankar, V.S., et al. (eds.). The Mahābhārata: For
the First Time Critically Edited, 19 vols. (Poona: Bhandarkar Oriental Research Institute, 1933–
59). It is conventional to use as his name either Hanūmān or Hanumān, and as noted by John
Brockington (p. 134, note 1), the spelling with long u is more frequent in the Mahābhārata: see
John Brockington, “Hanumān in the Mahābhārata,” Journal of Vaishnava Studies 12, no. 2
(2004): 129–35. However, for the sake of consistency with other names cited in the stem form,
I will use Hanūmat in this article.
2 Hiltebeitel lists (on pp. 467–69 [pp. 23–24 in this volume]) the 67 episodes the Mahābhārata
names as upākhyāna, which he translates as “subtales.” He comments, “I will favor the transla-
tion ‘‘subtale’’ for upākhyāna, with perhaps a hint of subtext.” See Alf Hiltebeitel, “Not Without
Subtales: Telling Laws and Truths in the Sanskrit Epics,” Journal of Indian Philosophy 33 (2005):
455–511; 455, note 2.
working as a committee over a generation or two between 150 bce and the year
1.3 He adds that, “To speak of the temporal priority of the Mahābhārata over
the Rāmāyaṇa and Manu is thus not to rule out the possibility that the last two
might have been started before the Mahābhārata was finished.”4 Hiltebeitel
has also maintained that the Mahābhārata provided the pattern for the
Rāmāyaṇa, and I quote: “… the Rāmāyaṇa poet is familiar with the Mahābhā
rata’s archetypal design and intent upon refining it.”5 Hiltebeitel’s view is that
Vālmīki composed the Rāmāyaṇa by refining the Mahābhārata’s (already exist-
ing) mode of expression into kāvya, refining also its structure and style by
reducing and incorporating into the main story the subtales for which the
Mahābhārata is famous.
Some have assumed or argued that the size of the Mahābhārata, the sheer
number of words, is evidence of a long compositional history—it would take
a long time to write so much, perhaps 400 or 800 years. Some have assumed
or argued that the diversity of literary styles and religious ideas would require
hundreds of years to be composed.6 These, however, are assumptions rather
than persuasive arguments. As a counter-example, I can cite Isaac Asimov,
who wrote some 500 books—on popular science, history, chess, and science
fiction—while also serving as professor of biochemistry at Boston University.
I mention him not only because of the number of his many works, but also
their diversity: he published in all ten major categories of the Dewey Decimal
System, so he was not simply producing quickly written romance novels.
Indeed, as another example, Alf Hiltebeitel has written perhaps as much about
3 See Alf Hiltebeitel, Dharma: Its Early History in Law, Religion, and Narrative (Oxford: Oxford
University Press, 2011), 190–91; see also 11, and his Rethinking the Mahābhārata: A Reader’s
Guide to the Education of the Dharma King (Chicago: The University of Chicago Press, 2001),
18–20.
4 Hiltebeitel, Dharma, 200.
5 Ibid., 413.
6 See, for example, E. Washburn Hopkins, The Great Epic of India: Its Character and Origin (New
York: Charles Scribner’s Sons, 1902), especially 396–400, in which he imagines a work prior to
the Mahābhārata that did not feature the Pāṇḍavas, with growth and “intrusions of didactic
matter” (398) in stages from 400 bce to 400 CE, without much evidence to support such a
chronology. An even more elaborate pattern of development in stages is proposed by James
L. Fitzgerald in “Mahābhārata,” in Brill's Encyclopedia of Hinduism, vol. 2, ed. Knut A. Jacobsen,
Helene Basu, Angelika Malinar, and Vasudha Narayanan (Leiden: E.J. Brill, 2010). A similar but
less elaborated view of its development over time can be seen in J.A.B. van Buitenen, The Ma
hābhārata, vol. 1 (Chicago: The University of Chicago Press, 1973), xxiii–xxv; see also his com-
ment (ibid., xvi): “Whatever historical realities may also have been woven into the epic, it is
not an accident of dynastic history; however fortuitous its career of expansion, the epic is not
an accident of literary history. The grand framework was a design.”
The Tale of an Old Monkey and a Fragrant Flower 189
I will now summarize the passage in which Bhīma meets Hanūmat. Draupadī
sees a remarkable flower brought by the wind and asks Bhīma to fetch more.8
He climbs the mountain Gandhamādana (“intoxicating with fragrance”), along
the way destroying a banana grove. Then he encounters a huge monkey, who
yawns and slaps his tail on the ground, the sound echoing all around (the text
tells us that his name is Hanūmat). Bhīma roars defiantly, and Hanūmat asks
why he has been awakened when he is ill. He states that while animals do not
know dharma, humans endowed with reason should have compassion for all
creatures, but Bhīma has destroyed dharma by harming the forest and its crea-
tures. He advises Bhīma to turn back (146).
Bhīma identifies himself as a kṣatriya named Bhīmasena Pāṇḍava, and a
son of Vāyu. The text identifies Hanūmat as another son of the Wind, and he
again advises Bhīma to turn back—he will not make way for Bhīma. When
Bhīma insists, Hanūmat says that he does not have the strength to rise, so
Bhīma should just jump over him. Bhīma declines the offer so as not to humble
him, while asserting that he could have jumped over both the monkey and
the mountain, like Hanūmat jumped over the ocean. Hanūmat asks for details,
and Bhīma replies that Hanūmat is his brother, famous from the Rāmāyaṇa,
7 Ādhyāya colophons in the manuscripts consulted for the Critical Edition have a variety of
titles, including both long and short u, the word samāgama (“meeting”) replaced by saṃvāda
(“discussion”), and some that omit reference to Bhīma: Hanumad-Vākyaṃ (“Hanūmat’s
Speech”) and Hanūmad-Darśanaṃ (“The Vision of Hanūmat”). I call this passage “Hanū
mād-bhīma-samāgama” because of the prevalence of this title in devanāgarī and Kāśmīrī
manuscripts.
8 In stating that the flower is “floating down a river” (p. 180), Van Buitenen is wrong—it is wind-
borne, appropriately for an episode featuring two sons of Vāyu. Nonetheless, there are reasons
to see the Samāgama and “Five Indras” episodes as linked to one another, as he indicates and
as I discuss below. See J.A.B. van Buitenen, trans. The Mahābhārata, vol. 2. (Chicago: The
University of Chicago Press, 1975).
190 Sullivan
Hanūmat’s role in which Bhīma summarizes in one sentence: “For the sake of
Rāma’s wife, that Indra of monkeys jumped the 100-yojana sea in a single leap”
(Mahābhārata 3.147.12). Bhīma claims to equal Hanūmat in strength and in bat-
tle, so he demands that the monkey move or Bhīma will defeat him. Hanūmat,
regarding Bhīma as overly impressed with his own strength, asks that Bhīma
take pity on him and just move his tail to pass by. Unable move the tail at all,
Bhīma asks to be forgiven, and asks the identity of the monkey: is he a Siddha, a
Deva, a Gandharva, or a Guhyaka? Hanūmat identifies himself by name for the
first time, and summarizes his interaction with Rāma, whom he calls “Viṣṇu in
human form” (viṣṇur mānuṣarūpeṇa; 3.147.28). Hanūmat recounts that after
Rāma had recovered his wife, Hanūmat asked, “May I live as long as the story of
Rāma (rāmakathā; 3.147.37) lives on in the worlds,” which Rāma granted.
Bhīma bows to Hanūmat and asks to see the form in which Hanūmat leaped
the ocean. Hanūmat replies that no one can see that form from another yuga,
as all creatures comply with the time in which they live. Bhīma asks for knowl-
edge of the yugas and Hanūmat teaches him the characteristics of each of the
four eons in detail (Mahābhārata 3.148). Bhīma refuses to go until he sees
Hanūmat’s old form, so Hanūmat, in order to grant his brother a favor, grows to
a mountainous size which both delights and surprises Bhīma. When Bhīma
closes his eyes, Hanūmat announces that this is all Bhīma can stand to see,
though he could grow as large as he wants. Bhīma asks that Hanūmat reduce
his form, and wonders why Rāma didn’t have Hanūmat alone defeat Rāvaṇa, to
which Hanūmat replies that he did not want to detract from Rāma’s glory.
Hanūmat teaches Bhīma about dharma: he advocates worship of the gods with
offerings and devotion (bhaktyā; 3.149.24), and urges Bhīma to adhere to his
svadharma, namely kṣatriya-dharma (37), then goes on to describe a king’s
duties.
Only then does Hanūmat resume his smaller form; he embraces Bhīma, tell-
ing him that he reminds Hanūmat of Rāma. He offers Bhīma a boon, and
Bhīma says that with Hanūmat as protector the Pāṇḍavas will conquer all ene-
mies. Hanūmat offers to do a kindness: he will add his roar to Bhīma’s on the
battlefield while stationed on Arjuna’s flagstaff. With that, Hanūmat disap-
pears (antaradhīyata; 3.150.15), asking Bhīma to keep his location secret. The
remainder of the story features Bhīma completing his search for the flowers by
killing a contingent of Yakṣas and Rākṣasas who had opposed his flower-gath-
ering, after which Draupadī and the Pāṇḍava brothers (other than Arjuna)
were reunited with Bhīma, and Draupadī’s desire for these flowers was fulfilled
(153).
The Tale of an Old Monkey and a Fragrant Flower 191
The Analysis
9 Quoting the translation of Van Buitenen, Mahābhārata, vol. 2, 500: navāvatāraṁ rūpasya
(Mahābhārata 3.146.33). Sutton is thus incorrect when he states that the Mahābhārata
“has a more limited understanding of the concept and in fact never once in the Critical
Edition uses the term avatāra.” See Nicholas Sutton, Religious Doctrines in the Mahābhā
rata (Delhi: Motilal Banarsidass, 2000), 156.
10 See David L. Gitomer, “Rākṣasa Bhīma: Wolfbelly among Ogres and Brahmans in the San-
skrit Mahābhārata and the Veṇīsaṃhāra,” in Essays on the Mahābhārata, ed. Arvind
Sharma (Leiden: E.J. Brill, 1991), 310.
11 James McHugh, Sandalwood and Carrion: Smell in Indian Religion and Culture (New York:
Oxford University Press, 2012), 93–95. He cites Mānavadharmaśāstra 2.175 and 177 on the
prohibition of flowers to ascetics.
192 Sullivan
Bhīma’s desire-driven quest is the context for the female residents of the forest
regarding Bhīma as if he were “a new incarnation of beauty” while seated
beside their lovers, gazing with interest at him. But this noun avatāra, of
course, has other and very significant connotations that have nothing to do
with the erotic, but with attitudes of devotion to God. Just as the erotic rasa is
being suggested by the setting, Draupadī’s demand, etc., the presence here of
the term avatāra may be taken as suggesting an attitude of religious devotion
(bhakti), a subtle cue to the audience just before Bhīma meets Hanūmat. Use
of the word avatāra at the beginning of this passage is no accident, and is
worth bearing in mind as we look at other aspects of the episode.
This Samāgama episode features one of the main characters, Bhīma, in his
current life so it is part of the main story of the Mahābhārata, not a “mirror
story” such as Nala’s that features other characters.12 Bhīma is very much in
character: he destroys a banana grove, is aggressive, and kills Yakṣas for the
sake of flowers. There are, however, two surprising incidents in this episode
regarding Bhīma. The first is his response to Hanūmat’s suggestion that Bhīma
just jump over him: Bhīma refers to having come to know the Supreme Soul
(nirguṇaḥ paramātmeti) beyond all qualifications, so he won’t humble or insult
him by jumping over (Mahābhārata 3.147.8–9). The occasions in which Bhīma
offers a spiritual reason for not being heroic are few and far between in the
Mahābhārata—indeed, this must be unique. The second surprise regarding
Bhīma is that he does not accept the offer of extraordinary food from Hanū
mat—roots and fruit that are similar to amṛta (146.81). This too must be the
only occasion in the Mahābhārata on which Bhīma declines an offer of food;
his nickname “Wolfbelly” (Vṛkodara) is due to his appetite, an excess of which
is stated to be partly responsible for his death in the Himālayas.13 But he returns
to his usual character once Hanūmat is gone. It is noteworthy, however, that
this episode involves only Bhīma among the main characters, and he does not
mention to anyone else that he had met Hanūmat. The end of the episode
reunites him with his brothers and wife, but Bhīma neglects to mention this
particular event of his day: that he has met a God and been blessed by him,
that this God offered to wipe out their enemies for them, and that this deity
will from now on reside on Arjuna’s flagstaff. Nor does he mention that he has
already met Hanūmat when, later in our text, a fuller account of the story of
Rāma is told to all of them, the “Rāmopākhyāna” (Mahābhārata 3.258–75).
Hanūmat asked that his location be kept secret, but Bhīma keeps secret the
entire encounter.
Bhīma comments that he is aware of the exploits of Hanūmat in the Rāmā
yaṇa, a story he knows by that name. The Critical Edition text has no variant
readings for the line in which the Rāmāyaṇa is named.14 Van Buitenen, in his
brief introductory remarks on the Samāgama asks, “Has this episode been the
inspiration of the later retelling of The Rāmāyaṇa story?”—by which he means
the “Rāmopakhyāna.”15 To me, the reverse is the more likely, as I see this epi-
sode as dependent on a number of other episodes and subtales which would
have had to be in place in our text for this episode to have been created, as it
echoes themes of those other passages. As Philip Lutgendorf states, “… Bhīma’s
quest for the saugandhika flower offers a masterful display of one of the
Mahābhārata’s most striking features: its fugue-like variation on its own
themes ….”16 In my view, indeed, this episode is best understood, by analogy
with the fugue, or (perhaps better) Karṇātak music, as restating and elaborat-
ing themes already introduced elsewhere in the work.17
Are we to conclude from use of the title Rāmāyaṇa that this whole Samā
gama episode was composed once Vālmīki’s text was known to Vyāsa’s
committee, and to audiences? I do not think that this conclusion is war-
ranted. It is important to notice that the title Rāmāyaṇa occurs elsewhere in
14 bhīma uvāca
bhrātā mama guṇaślāghyo buddhisattvabalānvitaḥ |
rāmāyaṇe ’tivikhyātaḥ śūro vānarapuṁgavaḥ || (Mahābhārata 3.147.11)
15 See Van Buitenen, Mahābhārata, vol. 2, 180.
16 See Philip Lutgendorf, Hanuman’s Tale: The Messages of a Divine Monkey (Oxford: Oxford
University Press, 2007), 281–82.
17 As A.K. Ramanujan observed, “I’d suggest that the central structuring principle of the epic
is a kind of repetition. One might say that repetition or replication is the central principle
of any structuring …. Indian artworks, like the Hindu temple, or the decads (pattu) of
Tamil classical or bhakti poetry, or the rāgas of Karnatak music, are built on the principle
of interacting structures of repetition and elaboration and variation.” See A.K. Rama
nujan,“Repetition in the Mahābhārata,” in Essays on the Mahābhārata, ed. Arvind Sharma
(Leiden: E.J. Brill, 1991), 421.
194 Sullivan
18 Van Buitenen interestingly conforms to the same usage in referring to the “Rāmopakhyāna”
subtale as “The Rāmāyaṇa story.” Mahābhārata, vol. 2, 180. His discussion of the two
(Mahābhārata, vol. 2, 207–14, quoting 213–14) concludes that the tale, “after its contents
were fixed in the story of Rāma (“Rāmopakhyāna”), underwent further development,
acquired a new beginning and a new end, attracted subsidiary elements, and became
known as the original poem (ādikāvya) of Vālmīki. The ones responsible for the inclusion
of the story of Rāma in The Book of the Forest either did not know of Vālmīki’s poem, or
knew that the story of Rāma was different from it.” Brockington demonstrated close corre-
spondence between the Northeast recension of the Rāmāyaṇa and the “Rāmopakhyāna”
but also acknowledges the latter’s influence on the former, a literary feedback loop; see
John L. Brockington. The Sanskrit Epics (Leiden: E.J. Brill, 1998), 473–77; and in greater
detail, Epic Threads: John Brockington on the Sanskrit Epics, ed. Greg Bailey and Mary
Brockington (Delhi: Oxford University Press, 2000), 288–325.
The Tale of an Old Monkey and a Fragrant Flower 195
as an Indra pole” (146.60); and his tail is described as “stretched high like Indra’s
rainbow” (147.18). Twice he is an “Indra of monkeys” (147.1 and 12). In addition
to emphasizing his power and divinity, these comparisons to Indra may also
have been made to reinforce the idea of his great antiquity (another indicator
of immortality), a subject to which I shall soon return.
What then are the other episodes in the Mahābhārata, besides the “Rāmo
pākhyāna,” that the Samāgama echoes, or with which it resonates? It is parallel
to the other two episodes in this parvan in which one of the trio of elder
Pāṇḍava brothers encounters a divine relative: Arjuna meets his father Indra
and Yudhiṣṭhira meets his father Dharma in disguise as a riddling Yakṣa. The
other two meetings feature deities who appear in the Mahābhārata’s main
story elsewhere. This is not so regarding Bhīma’s meeting with Hanūmat
(except insofar as Hanūmat could be regarded as present on Arjuna’s flagstaff,
but, as indicated below, he is not explicitly identified in these other references).
These meetings occur in our text with the pattern youngest to eldest brother,
so Bhīma’s encounter is appropriately the middle one. In each case, the
Pāṇḍava has an encounter with a divine family elder who tests and teaches
him: as James Laine has indicated, these “visions of God” are initiatory encoun-
ters.19 Despite the dazzling nature of Bhīma’s initiation, his is surely the least
transformative of the three encounters; perhaps it was added to complete a
pattern, echoing themes from elsewhere in the text.
Additionally, this episode is linked to the “Burning of the Khāṇḍava Forest.”20
This episode that ends the Ādiparvan includes Arjuna receiving from Varuṇa
the distinctive chariot with which he triumphs in the Mahābhārata:
We see the same comparison made between the flagstaff in this episode
(śakrāyudhasamā) and Hanūmat’s arching tail in the Samāgama (indrāyudham):
both are like Indra’s bow. Throughout the Mahābhārata there are many refer-
ences to Arjuna’s chariot being “monkey-bannered” (kapiketu, etc.). The
“Hanūmād-bhīma-samāgama” is unique in the Critical Edition in identifying
the “divine monkey” on Arjuna’s banner as Hanūmat, and supplying a reason
for the monkey to be his emblem.22 As we have seen, the text states that
Hanūmat seeks to do a favor for his brother Bhīma, so he will add his roar to
Bhīma’s on the battlefield, and he will reside on Arjuna’s banner.23 This means
that Arjuna has been driving around in his chariot for some years with a divine
monkey on its flagstaff before Hanūmat announces his kind offer to reside
there. Notice that Hanūmat does not say that he has been there already for
several years; he seems only now to be taking up residence on Arjuna’s flag-
staff. This, and the fact that Hanūmat occupies Arjuna’s rather than Bhīma’s
own banner, suggest to me that the numerous references to Arjuna’s monkey
emblem were already in place throughout the text, and this episode is amplify-
ing on those, explaining the identity of that divine monkey. In other words, this
episode is dependent on the many references to Arjuna’s monkey banner, and
takes its place in the text at the end of its period of composition, however long
that was, in explanation of the numerous existing monkey-banner references.
Since we know of no connection between Hanūmat and Arjuna—after all,
they never met—we must wonder why Arjuna has a monkey as his emblem,
here uniquely identified as Hanūmat.
The only possible answer I see for this is Vedic. There is one Vedic poem in
which Indra is associated with a monkey, Vṛṣākapi, the “bull-monkey” (Ṛgveda
10.86). This poem is difficult to interpret, and is the only one in which Vṛṣākapi
appears. As Stephanie Jamison indicates, the poem may have been part of the
horse sacrifice’s liturgy.24 In any case, this Vedic poem would provide a reason
for Arjuna, son of Indra, to have as his battle-emblem a divine monkey—
namely, that this monkey is associated with Indra, and therefore with Vedic
sacrifice and timeless truth. Clearly Vṛṣākapi was not Hanūmat in the Ṛgveda,
but the Mahābhārata here identifies the monkey banner with Hanūmat and
that association has persevered.25 Indeed, Hanūmat and Bhīma remain associ-
ated in the popular imagination in a variety of ways, among which these can be
noted:
The Samāgama episode thus draws on the association between Indra and
Arjuna, well established throughout our text, and associates Hanūmat with
both Indra and Arjuna. The rest of the Critical Edition, however, does not know
Hanūmat as the identity of the monkey on Arjuna’s banner. To me this suggests
24 See Stephanie W. Jamison, Sacrificed Wife, Sacrificer’s Wife: Women, Ritual, and Hospitality
in Ancient India (New York: Oxford University Press, 1996), 74–88.
25 As noted by Brockington in John L. Brockington, “Hanumān in the Mahābhārata,” Journal
of Vaishnava Studies 12, no. 2 (2004): 133.
26 See Philip Lutgendorf, “Monkey in the Middle: The Status of Hanuman in Popular Hindu-
ism,” Religion 27 (1997): 313.
27 See Phillip B. Zarrilli, Kathakali Dance-Drama: Where Gods and Demons Come to Play
(London: Routledge, 2000), 101–17 for a translation of the dance-drama Kalyāna-Saugan
dhikam, composed about the year 1700, with pictures and discussion.
28 See Joseph S. Alter, The Wrestler's Body: Identity and Ideology in North India (Berkeley: Uni
versity of California Press, 1992), 261.
198 Sullivan
that this identity was entered into the text at the very end of its compositional
history.
Another episode linked to the Samāgama is the Bhagavadgīta. Hanūmat com-
plies with Bhīma’s request to see his “incomparable form” (rūpam apratimaṁ;
Mahābhārata 3.148.3) in which he leaped over the ocean, just as Kṛṣṇa fulfills
Arjuna’s request to see his supreme form (rupam aisvaryaṃ; Bhagavadgītā
11.3) in the Gītā. In that incomparable form, Hanūmat delivers teachings on
kṣatriya-dharma and rājā-dharma as well as time and the four-yuga system
of world eons.29 Such teachings parallel themes in the Bhagavadgītā, as when
Kṛṣṇa refers to himself as Time (Bhagavadgītā 11.32) and insists that Arjuna ful-
fill his dharma as a kṣatriya. Another parallel between the Samāgama and the
Gītā is their references to Viṣṇu. Hanūmat identifies Rāma as “Viṣṇu in human
form” (147.28), and Kṛṣṇa is addressed twice as “Viṣṇu” when Arjuna is viewing
the supreme form Kṛṣṇa reveals (Bhagavadgītā 11.24 and 30). In addition, both
Hanūmat and Kṛṣṇa promote devotion as a religious practice. Hanūmat advo-
cates that Bhīma worship the gods with offerings and bhakti (149.24), and while
this is less specific as to the object of that devotion than Kṛṣṇa’s admonition
to be devoted in particular to him (Bhagavadgītā 18.65), the religious practice
is the same. And note that the vision of Kṛṣṇa’s form is a secret available only
to Arjuna,30 as the vision of Hanūmat and his location is to be Bhīma’s secret.31
This episode magnifies the importance of Hanūmat by comparing him implic-
itly to Kṛṣṇa; the depiction of the two of them performing such similar actions
and imparting such similar teachings, each to a Pāṇḍava brother, suggests
that we view Hanūmat and Kṛṣṇa as comparable divine figures. Certainly this
29 Luis González-Reimann has argued that the yuga teaching is not integral to the Mahābhā
rata, and not consistently presented either; Mārkaṇḍeya’s discourse (187) disagrees with
Hanūmat’s (147) in certain respects; see Luis González-Reimann, The Mahābhārata and
the Yugas: India’s Great Epic Poem and the Hindu System of World Ages (New York: Peter
Lang, 2002), 102–6. In my view, the four-yuga idea may be “integral” to the Mahābhārata
even if it is not discussed frequently in the text. That the four-yuga conception of time is
taught may be seen as a likely indicator of the relative lateness of this passage, but that
does not mean that it is irrelevant to the conception of the text as we have it in the Critical
Edition. Certainly, our understanding of the historical relationship between the Mahāb
hārata and Rāmāyaṇa cannot be based on the traditional understanding that the latter
describes events of a yuga prior to the Mahābhārata’s events—even if the Mahābhārata
depicts Hanūmat as ancient.
30 Kṛṣṇa states that no one but Arjuna has ever beheld his supreme form, and it cannot be
seen by anyone but Arjuna (Bhagavadgītā 11.47–48 = Mahābhārata 6.33.47–48).
31 In like fashion, Yudhiṣṭhira is the only one to get a vision of his father Dharma, since
Dharma talks to him only when Yudhiṣṭhira’s four brothers are unconscious (end of book
3) or already dead (book 18).
The Tale of an Old Monkey and a Fragrant Flower 199
episode also has a humorous dimension: the audience knows the identity of
Hanūmat while Bhīma does not, and so he is humbled. This contrasts with
the serious tone of the Gītā, but the divinity of Hanūmat is no joking matter.
That the Mahābhārata’s audience needs no introduction to Hanūmat and is in
on the joke from the outset is worth noticing. The fact that these two Pāṇḍava
brothers each received such similar teachings, each having begged forgive-
ness and asked the deity to reveal his supreme form, after which each warrior
requested that he resume his prior, smaller form, can only indicate that one
account is based on the other. This suggests to me that the Bhagavadgītā, so
much more dramatic and well-integrated into the Mahābhārata as a whole,
has provided the model and the Samāgama resonates with the Gītā, which it
here foreshadows in the text.
Another episode linked to the Samāgama is parallel to it and follows it
almost immediately in the text (157–59). This passage features Bhīma going
on a quest for more saugandhika flowers up Gandhamādana mountain at
Draupadī’s request, where he finds the flowers, kills Yakṣas, meets Kubera who
blesses him, and collects more flowers for Draupadī (though she had a bunch
of them a couple of chapters earlier!).32 Van Buitenen had thought that the
second ascent of the Gandhamādana was “a recast of the first.”33 Grünendahl
points out correctly that the Parvasaṃgraha table of contents (Mahābhārata
1.2) shows no awareness of the “Samāgama,” only the second ascent of Gandha
mādana, which he deems the original version.34 Themes in the two paired
32 The other three Pāṇḍava brothers and Draupadī joined Bhīma (Arjuna not having
returned as yet), and Bhīma handed over the lotus flowers (Mahābhārata 3.153.28):
anuśāsya ca kaunteyaṁ padmāni pratigṛhya ca |
tasyām eva nalinyāṁ te vijahrur amaropamāḥ ||
Danielle Feller rightly observes, “After this great effort on Bhīma’s part to get the lotuses
for Draupadī, the end of the story is something of an anti-climax … we would expect
Bhīma to give them to Draupadī with some pomp and ceremony.” Feller, “Bhīma’s Quest
for the Golden Lotuses (Mahābhārata 3.146–53 and 3.157–59),” 91. I take this verse as indi-
cating that Draupadī does receive the flowers for which she had sent Bhīma, not as Feller
does when she states, “We do not even know if Draupadī herself … ever receives them.”
Ibid., 91–92.
33 Van Buitenen, Mahābhārata, vol. 2, 201–2.
34 Grünendahl argues that efforts have been made by the interpolator of the Samāgama to
accommodate it, and mark what had been the original ascent account (155–58) as the
second; see Reinhold Grünendahl, “Zu den beiden Gandhamādana-Episoden des Āraṇya
kaparvan,” Studien Zur Indologie und Iranistik 18 (1993): 110–11. However, as noted by Hilte-
beitel both accounts have Southern Recension interpolations, with the Samāgama
episode having even more than the other, a finding that does not support Grünendahl’s
idea that the text’s second ascent (155–58) was necessarily in place earlier than the
200 Sullivan
[T]he author of this passage, being intimately familiar with the epic con-
tents before him and in particular with the special rapport between the
Pāṇḍavas’ deeds while in exile and their final acts, naturally reflected
these themes within his account of the Gandhamādana ascent and in so
doing deepened and enriched the text’s narrative integrity.36
“Samāgama.” See Alf Hiltebeitel, “On Sukthankar’s ‘S’ and Some Shortsighted Assessments
and Uses of the Pune Critical Edition (CE).” Journal of Vaishnava Studies 19, no. 2 (2011):
109.
35 As argued by both Grünendahl, “Zu den beiden Gandhamādana,” and Christopher Austin,
“Draupadī’s Fall: Snowballs, Cathedrals, and Synchronous Readings of the Mahābhārata,”
International Journal of Hindu Studies 15, no. 1 (2011): 111–37.
36 Austin (“Draupadī’s Fall,” 130) also states: “… we are witnessing, I believe, an echo from the
epic conclusion bouncing back into a later contribution to Book 3.”
37 As noted by Grünendahl (“Zu den beiden Gandhamādana,” 125–28) for Mahābhārata
3.140 through 3.152. He also cites Mahābhārata 2.10.22, describing Kubera’s sabhā, which
floats in the sky like a peak of Mount Kailāsa, at which Śiva and his gaṇas reside.
38 3.146.78–81 and 3.147.40–41. Biardeau, Le Mahābhārata, vol. 1, 546–47, and Feller, “Bhīma’s
Quest,” 85 also state that the mountain is a heavenly site.
The Tale of an Old Monkey and a Fragrant Flower 201
linked thematically, The Five Indras (1.189). I see these as sharing a theme of a
quest for a special flower that results in a revelation: Indra climbs a mountain
seeking the source of the golden lotus flowers floating down the River Gaṅgā
and aggressively encounters Śiva, who humbles him. Śiva decrees, and
Nārāyaṇa agrees, that this Indra and four former Indras will be born as the five
Pāṇḍavas, and that the Goddess Śrī will take birth as their wife. Vyāsa then
reveals this secret to King Drupada, so this revelation about the divine plan
comes from a family elder. Perhaps the saugandhika flower episode is echoing
the pattern of the quest for a special flower in The Five Indras with its revela-
tion to a Pāṇḍava family member by a family elder. This would be another
instance of repetition, elaboration, and variation on an important theme.
I cannot resist commenting on one other way in which this episode is sug-
gestive and thematically rich. Not long after Bhīma meets Hanūmat, the
Pāṇḍavas spend their year incognito, and Bhīma announces that his disguise
would be as cook and wrestler.39 As Joseph Alter observes, “Kings have kept
wrestlers because the physical strength of the wrestler symbolizes the political
might of the king.”40 Obviously, Bhīma is well suited to such symbolic repre-
sentation. Not only does he defeat some professional wrestlers, he mangles the
general Kīcaka with his bare hands to protect Draupadī. I mention Bhīma’s
depiction as a wrestler because of the strong association modern wrestlers in
India have with Hanūmat, whom they worship, and with Bhīma, who is also
praised and commemorated at their gymnasiums.41 Interestingly, Bhīma
defends Draupadī from the relatives of Kīcaka by magnifying his size, after
which he found it necessary to rearrange his clothing before smashing them
with an uprooted tree.42 Like his elder brother Hanūmat, Bhīma too here man-
39 Mahābhārata 4.2.1–7, where the term (used twice) for wrestler or pugilist is niyodhaka-.
40 See Alter, who also states: “Royal courts and princely estates have sponsored wrestlers
probably since the time of Kansa, Krishna, Ravana, and the Pandava brothers. However,
there is no detailed historical record of this and no way of telling whether wrestling
patronage has changed over time. In all likelihood, the formal aspect of patronage has not
changed significantly ….” The Wrestler's Body, 71–72.
41 Alter comments, “In spite of the incipiently sectarian tone of much wrestling rhetoric—
where images of Shiva, shakti, and previous Pandava war heroes abound—there is usu-
ally also a more pervasive tone of secular, non-communal fraternity.” Ibid., 262.
42 Mahābhārata 4.22.17–18.
ity uktvā sa mahābāhur vijajṛmbhe jighāṁsayā |
tataḥ sa vyāyataṁ kṛtvā veṣaṁ viparivartya ca |
advāreṇābhyavaskandya nirjagāma bahis tadā ||
sa bhīmasenaḥ prākārād ārujya tarasā drumam |
śmaśānābhimukhaḥ prāyād yatra te kīcakā gatāḥ ||
202 Sullivan
ifests an enormous form for a heroic act. While I find no textual basis in the
Mahābhārata for thinking that his meeting with Hanūmat led Bhīma to his
disguise as a wrestler, or his ability to enlarge his form, it is a tempting idea; it
would be so appropriate!
The Samāgama may be a relatively late addition to the Mahābhārata, but it
is incontestably part of the Critical Edition, and we are exceedingly limited in
our ability to envision any version of the Mahābhārata prior to the Critical
Edition; indeed, to imagine any earlier version is precisely that, an act of imagi-
nation.43 This episode is linked thematically to many other episodes, even if
the Critical Edition has no other references to Bhīma meeting Hanūmat. The
Samāgama episode uses the term bhakti—does this help us date this episode?
I note that the last verse of the Śvetāśvatara Upaniṣad advocates that one cul-
tivate bhakti for God and for one’s guru. Patrick Olivelle has stated that the
Śvetāśvatara Upaniṣad was “composed probably in the last few centuries bce.”44
If he is correct, such a date for this use of the term bhakti would correlate well
with Alf Hiltebeitel’s theory that the Mahābhārata was written between 150
bce and the year 1.
Perhaps the feature of this episode that is most problematic with regard to a
date of composition about 100 bce for the Mahābhārata as a whole, as Alf
Hiltebeitel envisions, is that it presents Hanūmat as a deity. The episode does
not exactly depict Bhīma worshiping Hanūmat—he asks if Hanūmat is a deva
or some other type of being greater than human, but gets no answer, he bows,
he asks forgiveness, as one might of a deity. Although Hanūmat advocates
bhakti, he does not specify that it should be directed toward him in particular,
but he is certainly shown to be divine. In that respect, this episode represents
Hanūmat as a more exalted being than does the Rāmāyaṇa or the Rāmo
pākhyāna! I say this in light of the depiction of Hanūmat in the “Samāgama,” a
depiction clearly intended to parallel that of Kṛṣṇa in the Bhagavadgītā. The
text shows Kṛṣṇa and Hanūmat each revealing to a Pāṇḍava warrior his
supreme form, which is overwhelming to the warrior, who then requests that
he resume his more familiar form. In addition, each deity transmits teachings
on the importance of following one’s dharma, specifically kṣatriya-dharma,
teachings transmitted privately and as a secret. In both the Mahābhārata and
45 Alf Hiltebeitel, “The Nārāyaṇīya and the Early Reading Communities of the Mahābhā
rata,” in Between the Empires: Society in India 300 bce to 400 CE, ed. Patrick Olivelle (New
York: Oxford University Press, 2006), 227.
204 Sullivan
References
Primary Sources
The Mahābhārata for the First Time Critically Edited. Edited by Vishnu S. Sukthankar,
Shripad K. Belvalkar, Parashuram L. Vaidya, et al. 19 vols. Poona: Bhandarkar Oriental
Research Institute, 1933–66.
The Mahābhārata, vol. 1. 1. The Book of the Beginning. Translated by J.A.B. van Buitenen.
Chicago: University of Chicago Press, 1973.
The Mahābhārata, vol. 2. 2. The Book of the Assembly Hall; 3. The Book of the Forest.
Translated by J.A.B. van Buitenen. Chicago: University of Chicago Press, 1975.
Secondary Sources
Adluri, Vishwa. “Frame Narratives and Forked Beginnings: Or, How to Read the
Ādiparvan.” Journal of Vaishnava Studies 19, no. 2 (Spring 2011): 143–210.
Alter, Joseph S. The Wrestler’s Body: Identity and Ideology in North India. Berkeley:
University of California Press, 1992.
Austin, Christopher. “Draupadī’s Fall: Snowballs, Cathedrals, and Synchronous Readings
of the Mahābhārata.” International Journal of Hindu Studies 15, no. 1 (2011): 111–37.
Bailey, Gregory M. and M. Brockington, eds. Epic Threads: John Brockington on the
Sanskrit Epics. Delhi: Oxford University Press, 2000.
Biardeau, Madeleine. Le Mahābhārata: Un récit fondateur du brahmanisme et son inter-
prétation, 2 vols. Paris: Éditions du Seuil, 2002.
Brockington, John. “Hanumān in the Mahābhārata.” Journal of Vaishnava Studies 12, no.
2 (2004): 129–35.
Brockington, John L. The Sanskrit Epics. Leiden: E.J. Brill, 1998.
Feller, Danielle. “Bhīma’s Quest for the Golden Lotuses (Mahābhārata 3.146–53 and
3.157–59).” In Battle, Bards and Brāhmins. Papers of the 13th World Sanskrit Conference,
Volume II, edited by John Brockington, 79–99. Delhi: Motilal Banarsidass, 2012.
Fitzgerald, James L. “Mahābhārata.” In Brill’s Encyclopedia of Hinduism, vol. 2. Edited by
Knut A. Jacobsen, Helene Basu, Angelika Malinar, and Vasudha Narayanan, 72–94.
Leiden: E.J. Brill, 2010.
Gitomer, David L. “Rākṣasa Bhīma: Wolfbelly among Ogres and Brahmans in the Sanskrit
Mahābhārata and the Veṇīsaṃhāra.” In Essays on the Mahābhārata, edited by Arvind
Sharma, 296–323. Leiden: E.J. Brill, 1991.
González-Reimann, Luis. The Mahābhārata and the Yugas: India’s Great Epic Poem and
the Hindu System of World Ages. New York: Peter Lang, 2002.
Grünendahl, Reinhold. “Zu den beiden Gandhamādana-Episoden des Āraṇyakaparvan.”
Studien Zur Indologie und Iranistik 18 (1993): 103–38.
Hiltebeitel, Alf. “Authorial Paths Through the Two Sanskrit Epics, Via the Rāmopākhyāna.”
In Epic Undertakings: Papers of the 12th World Sanskrit Conference, vol. 2, edited by
Robert P. Goldman and Muneo Tokunaga, 169–214. Delhi: Motilal Banarsidass, 2009.
The Tale of an Old Monkey and a Fragrant Flower 205
———. “The Burning of the Forest Myth.” In Hinduism: New Essays in the History of
Religions, edited by B.L. Smith, 208–24. Leiden: E.J. Brill, 1976.
———. Dharma: Its Early History in Law, Religion, and Narrative. New York: Oxford
University Press, 2011.
———. “The Nārāyaṇīya and the Early Reading Communities of the Mahābhārata.” In
Between the Empires: Society in India, 300 bce to 400 CE, edited by Patrick Olivelle,
227–56. New York: Oxford University Press, 2006.
———. “Not Without Subtales: Telling Laws and Truths in the Sanskrit Epics.” Journal
of Indian Philosophy 33, no. 4 (2005): 455–511.
———. “On Sukthankar’s ‘S’ and Some Shortsighted Assessments and Uses of the Pune
Critical Edition (CE).” Journal of Vaishnava Studies 19, no. 2 (2011): 87–126.
———. Reading the Fifth Veda: Studies on the Mahābhārata. Essays by Alf Hiltebeitel,
vol. 1. Edited by Vishwa Adluri and Joydeep Bagchee. Leiden: E.J. Brill, 2011.
———. Rethinking the Mahābhārata: A Reader’s Guide to the Education of the Dharma
King. Chicago: University of Chicago Press, 2001.
Hopkins, Edward Washburn. The Great Epic of India: Its Character and Origin. Calcutta:
Punthi Pustak, 1969. First published 1902 by Charles Scribner’s Sons.
Jamison, Stephanie. Sacrificed Wife/Sacrificer’s Wife: Women, Ritual and Hospitality in
Ancient India. New York: Oxford University Press, 1996.
Laine, James W. “Out of Character: Marginal Voices and Role-Transcendence in the Ma
hābhārata’s Book of the Forest.” Journal of Indian Philosophy 19 (1991): 273–96.
———. Visions of God: Narratives of Theophany in the Mahābhārata. Vienna: Institut
für Indologie der Universität Wien, 1989.
Lutgendorf, Philip. Hanuman’s Tale: The Messages of a Divine Monkey. Oxford: Oxford
University Press, 2007.
———. “Monkey in the Middle: The Status of Hanuman in Popular Hinduism.” Religion
27 (1997): 311–32.
McHugh, James. Sandalwood and Carrion: Smell in Indian Religion and Culture. New
York: Oxford University Press, 2012.
Olivelle, Patrick, trans. Upaniṣads. Oxford: Oxford University Press, 1996.
Ramanujan, A.K. “Repetition in the Mahābhārata.” In Essays on the Mahābhārata, edited
by Arvind Sharma, 419–43. Delhi: Motilal Banarsidass, 2007.
Sutton, Nicholas. Religious Doctrines in the Mahābhārata. Delhi: Motilal Banarsidass,
2000.
Zarrilli, Phillip B. Kathakali Dance-Drama: Where Gods and Demons Come to Play.
London: Routledge, 2000.
206 Wulff Alonso
Chapter 7
Introduction
This book is one of the many ways in which we are recognizing Alf Hiltebeitel’s
contribution as a brilliant researcher and a lucid witness of nearly half a cen-
tury of research on the Mahābhārata.
It is difficult to define the major changes in the field during this time, and it
is even more difficult for someone like me who works in another field; how-
ever, we could certainly point to several elements common to various social
sciences, which could, perhaps, be further fleshed out by referring to closer
fields, such as studies of Greek mythology and epics.
The first major change is the impact of Parry’s and Lord’s perspectives on
oral composition and performance in the epics. The second change, which is
more complex, is the impact of French thought after World War II, the move-
ment that we could broadly define as “structuralism.” The role of the Annales
School, created under the direction of Lucien Febvre, who was director of the
Annales and of the Sixth Section of the École des Hautes Études en Sciences
Sociales, and of scholars such as Louis Gernet and Jean-Paul Vernant clearly
parallels Louis Dumont or, more specifically, Madeleine Biardeau of the Fifth
Section of the École and its Centre d’Études de l’Inde et de l’Asie du Sud.
It is certainly difficult to review all of these changes and to simultaneously
consider contributions as varied as those of Georges Dumézil, Claude Lévi-
Strauss, Biardeau, or Gernet. However, their most essential trait here, as in
other social sciences, is likely the call for understanding processes and phe-
nomena as a whole and not in isolation. For myths or epics, the question is the
same: consider the works as a whole and connect them with all that surrounds
them, from the creators, performers, and audience to the genres, tradition, and
society.
Surely, we must add two additional components: the impacts of gender
studies and the criticisms of Eurocentric models. Postcolonial concepts imply
not only setting aside perspectives that define other societies and their pro-
ductions as inferior to Western canons, but also discussing these perspectives
and their implications. The values that supported aesthetic valuations, for
example, were full of assumptions about the existence of a single (classical)
aesthetic and rationality and automatically implied assessments about the
capabilities of the societies that produced such texts or works of art.
In this landscape, Alf Hiltebeitel’s and Madeleine Biardeau’s1 ideas about
the Mahābhārata’s unity and internal consistency are not surprising. Their
ideas involve further strengthening and outcropping a position that was
despised and even ridiculed by those who adhered to the idea of an accumu-
lative Mahābhārata, the dominant paradigm. What is, perhaps, surprising is
the practically undisputed hegemony of these positions, when their basis—
analysts’ interpretations of the Homeric question—has been systematically
revised during the century between Hopkins and the present day. Although
Alf Hiltebeitel noted the upākhyānas as the new frontier of research on the
Mahābhārata,2 I would say that there is another disregarded frontier: historio-
graphical questions, which Vishwa Adluri and Joydeep Bagchee have begun to
explore.3 The existence of scholars resistant to the “invasion” of such analysts
will not be the only surprising consequence of Adluri’s and Bagchee’s work.
It is well-known that there are two distinct groups in the new unitary
paradigm: one that believes in a systematic reworking of earlier Mahābhā
ratas within a previous epic tradition, and the other—represented by Alf
Hiltebeitel—that believes in a work made ex novo. The latter would involve
the work of a team attempting to formulate a story that would develop a com-
plex ideology in competition with other ideologies, such as Buddhism, in the
context of redefining the monarchy under the shadow of bhakti, and after
Alexander the Great. This position confronts the traditional idea of a massive
accumulation of components and developments in all directions and defends
the idea of a complex work that seeks to define a competitive ideology, an
ideological pretention that I would describe more as the search for a popular
front than for a political party. It would also draw on all sources and resources
for its establishment and dispersion, articulating in writing the main story,
side stories and doctrinal perspectives, allowing different interpretations and
performances.
The upākhyānas have traditionally been considered by the dominant para-
digm as obvious examples of the cumulative component of the Mahābhārata,
often incompatible with the main story. New approaches require reconstruct-
ing the thought and aesthetic frames that the authors display in a way that
allows us to understand their functions in a work, which can easily be described
as experimental and foundational.
My point here is simple. I raise the question of the existence of a contrast
between the architectures of power and the role of Indra in the main and side
stories of the Mahābhārata. My conclusion is that there is such a contrast,
which I will attempt to define, and I will suggest answers to the questions of
why and how.
I will not address a major problem in defining the upākhyānas, which has
already been raised by Alf Hiltebeitel: the upākhyānas are not all the same.
For example, we must distinguish between those that explain the birth of
the key characters (Bhīṣma or Satyavatī), the prior or contemporary exploits
of the essential characters (Rāma Jāmadagnya or Kṛṣṇa), the stories that are
not directly related to the main plot (Nala and Damayantī), the reworkings of
Vedic stories (Indra and Vṛtra), and the stories that are narrated by characters
who tell another story that mirrors the current situation (for example, Kuntī
tells Kṛṣṇa a story for Yudiṣṭhira, to incite him to war in the Udyogaparvan
[Mahābhārata 5.131–4]). I believe that the contrast I suggest is valid for the dif-
ferent categories of upākhyānas.
I shall focus on the architecture of power. This concept aims at defining the
hierarchical framework deployed and put into play in epics. The worlds that
epic poets construct in polytheistic societies necessarily cover macrocosms
Supernatural Conflicts, Unanimities, and Indra 209
and microcosms4 and define hierarchical structures in both levels and the rela-
tions between them. Conflicts in epics, which are by definition conflicts of
power and hierarchy, are categorized in these structures and put them in play.
The presence of the macrocosm wherein deities play the leading roles in the
story (as is normally the case in the Ancient Near East) is obvious enough.
However, the Ancient Near East offers us an example, almost an exception, of
a story featuring a human being: the poem of Gilgamesh. To more precisely
define the concept, it may be useful to briefly refer to this poem, the earliest
story that could be defined as an epic.
The general framework of the story is clear: the difference between gods
and humans.5 Sacrifice is the sign of submission and feeds the gods. The pro-
tagonist is a Kṣatriya, the king of Uruk, Gilgamesh, the son of a goddess and
a man, but he is also a human, who is responsible, as king, for establishing a
connection between the two worlds: handling human affairs with justice and
maintaining the necessary links with the gods.
The first conflict is not a direct conflict with the gods but is between the king
and his subjects. Gilgamesh is a powerful king who, in some respects, is abusive
to his subjects, an aspect that relates to his excessive force and virility. The
focus, then, is the dharma of this Kṣatriya. The subjects complain to the gods
because he does not fulfill his role as king well. The gods create and send
Enkidu against him, at first a wild man, who is then civilized by a prostitute.
Their confrontation, though, ends in friendship.
This story opens two new conflicts. First, through their heroic activities,
together they both (literally) break more boundaries with the divine sphere,
which, in turn, leads to the second conflict: the amazed goddess Ishtar pro-
poses a sexual-marital liaison to Gilgamesh, which he refuses, offending her.
After a failed attempt to punish him for this offense and another offense to the
goddess, his inevitable punishment takes the form of the death of Enkidu.
It would not be useful here to deal with the core of the problem: the encoun-
ter of a powerful female with an inferior man, which is all but commonplace in
epics (think of the Iliad, the Odyssey, the Nibelungenlied, the Táin Bó Cúailnge,
and the Mahābhārata itself6). What is useful is underscoring that the punish-
ment is simply a reminder of the inferiority of human beings to gods, which
makes Gilgamesh’s quest for immortality impossible. The meeting with Uta-
napishti (the proto-Noah) allows him to understand the secret of the gods: the
flood that had wiped out all of humanity except for Uta-napishti and his fam-
ily. If a flood is needed for only one man to become immortal, there is little
hope for Gilgamesh. At the end of the poem, his learning and fate are clear: to
be a good king, to enjoy human life, and to tell his marvelous story.
The story of Gilgamesh is inseparable from the hierarchies of the human and
divine orders (which are the same), the king as an intermediary (incidentally,
an almost inevitable factor in monarchical societies prior to industrialization),
and the obvious importance of the king’s role and, in particular, his morality.
There is no epic without ethics; the epic is not a production created by (or
dedicated to) dull warriors who are only interested in pure military exploits.
Additionally, there is no possibility of understanding the story without a
global view and without taking into consideration its architecture of power.
The poem speaks of contrasts in these areas, neither of mediations (in spite of
Lévi-Strauss) nor of negotiations.
Some Unanimities
This approach could enable us to achieve enough perspective to apply this set
of criteria to the Mahābhārata. Its main story shows some clear points of cor-
relation with the epic poem of Gilgamesh:
1. The separation between the divine and human beings (although with
nuances that we will see in the case of the Mahābhārata) is reinforced
by the death of humans and is, in a sense, its demonstration.
2. The poem focuses on intermediaries between the human and divine
beings: the Kṣatriya-kings.
3. It is inseparable from the ethics and justice of the kings, who are
essential intermediaries between the divine and the human.
We could connect two further components with the theme of the flood:
4. It details a planned massacre, which is the secret of the gods. In the case
of the Mahābhārata, its narrative itself reveals the “secret of the gods” as
an apocalypse in the double sense of the word: a pending catastrophe
and a revelation of this secret.7
7 As hinted in James L. Fitzgerald, “Mahābhārata,” in The Hindu World, ed. Sushil Mittal and
Gene Thursby (New York: Routledge, 2004), 103 and 122, although in a more specific sense of
the word.
Supernatural Conflicts, Unanimities, and Indra 211
I would like to note here that these seven points (like the previous five)
correspond to the interpretation of the Iliad in our oldest indirect sources
on it, particularly Hesiod but also in the Homeric Cycle poems, the Cypria.9
According to this interpretation, the wars of Thebes and Troy were devised by
8 This theme, not present in the poem of Gilgamesh (see George, The Babylonian Gilgamesh
Epic, vol. 1, 509–10), is however part of another ancient near east story which most probably
influenced Greek versions; see Walter Burkert, The Orientalizing Revolution: Near Eastern
Influence on Greek Culture in the Early Archaic Age (Cambridge, MA and London: Harvard
University Press, 1995), 100 ff.
9 See Poetae Epici Graeci. Testimonia et Fragmenta, ed. Alberto Bernabé (Leipzig: Teubner, 1988),
Cypria fragm. 1; 2 ff. and the abstract of Proclus; Scholiast to Iliad 1, 5; Hesiod in Hesiod,
Theogonia, Opera et Dies, Scutum, ed. Friedrich Solmsen; Fragmenta Selecta, ed. R. Merkelbach
and M.L. West (Oxford: Oxford University Press, 1970), fragment 204; Hesiod, Works and Days
156 ff.; Fernando Wulff Alonso Grecia en la India: El repertorio griego del Mahābhārata (Madrid:
Akal Eds, 2008), 110 ff.; 147 ff.; The Mahābhārata and Greek Mythology, (New Delhi: Motilal
Banarsidass, 2014), 127 ff.; 175 ff.
212 Wulff Alonso
Zeus to exterminate the generation before us, the heroes’ generation, leading
to the present time. Zeus is a guardian of destiny who looks for the end of a
generation produced by the dangerous mixture of humans and gods. After
their glories and sins are eliminated comes an age of decadence.
Again, I find some differences between the Mahābhārata and this interpre-
tation of the Iliad that are particularly interesting for this discussion:
1. There is no god like Zeus, guardian of destiny, who controls and manipu-
lates the Olympian gods, who favor one side or the other. On the contrary, there
is substantial unanimity in the celestial forces.
2. The confrontation is more indirect and occurs elsewhere. The conflict is
with other supernatural beings, the Asuras, and takes place through the human
beings involved. The war is to be won by the celestial forces. There are no direct
clashes between the celestial forces and the Asuras in the main story. As in the
Greek world, the Asura-Titans had been previously defeated, but unlike in that
story, they are able to battle to dominate the coming era, the Kaliyuga. Both
Asuras and gods resort to the same expedient: partial incarnations, a very spe-
cific component in religious terms but also in terms of narrative structures.
The confrontation becomes substantial among humans incarnated by these
two groups of supernatural beings. There is a certain tendency for the celestial
forces to reincarnate in the Pāṇḍavas and for the Asuras in the Kauravas, but
this tendency is not so radical, as Karṇa, Bhīṣma, Droṇa, and Aśvatthāman,
fight with the Kauravas. As in the Iliad, there are sons of the celestial forces on
both sides (Bhīṣma, Karṇa/the Pāṇḍavas). The confrontation between them,
like clashes between pro-Achaean and pro-Trojan Greek gods, actually feeds
the final kill.
3. Although both human parties present similar religious perspectives, the
celestial forces’ project faces the Asuras’ project. Their project involves a king
that protects the dharma among human beings and their connections with the
celestial forces, defending, for example, justice, the Vedas, Brahmans, ascetics
and asceticism, and pilgrimage sites. This idea appears in the main and sec-
ondary stories as well as in the wisdom sections: good kings produce good
human worlds, whereas bad kings or the absence of kings implies droughts,
famines, Rākṣasas’ and Asuras’ attacks, perversion of the social order, and a
lack of sacrifices, ascetics, and food for the gods.
4. The celestial forces are defined in the text as a complex group with
two main participants, gods and great Ṛṣis, but also with other groups, guests
and servants of the gods (ancestors of the Kṣatriyas’ lineages, Gandharvas,
Apsarās …). There are no pro-Kaurava gods (though Sūrya specifically supports
Supernatural Conflicts, Unanimities, and Indra 213
his son Karṇa and Gaṅgā her son Bhīṣma)10 or great Ṛṣis, and Brahman war-
riors, such as Droṇa, reluctantly take this side, expressing his feelings of guilt,
degradation and disgrace.
5. Indra appears as the nominal king of the celestial forces. He, like Zeus,
contemplates the main events from his celestial abode, expressing approval
or rejection of Arjuna’s display after learning the art of fighting and his con-
frontation with Karṇa, Draupadī’s svayaṃvara, Yudiṣṭhira’s coronation in
Indraprastha, the sufferings of Draupadī in the sabhā of Hāstinapura, Kaurava’s
defeat by Arjuna in the kingdom of Virāṭa, and the vicissitudes of war, amongst
other things. Indra intervenes directly in few cases and usually in relation to
his son, Arjuna. For example, Indra welcomes Arjuna into his heaven, advises
him on how to obtain more weapons, disarms rival Karṇa (though before the
war), and supports Arjuna in their confrontations with him. Even his strange
appearance in the scene of the burning of the Khāṇḍava forest—a story, more-
over, that is not part of the conflict between the Kauravas and Pāṇḍavas—is
related to Arjuna, though entailing a confrontation with him, Kṛṣṇa and Agni.
6. Indra’s nominal hegemony is scarcely challenged in the main story. The
Khāṇḍava forest scene is precisely one of the few exceptions, though occasion-
ally his powerlessness is paramount, such as when the Brahmans officiating
the sacrifice commissioned by Janamejaya force him to drop the snake Takṣaka
that is about to be devoured by the fire (Mahābhārata 1.51.11-13).
7. However, at the same time, Indra is not the Zeus who controls and directs
gods acting on the earth. While Zeus manipulates Athena, Hera, or the pro-
Trojan gods (let alone the humans involved), the agents of the celestial forces
on Earth are fully independent: the great Ṛṣi Vyāsa and the bhakti god Kṛṣṇa.
Vyāsa, the perfect Brahman,11 directs the processes and gives birth to some of
the actors; Kṛṣṇa also directs the processes and specifically controls the two
killings so that he intervenes directly to achieve the double aim of the game:
the general massacre and the (relative) triumph on the side of justice. Vyāsa
tells the story and, within it, gives Saṃjaya voice and protects him until the end
of the first war, when his gift of vision and speech runs out (Mahābhārata
10.9.58). At the end of this war, he gives Gāndhārī voice in a battlefield turned
into a terrible cemetery (11.16.1–4) and Kṛṣṇa gives the dying Bhīṣma strength
and the divine eye of knowledge so that he can pass his wisdom to Yudiṣṭhira
10 The possible exception of Balarāma requires a specific treatment: see Andreas Bigger,
Balarāma im Mahābhārata: seine Darstellung im Rahmen des Textes und seiner Entwick-
lung (Wiesbaden: Harrassowitz, 1998), 44–51 for the main texts.
11 See Bruce M. Sullivan, Kṛṣṇa Dvaipāyana Vyāsa and the Mahābhārata: A New Interpreta-
tion (Leiden: E.J. Brill, 1990), 50–51; 57ff. for this aspect and Vyāsa’s activities.
214 Wulff Alonso
(12.52.15–21). We will return to this point, but it should be stressed now that it
is not accidental, as they both represent the two essential groups forming the
celestial forces, great Ṛṣis, and gods, although Kṛṣṇa is much more than that.
Kṛṣṇa is conceived as the reincarnation of the God of gods, Viṣṇu, which marks
one of the few doctrinal differences between the two sides: the Pāṇḍavas all
recognize Kṛṣṇa as such and act accordingly, but Duryodhana, the Kaurava
leader, neither recognizes Kṛṣṇa nor acts in a way that would suggest such rec-
ognition, even though a number of his followers do.
8. The purpose, the secret of the gods, is performed as a collective endeavor
by both great Ṛṣis and gods. It involves creating the characters, as their birth
itself is part of the plan, as a mystery of the gods (rahasyaṃ khalu idaṃ rājan
devānām iti naḥ śrutam; Mahābhārata 1.58.3), which applies both to their birth
itself and to the processes leading up to it in which the two groups take part,
sometimes together. The great Ṛṣis contribute to their generation through rape
(Satyavatī), misleading boons (Kuntī), not so misleading boons (Gāndhārī),
curses (Bhīṣma), semen spilling (Droṇa) or directly (Dhṛtarāṣṭra, Pāndu,
Vidura). The gods of bhakti are also involved in the births of the characters
of the play, for example, resurrecting a dead child (Parikṣit and Kṛṣṇa) or
through boons (Śiva and Ambā-Śikhaṇḍin; Śiva and the five Pāndavas),
whereas conventional gods are more conventionally involved in generating
them (the Pāṇḍavas, Karṇa). All of them, not only Viṣṇu-Kṛṣṇa, are involved
in partial incarnations. They jointly contribute to the slaughter by giving
weapons to both parties: conventional gods (Indra and the Four Guardians of
the World, Agni), the bhakti gods (Kṛṣṇa and Śiva to Arjuna and Śiva to
Aśvatthāman), and the great Ṛṣis (Rāma Jāmadagnya to Droṇa and other
characters).
Blurring Indra
12 See, for example, its use in the same verse Mahābhārata 13.41.16.
216 Wulff Alonso
story of Yayāti.13 Indra must also seek advice from his chaplain, in general (see
12.85, for example) and in moments of defeat to the Asura (12.124.18 ff.). It is
easy to understand why the author/s of the Mahābhārata claim that Indra
bows to Brahmandom (133.2).
The background of the question is obviously larger and can be clearly
observed when, during their first exile in Book 1, the Pāṇḍavas are presented as
helpless without a chaplain (Mahābhārata 1.159). Similarly, in another story,
the king Mucukunda blames his court priest Vasiṣṭha for his loss to Kubera’s
Rākṣasas; the seer, after severe penances, then drives them away and Mucu
kunda defends his dependence on the Brahman’s power to Kubera (12.75).
Indra’s power is continually and systematically challenged, threatened by
the great Ṛṣis but also by mere humans who use, for example, asceticism,14 the
instrument of accumulation of power par excellence. One of the most illustra-
tive stories in this regard is that of Kuśika, a king who practices awesome
austerities to obtain a powerful son, lord of the three worlds. Indra’s only
recourse is to become that child (Mahābhārata 12.49.1–6). When the authors
write this story, they emphasize that this is the same Indra who had defeated
the Asura Pāka, which recalls a highly meaningful parliament of Indra that the
poets introduce in the story of Nahuṣa, when after conquering Triśiras with
difficulty, Indra is defeated by Vṛtra and, powerless, states: “nothing can stand
up to him. In older times I was capable of doing it, but now I am impotent”
(5.10.1–2; Van Buitenen trans.). The question is not simply the obvious fact that
the new Indra is no longer the old Vedic Indra: the authors make the Vedic
Indra aware of now being the powerless new Indra.
In this sense, this story is directly connected to the very well-known story of
Nahuṣa (Mahābhārata 5.11–17), which provides another instructive example of
the concentration of elements that reflects this weakened Indra, as is to be
expected in a narrative that begins referring to the many sufferings of Indra
and his wife. This narrative is essential because it shows, as some of the previ-
ous examples do, the Mahābhārata author/s’ use of the old mythic legacy, for
example, when the intervention of Viṣṇu is needed for the defeat of Vṛtra, and
13 Mahābhārata 1.71 ff.; see 1.73.2–4: all the gods say to Indra, the god of the hundred sacri-
fices, that it is the time for him, Sacker of Cities, to destroy the enemy. Immediately he
changes into a breeze and mixes up the clothes of a group of girls to sow discord between
them.
14 See Monika Shee, Tapas und Tapasvin in den erzahlenden Partien des Mahābhārata (Rein-
bek: Verlag für orientalistische Fachpublikationen, 1986) for its value and narrative uses.
Supernatural Conflicts, Unanimities, and Indra 217
Viṣṇu is proclaimed as sovereign of the gods, the one who made Śakra the over-
lord of the gods after the killing of the Asura Bali.15
Everything is notable in Nahuṣa’s story, including Indra’s inability to face
Nahuṣa, a death king who has attained Indra’s heaven by his own merits, when
he becomes master of the world in his absence. Note that Nahuṣa justifies his
pretention of sleeping with Indra’s wife, arguing that Indra had committed
abuses, such as the violation of Ahalyā, the wife of Gautama, among many
other deceptions, lawless deeds, and cruelties, without being stopped by the
gods (Mahābhārata 5.12.5–7). The great Ṛṣis, with the mediation of Viṣṇu, rid
the world of Nahuṣa and reinstate Indra.
It is even more meaningful that Indra dwells concealed in the waters
far away out of shame and brahmanicide, that the death of an old Vedic
enemy becomes brahmanicide and that this crime must be expiated by this
strange king of the gods. Viṣṇu purifies him, and Indra must sacrifice for him.
Viṣṇu also reassures the gods prophesying the end of Nahuṣa (Mahābhārata
5.13.10 ff.).
Not incidentally, after the war, Vyāsa recommends to Yudisthira a purifica-
tion by sacrifices, as Indra had performed after beating Pāka and his enemies
(Mahābhārata 12.34.27–28; 34). We speak of gods, including Indra, conceived as
Kṣatriyas, as is explicitly stated: the first god created the divinities, and they
have succeeded by behaving according to Kṣatra law.16 Dependent as they are
on his chaplains, clearly distinct from them and limited by brahmanic power,
the gods are increasingly close to men and Kṣatriyas.17 The great Ṛṣis, like the
gods of bhakti or Brahmā, leave them the necessary but inferior task of
governing.
The very power of Indra over the gods, as noted many years ago by Holtz-
mann,18 not only depends on his martial virility but also on sacrifices (Mahā
bhārata 12.20.11; see also 14.94.4); his cultivation of the virtues of patience,
self-control, restraint of the senses (5.29.12); his positive attitude toward
in a human womb and the corresponding death. The first story, which is quite
traditional and even quite similar in appearance to Greek epics and Gilgamesh,
receives new meaning in such a context, just as the position of human beings
does.
Death also becomes another component of impairment for conventional
gods: in a context where there is a beyond for human beings after death and
there are different paths to the beyond that leave behind the very palaces of
the gods, the gods become passengers on the same airline but on more
advanced flights. If all gods without exception must be reborn as mortals and
mortals as gods (Mahābhārata 12.250.40), it is easy to understand the hermit
Mudgala’s doubts (247). After terrible austerities and a disturbing visit from
Durvāsas, the envoy of the gods informs him that he has earned heaven. He
asks him for information about heaven, which of course includes Indra’s
heaven and the Brahmā’s heaven beyond it, and refuses to go with him at the
prospect of falling back down again. Extinction, nirvāṇa, is the only worth-
while travel.
In another twist, even the old enemies, the Asuras, can be aware of the real-
ity of the world and its mutability. In a fascinating story (see Mahābhārata
12.216–7; 12.220; and more specifically 12.217.54 and 58; 12.220.41), after defeat-
ing Bali, Indra asks Brahmā where to find Bali to scoff at him. Although Brahmā
attempts to discourage Indra, the god looks for and finds him in the form of an
animal, and Bali answers his sarcasm, reminding Indra of the mutability of
everything and the role of fate, pointing out that many thousands of Indras
have passed, all powerful, and that he too shall pass.
There is no need to wait so long to see the dangers that lurk and new multi-
plications of Indras. In another story (Mahābhārata 1.27.5 ff.), a group of tiny
ascetics, the Vālakhilyas, feel offended by Indra for a minor issue and, out of
revenge, make a sacrifice to obtain a new Indra to all the gods. Indra becomes
very upset and appeals to Kaśyapa. The seer appeases them, stating that “this
Indra has been made lord of the Three Worlds at the order of Brahmā” (1.27.18;
Van Buitenen trans.) and suggests the solution of the birth of an “Indra of the
birds,” Garuḍa. Not surprisingly, Garuḍa overcomes Indra but not Viṣṇu some
lines after (1.28–29).
A framework of power where this situation can be imagined or where
Aurva, a child of the Bhārgava family, can be presented as newborn and about
to destroy the worlds by his anger at the killing of his family (Mahābhārata
1.170.10 ff.), in a world where the major forces are great Ṛṣis and bhakti gods,
the king of these kṣatriyized gods could not be but a nominal king, the king
of a fortress rather gradually abandoned as besieged. The undercurrents that
220 Wulff Alonso
undermine the building while constructing another with its ruins and other
materials have reduced these gods and their king to this precarious position.
The contrast between the Indra of the main story and the Indra of the upākhyā
nas is evident. At the same time, these Indras are perfectly compatible, first if
we take account of the different requirements of the script in both sections of
the work. The main story emphasizes the celestial forces unit, and internal
conflicts are not explored. As mentioned, the main story is not characterized
by direct collisions between Brahmans and great Ṛṣis, or gods and kings, or
gods and bhakti gods but by clashes between humans. Nothing should disturb
their slaughter. Of course, there is enough room for some exceptions and even
more for confrontations of men, in particular the Pāṇḍavas, with beings that
represent, in principle, the evil to overcome, such as Asuras and Rākṣasas, with
more ambiguous creatures, such as snakes, or with semi-divine and positive
beings, such as the Gandharvas. The Gandharvas finally blurr the boundaries
between hostility and friendship in the Pāṇḍavas’ case, but not in the case of
the Kaurava’s.
The few clashes between men and gods mark the direction in which hierar-
chies work. When Arjuna faces Śiva (Mahābhārata 3.40–1; 3.163), he does not
know who is he, and the sense of this story is the recognition of the inferiority
of the human and the granting of a terrible weapon for the slaughter of other
humans by the God, a weapon with which conventional gods and even Indra
are not familiar. The story of the confrontation of Arjuna with Indra himself in
the Khāṇḍava Forest (1.214–25) makes clear that Arjuna and Kṛṣṇa, Nara and
Nārāyaṇa, can again defeat conventional gods and even sets of gods, including
Indra. Finally, when Duryodhana pretends to kidnap Kṛṣṇa in the assembly
hall of Hāstinapura (5.129), Kṛṣṇa shows his divine form and all the gods,
Pāṇḍavas, and Vṛṣṇis appear as part of it, producing terror in his enemies. Of
course, Indra and Brahmā are but parts of that mystical body of the God who
is the Unmade Maker. It is worth remembering that Duryodhana also openly
despises the advice of the great Ṛṣis and receives their recriminations.20
20 The two clashes of Bhīma with Kubera’s familiars in Book 3 serve a specific function: they
negatively mark Bhīma as someone who does not know the limits between the human
and the divine spheres, as they mark Draupadī, who sends him, all in contrast with the
respectful Yudiṣṭhira’s behavior.
Supernatural Conflicts, Unanimities, and Indra 221
Even clashes between gods and Asuras are carried out by humans in which
they incarnate ad maiorem gloriam interfectionis. The poet/s of the Mahābhā
rata are interested in showing how celestial forces deal with two aims: the
general slaughter and the relative victory of the Kṣatriyas representing the
forces of dharma, in that way establishing the basis and foundation of the
future monarchy for the troubled times of Kaliyuga. A king of the celestial
forces is needed, by definition, as are great Ṛṣis-Brahmans and harmony
between them. They are not, however, interested in hiding his condition of
nominal king. Not coincidentally, Vyāsa, other great Ṛṣis, and Kṛṣṇa lead the
implementation process.
The narrative strategy that focuses on the killing of humans and in super-
natural consensus tends to shift internal conflicts, including the questioning of
the power of Indra and the debility of traditional gods, towards the upākhyā
nas; however, the weakness—the character almost nominal of his/their
power—is also visible. Two curses of great Ṛṣis on the gods producing the birth
of two main protagonists of the story show the continuity of the underlying
questions: the case of Dharma sentenced to rebirth as Vidura (Mahābhārata
1.101), and the Vasus convicted of rebirth as Bhīṣma and his brothers (1.93). The
case we have already observed of the five Indras, doomed to be reborn as the
five Pāṇḍavas, leads us back to the other great expression of power—a god of
bhakti, Śiva.
The poets of the Mahābhārata concentrate on conflicts and clashes in the
upākhyānas, which are merely mentioned or suggested in the main story.
However, we must not forget that it is precisely the juxtaposition of the main
story with the entire set of upākhyānas, not ignoring the wisdom sections,
which make the whole work and achieve its effect on the reader or listener.
This juxtaposition also, among other things, leads the reader to see the fragile
architecture of power in the main story and to conclude that neither great Ṛṣis
nor bhakti gods are interested in temporal power and that both are necessary
nuisances to be handed over to other inferior actors.
In contrast, the upākhyānas allow a rich and contradictory deployment of
imagination, ideological positions, and contrasts. It is a forest of stories in which
the protagonists and their categories, conflicts, confrontations, and narrative
resources are multiplied and explored. We are referring to confrontations or
contrasts of Asuras/gods; Asuras/great Ṛṣis; Asuras/Kṣatriyas; Asuras and great
Ṛṣis/gods, great Ṛṣis and bhakti gods; Brahmans or great Ṛṣis/gods; Brahmans
or great Ṛṣis/Kṣatriyas; Brahmans or great Ṛṣis/Brahmans or great Ṛṣis; older
Brahmans or Brahman fathers/younger Brahmans or sons; Ṛṣi learned in the
Vedas/ignorant of the Vedas; Brahman teacher/student; Brahman/his Brahman
brother/s; conventional god/conventional god; conventional god/bhakti god;
222 Wulff Alonso
bhakti god/Asura; bhakti god/great Ṛṣis; Kṣatriya and Śiva/great Ṛṣi; Brahman/
members of other castes and groups (hunters, housewives, Vaiśyas), Brahmans’
or great Ṛṣis’ ancestors/celibate Brahmans; ancestors of royal lineages and
inhabitants of heavens/gods. We have limited this list to male protagonists but
have not forgotten the number of stories in which the protagonists are female:
Brahmans, Kṣatriyas, ascetics, Asuras, goddesses, and the strong presence
of conflicts related to their offspring, between members of the different cat-
egories of beings involved over females, or conflicts between ascetic woman/
ascetic Kṣatriya; ascetic woman/Garuḍa, amongst others. There is room for
other protagonists, for example, different animals or trees, and the wind.
Few critics have failed to see that a crucial—perhaps the most crucial—ele-
ment is the conflict between the Brahmans and Kṣatriyas, which is still more
evident if we consider that this conflict is also projected as clashes between
conventional, kṣatriyized gods/great Ṛṣis or, more clearly, great Ṛṣis/Kṣatriyas
(Viśvāmitra/Vasiṣṭha, for example). We have merely pointed to its presence in
the stories of Dharma-Vidura and the Vasus-Bhīṣma, which are the type of
upākhyānas that are more directly connected to the main story. Contrasts, con-
tradictions, and competitions between Brahmans and Kṣatriyas do not imply
doubts about the fundamental strategic message involved: the need for col-
laboration between them, but quite the opposite, in the same way that stories
about the contradictions between asceticism and its values and the continuity
of Brahman or Kṣatriyas stocks do not question asceticism, Brahmans or
Kṣatriyas. In fact, the main function of Greek mythology is probably to rein-
force the difference between and the complementarity of the two basic
categories involved in its stories: human beings and gods. The world’ s order is
reinforced by contrasts and hubris is a perfect instrument for doing it
As noted, consensus among these groups dominates the main story: great
Ṛṣis and gods are on the same side, as are ascetics and Brahmans. They are all
worshipped by the two sides, though the inevitable presence and ritual activi-
ties of Brahmans in the Kaurava’s court allows us to see slight counterpoints,
such as the refusal of the Brahmans to officiate at the agnihotra after the abuses
suffered by Draupadī (Mahābhārata 2.72.20). In fact, the Asuras’ instruments,
the Kauravas, are not aware that they are their minions. Indra’s theme is not,
therefore, the only one in which the requirements of the script lead to a differ-
ent treatment between the main story and the upākhyānas.
These points can be reinforced if we consider the mortal sphere. Again, the
rule of consensus does not mean that the problem of the relationship between
kings and Brahmans is not essential. It is clear that the issue of the contrast
between them appears, but hardly as a direct conflict. When Duryodhana
despises the great Ṛṣis’ counsels, he also despises his mother’s, the court
Supernatural Conflicts, Unanimities, and Indra 223
main story but in this characterization of some of its most essential actors. The
Dharma King’s learning is linked to moderate his Brahmanical components
in present times, times of Kaliyuga full of heretics, barbarians, and Yavanas,
invaders or inhabitants of the kingdom of King Dharma himself. Let me stress
again that the construction of these Brahman warriors is inseparable from their
end, their disappearance, and their defeats by Kṣatriyas. Their own words, the
words of their enemies and those of their ancestors point to their inappropri-
ate condition as warriors. The core message is clear enough: Brahmans in arms
must only exceptionally have room in the new order needed in the Kaliyuga
time of crisis.
Additionally, it is no coincidence that Rāma Jāmadagnya’s defeat, the great
Ṛṣi who marks the previous transition between yugas, takes place before the
war leading up to Kaliyuga and that the victor was Bhīṣma, who was at the
same time the first human responsible for the origin of the war at the begin-
ning of the story and responsible for transmitting the wisdom of government
to Yudiṣṭhira, the new King, at the end of it.
Kṣatriyas’ and Brahmans’ unity is advocated and can be viewed as the conti-
nuity of the unity of the Celestial Forces in the main story. The authors of the
work explore the tensions between them in different ways: the authors use
them to construct (or deconstruct) the characters they invent, including their
blurred old Indra, by letting them overflow in the upākhyānas or developing
them in a more theoretical way in the discourses of the wisdom sections.
We have attempted to argue that there exists a contrast between the role of
Indra in the architecture of power in the main story and in the substories of the
Mahābhārata. We have also defended the compatibility of this two presenta-
tions, taking into account the different roles that the authors of the Mahābhārata
assigned to both sections of their work. We argue that in any case, Indra’s fragil-
ity as king of the gods is conspicuous. We also note that something similar can
be found in some of the most essential aspects of the work: the relationship
between Kṣatriyas and Brahmans, an essential aspect of the Vaiṣṇava sacral
monarchy proposed as a solution to the times of Kaliyuga.
We have noted how the stories and dialogues of the Mahābhārata project
religious elements, which allow us to grasp religious global rethinking and the
era that produced it. The trends undermining conventional gods of Vedic lin-
eage are obvious: ideas such as the universality of reincarnation, that ideal
final destination leaving behind the gods and their heavens; kalpas and yugas
Supernatural Conflicts, Unanimities, and Indra 225
and the image of a creator God who creates and recreates the world and the
very gods, and encompasses them all; ways of liberation that emulate or sur-
pass the merit of sacrifice (e.g., pilgrimage, asceticism, meditation, ethical
behavior), and, last but not least, devotion; or the fact that conventional gods
were depicted as needing each of these expedients to advance in the chain of
beings, as human beings or to maintain their power.
Some years ago, Shee noted how tapas was projected in the narrative sec-
tions of the Mahābhārata in the form of power, a prevailing and cumulative
instrument. I would note that one of its essential aspects is that tapas and all
such components (e.g., the god of bhakti, the kalpas, the path of incarnations)
come into play in the work as narrative resources. Substories do it much more
freely, concentrating and projecting these components in all directions, in
such a way that we could say that one of their roles is a more or less discrete
demolition of the traditional gods. A reader or listener could have asked after
reading or listening to them: What is Indra compared to ascetic power? What
is Indra compared to a real transcendent God? The old architecture of power,
in fact, the type of architecture of power present in the Gilgamesh epic or in
the Iliad has changed despite the necessary appearances. It is no longer a ques-
tion of Marduk succeeding Ea. We could say, as it were, that its demolition is
the message.
Into this construction flow the ever-changing trends and changes in Indian
religiosity, which can be viewed as a type of workaround that eventually
culminates in another articulation of the architecture of power, with the dis-
appearance of Indra and the conventional gods and the full exaltation of the
bhakti gods. However, it would be a mistake to confuse its place in time, in
the processes of change, with its own reality. It is a consistent system, a struc-
ture that makes sense by itself, although it does not correspond to the model
to which we are accustomed. It is not madness, and substories are part of its
method.
I have begun to mention the king’s role as articulator of the relationship
between the macrocosm and microcosm. When the work presents the offer of
a Vaiṣṇava king for the times of Kaliyuga, what appears is an impossible propo-
sition for the blurred Indra, a god, as we see, with no bhakti cults, no devotion,
no provisional beyond, and traditionally depicted as a character without the
new necessary virtues. He cannot meet the new demand, he cannot be the real
supernatural partner and protector of the new monarch. Although the work
has at least one story referring to his ritual role in this field (Mahābhārata
1.57.1–27), despite everything, the unfolding of the plot shows his limitations:
he gives King Vasu (or Uparicara) a chariot and a protective garland for battle
and, in particular, a bamboo pole for a festival of Indra at the end of the year, a
226 Wulff Alonso
ritual, the text says, continued into the present. But Indra does so because he is
worried that Uparicara, through his austerities, may aspire to become like him.
The frightened Indra is all-pervasive. The persistence of a link between monar-
chy, rituals, and Indra is inevitable, particularly if the rain holds as his sphere
of competence. Note that even in the case of this possible last sphere of activ-
ity, an ascetic (actually a very young ascetic), Ṛśyaśṛṅga, can force Indra to rain,
for fear of his curse. Indra, in the middle of all this humiliation, is called the
slayer of Bala and Vṛtra (balavṛtrahā; 3.110.3).21 King Māndhātar can also make
it rain after a twelve-year drought, whereas Indra, described at the time as the
Thunderbolt-wielder (vajrapaṇinaḥ; 3.126.39) just looks on. The good relation-
ship between Indra and the king does not erase the fact that the king does not
seem to need him to do so, which corresponds to the fact that he was born of a
male after the intervention of a Bhārgava, who had promised his father a son
“possessed of the might of austerities” who could even kill Indra (126. 20, Van
Buitenen trans.). In fact, this specialization of Indra places him in a subordi-
nated position, just as his condition as member of the group of the four World
Guardians, the lokapālas, the gods of the four points of the compass, frequent
in the Mahābhārata and quite popular in the Purāṇas. A nominal king of the
Three Worlds whose power is just a grant given by others simultaneously can
be a provincial governor and can change his position for the worse to become
only a provincial governor before practically disappearing. There is no safe
space for gods like him, no mysterium tremendum around such a god full of
metus tremendus.
The Mahābhārata is a witness in his fall and, simultaneously, one of his exe-
cutioners. In fact, the construction of this story makes sense precisely because
there is a god of bhakti. Clearly, “this grotesque butchery” that “would not have
happened” (Mahābhārata 11.27.20; Fitzgerald trans.), as Yudhiṣṭhira says at the
end of the slaughter, generates ideological and religious problems of para-
mount importance,22 for example, the meaning or just the possibility of
human action before the inevitability of fate and the divine plan or what to do
with the terrible image of the God who directs all this desolation and slaugh-
ter—these are feelings that are conveyed well by the Strīparvan. Nothing in
Indra suggests that a god like him could have been adapted to providing the
solutions offered by Kṛṣṇa, such as devotion or to provide a solid background
for a transcendent monarchy. Indra’s offer of a heaven in a string of heavens
21 Incidentally, like Enkidu, he is brought from the jungle to the city by a prostitute.
22 See, for example, Julian Woods, Destiny and Human Initiative in the Mahābhārata (Albany:
State University of New York Press, 2001) for some of the problems involved.
Supernatural Conflicts, Unanimities, and Indra 227
If they need Indra, Indra will need Viṣṇu to defeat Vṛtra—and Brahmā’s
counsel and the bones of the great Ṛṣi Dadhīca (Mahābhārata 3.98–9)—if it is
useful to allow him to be in charge of his nominal monarchy presiding over the
unanimous celestial forces, a great Ṛṣi and Kṛṣṇa direct actions and characters;
even his heaven, after all the heaven of the old warrior god, can be useful for
the dead warriors, a path well opened by the Upaniṣads. He was useful and
could continue his existence by becoming the figurehead of a ship that he
never steered and, at the same time, being ridiculed even in his own sex.
It is not a surprise that these new perspectives were accompanied in the
wisdom sections of the work with exhortations to implement an open and cre-
ative use of the old legacy, reinterpreting, without dogmatism, religious
knowledge and practices, including the old Vedic animal sacrifice, and reli-
gious (and non-religious) books. The entire story is intended to articulate the
popular dogma, but upākhyānas and wisdom sections, because of their greater
variability, may have served more particularly to vary, enrich and even correct
possible interpretations of the main story, to increase focus, integrate cults and
religious practices, and to attract people, allowing open debate among the
potential members of that front, and to offer more possibilities for adapting its
performance to different audiences or readers.25 Let me qualify a previous sen-
tence: it is not madness, and substories are part of its method.
I have attempted to suggest some perspectives that may explain differences
between the pantheon and the power dynamics of the substories and main
stories of the Mahābhārata, focusing, in particular, on Indra. So far, I have
avoided addressing controversies with Hopkins’s traditional perspective. It is
fortunate that we can follow it in three different works produced over close
to a century: Adolf Holtzmann in 1878, E. Washburn Hopkins in 1915 and John
Brockington in 2001.26 All three are interesting works, as they collect many
25 See James W. Laine, Visions of God: Narratives of Theophany in the Mahābhārata (Vienna:
Institut für Indologie der Universität Wien, 1989), 259–60 for his description of “a society
of competing ideologies” in “which the traditions that survived where those that gained
credibility and legitimacy by absorbing into a grand hierarchy all those possibilities that
other religions had to offer.” See the interesting remarks of Andrew J. Nicholson, Unifying
Hinduism: Philosophy and Identity in Indian Intellectual History (New York: Columbia Uni-
versity Press, 2010), 185, on the concept of “inclusivism” as a tool of assimilation, “a process
by which a multitude of various sects, philosophies, gods and modes of worship are
united under a single overarching concept”; and see also the remarks on the existence of
other parallel strategy, negation and even demonization in 185 ff., and 186, 195, 200–201,
203 in particular.
26 See Holtzmann, “Indra nach den Vorstellungen des Mahābhārata”; Edward Washburn
Hopkins, Epic Mythology (Strassburg: Trübner, 1915), particularly 122–41; and John L.
Supernatural Conflicts, Unanimities, and Indra 229
of the important texts, and I believe they are based on the same principles:
(1) The Mahābhārata reflects a Vedic Indra corresponding to the first phases (or
layers) of the work, the oldest ones; (2) However, newer layers can be observed
through analysis of the different treatments of Indra, arriving at the last stages
(or layers) when Indra becomes subordinate to Brahmā, Viṣṇu, or Śiva; (3) The
Vedic Indra is depicted as dominant, which would demonstrate the survival
of the oldest layer and Indra’s role in rituals, for example. Epithets reflecting
Indra as conqueror of his enemies, slayer of Bala and Vṛtra, or Thunderbolt-
wielder, would demonstrate his dominance.
Analyzing these elements and their variants is tempting. I am thinking, for
example, of the ideas of Holtzmann27 on the first Indra, the almighty god of
the heroic age (Heldenzeit) of the Indian people, a Hero-god (Heldengott) who
was a Nature-god (Naturgott) based on the storms. Also of note is his evolution
to another Indra when in a superior state, the growing speculative capacities of
people (erwachende speculative Geist des Volkes) perceived divine unity in the
multiplicity of the forces of nature and, therefore, transformed the divine
sphere into a more moral one in an age already dominated by priests and rep-
resented by Brahmā.
However, it is more useful to simply note that the first two proposals are
highly speculative (and have not been tested on the text) and that they start
from an unproven hypothesis, that is, the work is the result of the reworking of
an old epic core. There is no evidence of other epics that are contemporary to
this supposed old epic, and there is no sound evidence of the existence of epics
that are contemporary to the Mahābhārata we have, except the also splendidly
isolated Rāmāyaṇa—and let us remember that only new epics (for example,
the Aeneid) can appear in such an isolated way.
The third proposal is more than questionable. As we have seen before, the
presence of elements referring to the old Vedic Indra cannot let us forget that
even in the main story, Indra is not the real leader of the Celestial Forces, that
in the upākhyānas he is quite systematically derided, and that it is possible to
individuate the processes and forces at work that lead to his demolition with
the very internal evidence of the work.
From this perspective, it is easy to understand that some of their specific
arguments are untenable. The idea, for example, that epic epithets referring to
Indra would show the survival of the old conceptions can be perfectly coun-
tered with the evidence that, as we have observed, epic epithets appear
everywhere, even where Indra is ridiculed, and one of them (the thousand-
eyed Indra) is directly subject to ridicule. Obviously, such an epic creation had
to deal with Vedic hymns and stories and to reuse and recreate its useful parts.
It is safer to conclude that the prominence of certain such uses demonstrates
that there were no intermediate stages or other available epics or materials to
draw on. The same could be said about formulae such as “x attacked y like
Indra the Danavas.”28
Vedic Indra and Vedic myths were much more than a precedent that
accorded authority and prestige.29 They had to be used and developed. Epi
thets, formulae or the very myths can be seen as available materials, and the
supposed layers can be seen as different operational modalities, depending on
the authors’ interests in general and on the different parts of their work. This
idea is clearly consistent with the flexible use of the old heritage advocated
in doctrinal and religious matters and with the fact that the Mahābhārata’s
authors defended the superiority of their work on the Vedas.
The three mentioned authors aim to analyze the supposed different stages
in the work, without analyzing the work itself—the work as a self-contained
construction, as a whole to be interpreted, not mutilated or denied.
The options for use and reworking of the components available to the
authors of the Mahābhārata are certainly plentiful, but it is useful to remember
that they are not as many options as those of the Buddhists. Hopkins notes,30
for example, that “in Buddhistic narratives the excellence of a virtuous person
“makes hot the throne of Indra.” The epics have no such absurd figure.” We
must ask: Absurd to whom? Absurd for what? Certainly it is not easy to imag-
ine such formulae in the Epics (although it is enjoyable to imagine them) and
28 To end these notes, I can not but quote here an eighty-year old criticism to the position of
those authors who defended “a distinction based on differing conceptions of the gods
between the earlier and later parts,” noting that: “I make no attempt at a criticism …. Their
widely differing results cannot be considered reliable. I shall merely add one remark con-
cerning Dr. Finsler's method, because it is of most importance for my purpose. His prem-
iss is that certain descriptions of the gods are later than others and he judges the age of
various passages according to this presumption. If this premiss is questionable, he is
working with a petitio principii.” Martin P. Nilsson criticizing these premises on the Iliad
in Martin P. Nilsson, The Mycenaean Origin of Greek Mythology (Berkeley, Los Angeles,
London: University of California Press, 1932), 223–24, 228.
29 Danielle Feller, The Sanskrit Epics: Representation of Vedic Myths (New Delhi: Motilal
Banarsidass, 2004), 299–300.
30 Hopkins, Epic Mythology, 139.
Supernatural Conflicts, Unanimities, and Indra 231
not just for stylistic reasons, though the image of Indra in a permanent state of
alert of possibly losing his throne is paramount.
It is good to take this reference to indicate that the need for absorbing out-
side influences while opposing them, which largely explains the very existence
of the Mahābhārata, also involves Indra. Using Indra or not using Indra is not
a real option, because, without ever forgetting the conventional Vedic tradi-
tion, he was widely used in less conventional perspectives. So far as I know, the
first uses of the iconography of Indra occur on coins of the Greek kings in the
northwest of the subcontinent, producing syncretic representations of Indra-
Zeus, and the second uses are in Buddhist art.31 It would be tempting to follow
the Greek path. It suffices to cite one short example of a story we already know,
the story of Vipula. The characterization of Indra in this story as a god who
assumes different forms for his seductions (e.g., holding the thunderbolt, or as
a bird, a quadruped, the wind, a mortal, or invisible; Mahābhārata 13.40.28 ff.),32
arises the difficult problem of the Indian sources used,33 but it is obvious enough
that this problem does not exist in the case of the Greek Zeus. Iconography and
texts explore the same fields.34
31 Leona Anderson, Śakra in early Buddhist Art, MA Thesis, McMaster University, 1978, 18–19,
<http://digitalcommons.mcmaster.ca/opendissertation5624>.
32 See Fernando Wulff Alonso, “Indra en femenino: notas entre la India y Grecia a propósito
del Mahābhārata,” in Δῶρον Μνημοσύνης. Miscelánea de Estudios ofrecidos a Ma. Ángeles
Durán López,, ed. Aurelio Pérez Jiménez and Isabel Calero (Zaragoza: Libros Pórtico, 2011),
347–70.
33 Note Renate Söhnen-Thieme, “The Ahalyā Story Through the Ages,” in Myth and Myth-
making: Continuous Evolution in Indian Tradition, ed. Julia Leslie (Richmond, Surrey: Cur-
zon Press, 1996), 39: “If Indra is imagined as a “womanizer” in Indian literature, one may
well ask who were the women who were his prey … evidently there is only this one
instance [Ahalyā] and it may be worth while look at it more closely.” Although a very few
more examples could be added, the fact is that they are scanty and have nothing to do
with these alleged seductions.
34 I have avoided making overly explicit reference to my hypothesis on the use of Greco-
Roman materials in the Mahābhārata. However, a brief note could be useful here. The
Zeus of Socrates and Plato cannot be the Zeus of mythology, because their concept of
reincarnation entails ethics, a moral compass: a god who perpetrates crimes, sexual
assaults, or rapes, cannot preside over a world ruled by morality, where, after death,
humans are rewarded or punished in direct relation to their previous lives. I have sug-
gested that is one part of the components undermining Indra after the introduction of
reincarnation into the Vedic and Brahmanic tradition, in a parallel but separate process.
The Mahābhārata’s use, and recreation of Greek mythological tradition relating to Zeus
may have resulted—perhaps deliberately—in undermining Indra even further.
232 Wulff Alonso
without stories that, for example, explain the origins of characters and so many
other issues having clear developments in the work.37
It is fair to say that this ability to ignore substories has very interesting prec-
edents in the Iliad academy: for example, the tendency to deny no less a
reference than the Judgment of Paris as the origin of Hera’s and Athena’s hatred
of the Trojans (Iliad 24.25–30). Nevertheless, it is perhaps more surprising to
see that a clear reference to the warriors’ slaughter in Troy, identifying it with
the end of the race of half-god men (ἡμιθέων γένος ἀνδρῶν; 12.23), can be forgot-
ten and not connected to the beginning of the work, which directly associates
the killing of so many heroes with the fulfillment of the plan of Zeus (πολλὰς δ’
̈
ἰφθίμους ψυχὰς Ἄϊδι προί�αψεν ἡρώων … Διὸς δ’ ἐτελείετο βουλή; 1.3–5). Perhaps the
idea of a transparent Iliad based on the idea of a no less rational and transpar-
ent Greek religion was not compatible with a merciless and fatalistic image
projected on a work so directly connected to such “ancestors.” In my opinion,
the Mahābhārata’s authors did not make such a mistake.
References
Primary Sources
Hesiod. Theogonia, Opera et Dies, Scutum. Edited by Friedrich Solmsen. Fragmenta
Selecta. Edited by R. Merkelbach M.L. West. Oxford: Oxford University Press, 1970.
The Mahābhārata for the First Time Critically Edited. Edited by Vishnu S. Sukthankar,
Shripad K. Belvalkar, Parashuram L. Vaidya, et al. 19 vols. Poona: Bhandarkar Oriental
Research Institute, 1933–66.
The Mahābhārata, vol. 1. 1. The Book of the Beginning. Translated by J.A.B. van Buitenen.
Chicago: University of Chicago Press, 1973.
The Mahābhārata, vol. 2. 2. The Book of the Assembly Hall; 3. The Book of the Forest.
Translated by J.A.B. van Buitenen. Chicago: University of Chicago Press, 1975.
The Mahābhārata, vol. 3. 4. The Book of Virāṭa; 5. The Book of the Effort. Translated by
J.A.B. van Buitenen. Chicago: University of Chicago Press, 1978.
The Mahābhārata, vol. 7. 11. The Book of Women. 12. The Book of Peace, Part 1. Translated
by James L. Fitzgerald. Chicago: University of Chicago Press, 2004.
The Mahabharata of Krishna-Dwaipayana Vyasa Translated into English from the Original
Sanskrit Text. Translated by Kisari Mohan Ganguli. 5th ed. New Delhi: Munshiram
Manoharlal Publishers, 1990–91.
Poetae Epici Graeci. Testimonia et Fragmenta. Edited by Alberto Bernabé. Leipzig:
Teubner, 1988.
Secondary Sources
Adluri, Vishwa. “The Critical Edition and Its Critics: A Retrospective of Mahābhārata
Scholarship.” Journal of Vaishnava Studies 19, no. 2 (2011): 1–21.
———, ed. Ways and Reasons for Thinking about the Mahābhārata as a Whole. Pune:
Bhandarkar Oriental Institute, 2013.
Adluri, Vishwa, and Joydeep Bagchee. “Introduction.” In Reading the Fifth Veda: Studies
on the Mahābhārata, Essays by Alf Hiltebeitel, vol. 1, edited by Vishwa Adluri and
Joydeep Bagchee, xi–xxxvi. Leiden: E.J. Brill, 2011.
———. The Nay Science: A History of German Indology. New York: Oxford University
Press, 2014.
Anderson, Leona. Śakra in Early Buddhist Art. MA Thesis. Hamilton: McMaster University,
1978. <http://digitalcommons.mcmaster.ca/opendissertation5624>.
Biardeau, Madeleine. Le Mahābhārata: Un récit fondateur du brahmanisme et son inter-
prétation, 2 vols. Paris: Éditions du Seuil, 2002.
Bigger, Andreas. Balarāma im Mahābhārata. Seine Darstellung im Rahmen des Textes
und seiner Entwicklung. Wiesbaden: Harrassowitz Verlag, 1998.
Brockington, John L. “Indra in the Epics.” In Vidyārṇavavandanam. Essays in Honour of
Asko Parpola, edited by Klaus Karttunen and Petteri Koskikallio, 67–82. Helsinki:
Finnish Oriental Society, 2001.
———. The Sanskrit Epics. Leiden: E.J. Brill, 1998.
Burkert, Walter. The Orientalizing Revolution. Near Eastern Influence on Greek Culture in
the Early Archaic Age. Cambridge, MA and London: Harvard University Press, 1995.
Detienne, Marcel. L’invention de la mythologie. Paris: Gallimard, 1981.
Feller, Danielle. The Sanskrit Epics: Representation of Vedic Myths. New Delhi: Motilal
Banarsidass, 2004.
Fitzgerald, James L. “Mahābhārata”. In The Hindu World, edited by Sushil Mittal and
Gene Thursby, 53–74. New York: Routledge, 2004.
Hiltebeitel, Alf. Dharma: Its Early History in Law, Religion, and Narrative. New York:
Oxford University Press, 2011.
———. “The Mahābhārata and the Stories Some People Tell about It—Part 1.” Exemplar:
The Journal of South Asian Studies 1, no. 2 (2012): 2–26.
———. “The Mahābhārata and the Stories Some People Tell about It—Part 2.” Exemplar:
The Journal of South Asian Studies 2, no. 1 (2013): 1–14.
———. “Not Without Subtales: Telling Laws and Truths in the Sanskrit Epics.” Journal
of Indian Philosophy 33, no. 4 (2005): 455–511.
———. Reading the Fifth Veda: Studies on the Mahābhārata. Essays by Alf Hiltebeitel,
vol. 1. Edited by Vishwa Adluri and Joydeep Bagchee. Leiden: E.J. Brill, 2011.
———. Rethinking the Mahābhārata: A Reader’s Guide to the Education of the Dharma
King. Chicago: University of Chicago Press, 2001.
Supernatural Conflicts, Unanimities, and Indra 235
———. The Ritual of Battle: Krishna in the Mahābhārata. Albany: State University of
New York Press, [1976] 1990.
———. “‘You Have to Read the Whole Thing’: Some Reflections on Madeleine Biardeau’s
Mahābhārata.” In Du texte au terrain, du terrain au texte: Dialogues interdisciplinaires
au tour du l’oeuvre de Madeleine Biardeau. Journée 2011 du Centre d’Études de l’Inde
et de l’Asie du Sud. Paris, 2011. <http://ceias.ehess.fr/docannexe.php?id=1862>.
Holtzmann, Adolf. “Indra nach den Vorstellungen des Mahābhārata.” Zeitschrift der
Deutschen Morgendländischen Gesellschaft 32 (1878): 290–340.
Hopkins, Edward Washburn. Epic Mythology. Strassburg: Trübner, 1915.
Laine, James W. Visions of God: Narratives of Theophany in the Mahābhārata. Vienna:
Institut für Indologie der Universität Wien, 1989.
Nicholson, Andrew J. Unifying Hinduism: Philosophy and Identity in Indian Intellectual
History. New York: Columbia University Press, 2010.
Nilsson, Martin P. The Mycenaean Origin of Greek Mythology. Berkeley, Los Angeles,
London: University of California Press, 1932.
Shee, Monika. Tapas und Tapasvin in den erzahlenden Partien des Mahābhārata. Reinbek:
Verlag für orientalistische Fachpublikationen, 1986.
Söhnen-Thieme, Renate. “The Ahalyā Story Through the Ages.” In Myth and Mythmaking:
Continuous Evolution in Indian Tradition, edited by Julia Leslie, 39–62. Richmond,
Surrey: Curzon Press, 1996.
Sullivan, Bruce M. Kṛṣṇa Dvaipāyana Vyāsa and the Mahābhārata: A New Interpretation.
Leiden: E.J. Brill, 1990.
Woods, Julian. Destiny and Human Initiative in the Mahābhārata. Albany: State University
of New York Press, 2001.
Wulff Alonso, Fernando. “The Fourth Book of the Mahābhārata and its Greek Sources.”
In The Churning of the Epics and Purāṇas at the 15th World Sanskrit Conference, edited
by Simon Brodbeck, Alf Hiltebeitel, Adam Bowles, 71–95. New Delhi: Rashtriya
Sanskrit Sansthan and D.K. Printworld, in press.
———. Grecia en la India. El repertorio griego del Mahābhārata. Madrid: Akal Eds, 2008.
———. “Greek Sources in the Mahābhārata.” In Ways and Reasons for Thinking about
the Mahābhārata as a Whole, edited by Vishwa Adluri, 155–83. Pune: Bhandarkar
Oriental Research Institute, 2013.
———. “Heracles in the Mahābhārata.” Rivista degli Studi Orientali 71 (2008): 73–101.
———. ““Indra en femenino: notas entre la India y Grecia a propósito del Mahābhārata.”
In Δῶρον Μνημοσύνης. Miscelánea de Estudios ofrecidos a Ma. Ángeles Durán López,
edited by Aurelio Pérez Jiménez and Isabel Calero, 347–70. Zaragoza: Libros Pórtico,
2011.
———. The Mahābhārata and Greek Mythology. New Delhi: Motilal Barnasidass, 2014.
———. “Mitos, divinidades, reencarnación: entre Platón y la India”. In Realidad,
Fantasía, Interpretación: Funciones y pervivencia del Mito Griego, Estudios en Honor
236 Wulff Alonso
del Profesor Carlos García Gual, edited by Aurelio Pérez Jiménez, 571–85. Zaragoza:
Libros Pórtico, 2014.
———. El Peligro Infinito: Diosas, héroes y mujeres poderosas en cinco grandes épicas.
Madrid: Marcial Pons, 2015.
Chapter 8 237
Chapter 8
One of the most striking features of the Sanskrit Mahābhārata is the intricate
maze of subnarratives that courses through nearly every book. Its vast array of
side stories, digressions, and anecdotes make the epic appear as dense and
sprawling as a modern-day metropolis, and equally as impenetrable for its
modern readers. Like tourists visiting Tokyo, London, or Mumbai, we may tra-
verse the Mahābhārata’s broad central avenues with relative confidence and
gaze in wonder at its majestic skyscrapers, but few of us dare to venture into its
tiny bylanes and alleyways, lest we become hopelessly and irreparably lost. The
current generation of readers is fortunate to have an experienced guide in our
midst, who knows the epic’s streets by heart and who has steered us through its
more shadowy neighborhoods, helping us to appreciate their delicate and
extraordinary beauty along the way. More so than any other modern scholar,
Alf Hiltebeitel has insisted that we treat the Mahābhārata’s narrative complex-
ity not as a problem to be solved, not as calcification to be scraped away from
a more real core, but as a literary testament to the genius of its composers. He
has emboldened us to situate the epic’s composition in a specific time and
place—the post-Mauryan milieu of North India, “between the mid-second
century bc and the year zero”2—and to read the text as a coherent and mas-
terly written institution within an emergent early Indian public culture.
Hiltebeitel especially encourages us to read the Mahābhārata for its “sub-
tales” (upākhyānas) in order to appreciate the epic’s “archetypal design.”3 It is
still an open question, however, how this complicated design of embedded sto-
ries and speeches would have impacted the cultural lives of its early Indian
1 A preliminary version of this paper was presented in October 2012 at the Annual Conference
on South Asia in Madison, Wisconsin; I thank Vishwa Adluri for organizing the panels and the
present volume, as well as Alf Hiltebeitel, Robert and Sally Goldman, and other scholars in
attendance for comments and suggestions that have proven invaluable in its revision.
2 Alf Hiltebeitel, Rethinking the Mahābhārata: A Reader’s Guide to the Education of the
Dharma King (Chicago: University of Chicago Press, 2001), 18.
3 Alf Hiltebeitel, “Not Without Subtales: Telling Laws and Truths in the Sanskrit Epics,”
Journal of Indian Philosophy 33, no. 4 (2005): 479.
artifacts from what James Hegarty calls the “significant past.”12 A more thor-
ough investigation of this museological analogy is forthcoming, but I may
outline my basic principles of comparison here. Just as the layout, lighting, and
text panels enable the otherwise absent curator to regulate a museum-goer’s
appreciation of the value of the items on display, I suggest that the epic’s anony-
mous composers introduced a number of textual devices within the primary
narrative in order to control its audience’s reception of its embedded subnar-
ratives. First, the epic’s two outer frames establish a kind of discursive “lighting”
that shapes the appearance of the narrative materials contained within.13
Second, epic subnarratives are typically told within question-and-answer dia-
logues between sagacious figures and the Kaurava or Pāṇḍava princes; like the
“text panels” placed next to exhibits in a museum, these provide immediate
and authoritative commentary on the value of the subnarrative.14 Last but not
least, the unilinear arrangement of subnarratives induces intertextualities
between otherwise independent story material. Just as museum exhibits are
ordered through chronological or thematic sequences that subtly restrict how
a visitor evaluates them, subnarratives within the Mahābhārata have been
sequenced to create narrative trajectories that determine how its audiences
should interpret the nonstandard characters and events portrayed within
them. It is through such museic devices, I suggest, that the Mahābhārata came
to have a normative impact upon early Indian public life.
On the other hand, the Mahābhārata was not actually a museum, and
ultimately, the analogy must remain a heuristic device for jump-starting the
historical analysis of the text itself. How was this work of Sanskrit literature
15 On the relationship of “fixed” and “fluid” categories in Indian cultural traditions, see
Wendy Doniger, “Fixed and Fluid Texts in India.” Here, by invoking the “fluidity” of the
epic narrative tradition, I do not mean to argue against the validity of the critically recon-
structed text of the written Sanskrit Mahābhārata, but rather, as Franklin Edgerton had
acknowledged, the idea that “before the establishment of this text, the ancestor of all our
MSS., there were already different versions of the [Mahābhārata] stories.” Franklin
Edgerton, “Introduction,” in the Sabhāparvan (Poona, Bhandarkar Oriental Research
Institute, 1944), xxxvi. I thank Vishwa Adluri for this reference as well as a number of
other comments that have considerably strengthened this essay.
Pride and Prostitution 241
As we will see, this shocking story of pimping and promiscuity is not told in
isolation, but on the heels of two rather different stories, creating a three-part
narrative “installation.” Furthermore, the epic employs two distinct modes of
referentiality as a kind of “lighting” to illuminate the Gālavacarita. Its main
characters, we will find, are all Vedic personae, while the events take place at
important sites of post-Mauryan power. Finally, Nārada’s introductory and
concluding remarks work as “text panels” through which epic’s composers
sought directly to regulate the reception of the story. What exactly did they
want epic audiences to think? To foreshadow my conclusions, I will argue that
despite its seemingly racy contents, the epic’s Gālavacarita is ultimately a story-
exhibit meant for mainstream public audiences in post-Mauryan North India.
Within it, there appears a unique fusion of morality and political discourse, or
“dharmārtha,” that encouraged the consolidation of power through kindness
and friendship rather than despotism and conquest.
But first, what should we make of the exhibit itself? A number of feminist
historians have read the curious tale of Mādhavī’s prostitution for its implica-
tions regarding the subordination of women in early Indian society. For rather
different reasons, it also captured the attention of the comparative mytholo-
gist Georges Dumézil, who discovered it to be a narrative “relic” with a long
“pre-Indian” history. Both lines of inquiry have yielded insights into this
remarkable narrative that are well worth considering before embarking on our
“museulogical” approach.
16 Madhusraba Dasgupta, “Usable Women: The Tales of Ambā and Mādhavī,” in Faces of the
Feminine in Ancient, Medieval, and Modern India, ed. Mandakranta Bose (New York:
Oxford University Press, 2000), 53.
242 Sathaye
even the most patriarchal of texts, the dharmaśāstras, do not condone such
commodification within its idealistic social vision, and in many cases explicitly
prohibit it for high-caste women.23 There remains, therefore, a deeper cultural
ambivalence in the Mahābhārata surrounding Mādhavī’s strange polyandry—
just as in the more famous polyandrous marriage of Draupadī.24
Though she may ultimately be a pawn of the men around her, Mādhavī does
engage in one clear “act of autonomy” at the close of the narrative.25 After she
gives birth to four sons, and after Gālava is done with her, Yayāti arranges for a
marriage ceremony (svayaṃvara) in which his daughter is granted the right to
choose her own husband. But Mādhavī selects none of the suitors, and instead
enters the forest as a renunciant, choosing to live out her days “like a fawn
(mṛgavatī).”26 Ramaswamy reads this to be an essentially soteriological act,
for “Mādhavī’s salvation lies in her silence.”27 while Bhattacharji interprets it
as a form of social protest.28 Dasgupta splits the difference, explaining that
renunciation induces a “breach between Madhavi’s habitual compliance and
her new self-assertion,” and views the story to be “an ethical puzzle rather than
a straightforward model” for Brahmanical patriarchy.29
The Mahābhārata never makes the solution to this puzzle entirely clear.
Feller’s study shows that that the epic’s audiences, both male and female, would
certainly have found these events to be morally problematic. She proposes
therefore that it falls within the scope of the legal discourse on āpaddharma,
or “emergency law,” but I find this argument difficult to sustain. Indeed, a
number of legally viable options were open to Gālava, including simply beg-
ging for the horses from the three kings, as he had initially done in Yayāti’s
court. He wasn’t forced to prostitute Mādhavī. He also could have offered the
girl straightaway to Viśvāmitra, as his teacher himself quips. Furthermore, the
extraordinary circumstances necessary for invoking āpaddharma—the total
breakdown of moral order, the need for physical survival, or a clear and pres-
23 Danielle Feller, “The Strange Story of Princess Mādhavī,” paper presented at the Fifth
Dubrovnik International Conference on the Sanskrit Epics and Puranas, August 11–16,
2008.
24 Alf Hiltebeitel, The Ritual of Battle: Krishna in the Mahābhārata (Albany: State University
of New York Press, 1990 [1976]), 223.
25 Dasgupta, “Usable Women: The Tales of Ambā and Mādhavī,” 53.
26 Perhaps conflating it with the Ambā legend, Dasgupta speculates that Mādhavī had fallen
in love with Gālava, who did not reciprocate (ibid., 52); such a reading is not found in the
Critical Edition text.
27 Ramaswamy, “Gender and Transcendence in Early India,” 235.
28 Bhattacharji, “Economic Rights of Ancient Indian Women,” 511, col. 3
29 Dasgupta, “Usable Women: The Tales of Ambā and Mādhavī,” 53.
244 Sathaye
the kanyās exemplify how realworld women should not behave, for all of them
engage in quasi- or fully adulterous sexual relationships, think and speak for
themselves, and have fiery, self-assured personalities.
On the other hand, the Five Maidens are not criticized within the Hindu
tradition, but instead venerated as auspicious figures who “eliminate the great-
est of sins” (mahāpātakanāśakāḥ). In searching for the reasoning behind this
paradox, Bhattacharya rather unconvincingly argues that the kanyās represent
the Jungian Anima archetype.34 Simon Brodbeck, in reviewing Bhattacharya’s
work, demystifies the situation by pointing to “specific tensions between
matriliny and patriliny and between matrilocy and patrilocy” that run through
the stories of kanyā-types in the epics.35 In contrast to the satīs, the kanyās
represent an older “relic of a disused system” of women-centered marriage and
inheritance in which royal consorts were connected to the king’s lands and
fortune (Śrī).36 This system, Brodbeck notes, came into conflict with the patri-
lineal/patrilocal system of the dharmaśāstras. It is an observation that helps to
clarify why kanyā-types were significant for the Mahābhārata’s composers:
they exposed a misalignment between the preferred, patrilineal model of king-
ship and the alternate systems of certain older Kṣatriya lines that were being
incorporated into the Hindu mainstream at the time of the epic’s composition.
A “relic” of times gone by, the only option for Mādhavī—and the old matrilin-
eal/matrilocal system she represents—was to disappear.
A.K. Ramanujan (Berkeley, CA: University of California Press, 1986), 41–75; see Gloria
Goodwin Raheja and Ann Grodzins Gold, Listen to the Heron’s Words: Reimagining Gender
and Kinship in North India (Berkeley: University of California Press, 1994), 32–38.
34 That is to say, Bhattacharya posits them to be embodiments of a feminine inner personal-
ity, the Anima, embedded within the male subconscious, the Animus, which serves to
connect the (male) individual to a collective unconscious.
35 Simon Brodbeck, Review of Pancha-Kanya: The Five Virgins of Indian Epics. A Quest in
Search of Meaning, by Pradip Bhattacharya, South Asia Research 26 (2006): 103.
36 Ibid., 104.
246 Sathaye
legend of Medb, the principal figure of the Irish epic Táin Bó Cúailnge.37 On the
most basic level, Dumézil observed that the name of the two princesses derive
from the same Indo-European root, *medhuā-, meaning “intoxicating” and
related to Sanskrit madhu (“honey, sweet”) and English “mead.”38 Second, he
noted an obvious parallel between the two storylines: both queens marry four
different kings in the course of their career. Finally, Dumézil explored deeper
structural affinities between the two stories, arguing both Medb and Mādhavī
to be feminine personifications of an Indo-European conceptualization of
royal power.39 That is, both stories operate on “an Indo-European theory of the
nature, the chances, the risks of royalty and, above all, the qualities it requires.”40
These qualities are specifically embodied first by the princesses’ husbands and
then her sons, who represent a tripartite division of social functions.41 Most
important for Dumézil’s analysis is the congruence between the personae
of their fathers—Eochaid Feidlech and Yayāti. Noting that both kings clash
with their sons but have obedient, self-sacrificing daughters, Dumézil main-
tained that their saga illustrates “a vast theory of kingship”42 that was active
in archaic Indo-European thought. Within this mythological scheme, Yayāti/
Eochaid embodies an originary, universal kingship (pIE *rēg-). His grandsons
further divide this royal power into the three functions that order traditional
society, while the daughter, Mādhavī/Medb, serves as a matrilocal pathway for
its distribution.
Building off of Dumézil’s insights, a number of Indologists have explored
how the Mahābhārata “indigenized” the Indo-European tradition.43 J.A.B. van
Buitenen, for example, argues that this older “raw material” has been repack-
aged to form a “sequel” to the Mahābhārata’s other legends of Yayāti.44 As
37 Georges Dumézil, The Destiny of a King, trans. Alf Hiltebeitel (Chicago: University of Chi-
cago Press, 1973).
38 Ibid., 81–4.
39 Ibid., 83.
40 Ibid., 94.
41 Ibid., 28–29. On Dumézil’s trifunctional theory of Indo-European social thought, see
Nicholas J. Allen, “Hinduism, Structuralism and Dumézil,” in Miscellanea Indo-Europea,
ed. Edgar C. Polomé, Journal of Indo-European Studies Monographs 33 (Washington, DC:
Institute for the Study of Man, 1999), 241–60; and Emily B. Lyle, “Dumézil’s Three Func-
tions and Indo-European Cosmic Structure,” History of Religions 22 (1982): 25–44.
42 Dumézil, The Destiny of a King, 103.
43 Alf Hiltebeitel, “Comparing Indo-European ‘Epics’,” History of Religions 15 (1975): 94.
44 J.A.B. van Buitenen, “Introduction [to Book 5. The Book of the Effort],” in The Mahābhā
rata: 4. The Book of Virāṭa; 5. The Book of the Effort (Chicago: Chicago University Press,
1978).
Pride and Prostitution 247
45 Alf Hiltebeitel, “Dumézil and Indian Studies.” Journal of Asian Studies 34 (1974): 137; see
also Hiltebeitel, “Comparing Indo-European ‘Epics’,” 94.
46 Alf Hiltebeitel, The Ritual of Battle: Krishna in the Mahābhārata (Albany: State University
of New York Press, 1990 [1976]), 223.
47 Dumézil, The Destiny of a King, 107.
48 The common early Indological theory that Brahmans had Bowdlerized or otherwise tam-
pered with the Mahābhārata (e.g., Adolf Holtzmann Jr., Zur Geschichte und Kritik des
Mahābhārata [Kiel: C.F. Haessler, 1892]; Frederick Eden Pargiter, Ancient Indian Historical
Tradition [London: Oxford University Press, 1922]) remains problematic, at best—see
Vishwa Adluri and Joydeep Bagchee, The Nay Science: A History of German Indology (New
York: Oxford University Press, 2014) for an important critique of the German nationalist
discourses embedded within the idea of a Brahmanical textual takeover.
49 Dumézil, The Destiny of a King, 98.
50 Ibid., 78.
51 Ibid., 79.
52 Ibid., 79.
248 Sathaye
53 Ibid., 106–7.
54 Simon Brodbeck, The Mahābhārata Patriline: Gender, Culture, and the Royal Hereditary
(Aldershot: Ashgate Publishing, 2009).
Pride and Prostitution 249
and what Mary Brockington deems “backwards composition.”55 That is, the
first plot incident contours the interpretation of the second, and the second
contours the third. Thus, before audiences get to the morally shocking treat-
ment of Mādhavī, they are already made to think of it as abnormal, if not
abhorrent, by the events of the Śāṇḍilī subplot.
Before taking him to meet King Yayāti and his daughter Mādhavī, Garuḍa,
who has come to help his friend Gālava, first carries the young Brahman to a
mountain named “Ṛṣabha” where they are shown hospitality by a female
ascetic named Śāṇḍilī.56 After a nice meal, the pair fall into a kind of “food
coma” (annamohitau) (Mahābhārata 5.111.3d), but Garuḍa soon awakens to dis-
cover that his wings have been clipped, so that “the bird resembled a lump of
flesh with a head and feet.”57 Gālava asks the bird, “What inauspicious, morally
despicable thing have you been thinking of?”58 Garuḍa confesses that he had
contemplated taking the woman around to visit Prajāpati, Mahādeva, and
Viṣṇu (5.111.8d–9ab). He begs for mercy, explaining that he had done it only
“with good intentions” (priyakāmyayā) (5.111.10b). After admonishing the eagle
for insulting her, Śāṇḍilī forgives him and restores his original form, on the
condition that he “is never to disrespect women, even if they be worthy of such
abuse.”59
Garuḍa makes the promise, and the pair move on to see Yayāti—but the
moral landscape has already been given shape: before the epic audiences may
hear about how Mādhavī is subordinated, they are shown a strong, indepen-
dent female who resists male dominance. Her subject-position, we are told, is
based upon a perfected spiritual state (siddhi) that she has acquired through
55 James M. Hegarty, “Encompassing the Sacrifice: On the Narrative Construction of the Sig-
nificant Past in the Sanskrit Mahābhārata,” Acta Orientalia Vilnensia 7 (2006): 108; Mary
Brockington, “The Art of Backwards Composition: Some Narrative Techniques in Vālmīki’s
Rāmāyaṇa,” in Composing a Tradition: Concepts, Techniques and Relationships, Proceed-
ings of the First Dubrovnik International Conference on the Sanskrit Epics and Purāṇas,
August 1997, ed. Mary Brockington and Peter Schreiner (Zagreb: Croatian Academy of Arts
and Sciences, 1999), 99–110.
56 The location of this Ṛṣabha is uncertain. In the Rāmāyaṇa (4.39.39), Mount Ṛṣabha is
placed in the Himālaya near Kailāsa. In the Bhāgavata Purāṇa (10.79.15), Ṛṣabha is in the
Tamil country, while the sixteenth-century Caitanyacaritāmṛta (Madhyalīlā 9.166) locates
it specifically in the Palani Hills north of Madurai. The Mahābhārata here describes
Mount Ṛṣabha as located on the “chest of the ocean” (sāgarorasi) (5.110.22b), also sugges-
tive of a southern location.
57 māṃsapiṇḍopamo 'bhūt sa mukhapādānvitaḥ khagaḥ | (Mahābhārata 5.111.5ab)
58 kiṃ nu te manasā dhyātam aśubhaṃ dharmadūṣaṇam | (Mahābhārata 5.111.7ab)
59 ninditāsmi tvayā vatsa na ca nindāṃ kṣamāmy aham |
na ca te garhaṇīyāpi garhitavyāḥ striyaḥ kva cit || (Mahābhārata 5.111.16cd)
250 Sathaye
60 This good behavior involves an orientation towards both dharma and artha, a matter to
which we will return. The epic states a proverb here: “A person attains dharma through
good behavior, and through good behavior one also gets money; a person achieves pros-
perity through good behavior, and good behavior nullifies bad omens” (ācārāl labhate
dharmam ācārāl labhate dhanam | ācārāc chriyam āpnoti ācāro hanty alakṣaṇam ||;
Mahābhārata 5.111.15).
61 tāṃ dṛṣṭvā cārusarvāṅgīṃ tāpasīṃ brahmacāriṇīm |
grahītuṃ hi manaś cakre rūpāt sākṣād iva śriyam || ) (Mahābhārata 5*453 after 5.111.4b).
This verse is a marginal insertion found in one Devanāgarī manuscript, labeled “K1” by the
Mahābhārata’s critical editors, that is dated 1566 CE and allied with the Kashmiri Śāradā
manuscripts.
Pride and Prostitution 251
Given Garuḍa’s tarnished moral record, why then does Gālava so readily
trust him as a friend? The question never comes up because the first subplot of
the story has already provided an answer: Gālava is desperate, thinking himself
to be at the end of his rope. The story begins with Gālava attending to his
teacher Viśvāmitra, who is tested by the god Dharma as he attempts to gain
Brahmanhood. Dharma takes on the disguise of his archrival Vasiṣṭha and
comes begging for alms. Viśvāmitra produces a pot containing homemade,
steaming-hot rice porridge, but Dharma tells the sage that he has already eaten,
asks him to wait a while, and leaves. Holding the pot on his head, Viśvāmitra
patiently remains motionless for one hundred years—attended upon by the
dutiful Gālava. When Lord Dharma returns disguised as Vasiṣṭha, he is
impressed with Viśvāmitra’s forbearance and dedication, and declares him to
have become a Brahman.
Viśvāmitra then dismisses his loyal student, but Gālava insists on giving a
gurudakṣiṇā, a graduation fee, to his teacher. Viśvāmitra repeatedly declines
the offer, but Gālava refuses to take “No” for an answer. Losing his patience,
Viśvāmitra demands an impossible gift of eight hundred śyāmakarṇa horses—
pure white stallions each with one black ear (Mahābhārata 5.104.26). On
hearing this request Gālava fall into intense self-pity, for he has no money, no
friends, and no hope to keep his promise. Contemplating suicide, he makes a
final appeal to Viṣṇu, and it is precisely at that moment that Garuḍa comes to
his rescue.
It is important to consider what Viśvāmitra might represent in the Gālava
carita. As a king who turns himself into a Brahman, Viśvāmitra is a robust
counter-normative figure in Hindu mythology, a man who achieves the impos-
sible due to extraordinary hubris and personal will.62 He is a terrifying Oedipal
father figure in both epics,63 but the Mahābhārata especially exoticizes Viśvā
mitra as the Brahman “Other,” a persona to be feared and respected for his
violent power and extraordinary accomplishments, but never to be emulated
by “normal” Brahmans.64 By invoking Viśvāmitra, I suggest that the composers
precondition what their audiences are to think of Gālava—one expects that he
62 Adheesh Sathaye, Crossing the Lines of Caste: Viśvāmitra and the Construction of Brahmin
Power in Hindu Mythology (New York: Oxford University Press, 2015).
63 Robert P. Goldman, “Fathers, Sons and Gurus: Oedipal Conflict in the Sanskrit Epics,” Jour-
nal of Indian Philosophy 6 (1978): 325–92 and Robert P. Goldman, “Matricide, Renuncia-
tion, and Compensation in the Legends of Two Warrior-Heroes of the Sanskrit Epics,”
Indologica Taurinensia 10 (1982): 117–31.
64 Adheesh Sathaye, “The Other Kind of Brahman: Rāma Jāmadagnya and the Psychosocial
Construction of Brahman Power in the Mahābhārata,” in Epic and Argument in Sanskrit
Literary History: Essays in Honor of Robert P. Goldman, ed. Sheldon Pollock (New Delhi:
Manohar, 2010), 185–207.
252 Sathaye
will operate at the margins of social propriety, following in the footsteps of one
of the most notorious examples of such behavior. The demonstration of
Viśvāmitra’s extraordinary perseverance in achieving Brahman status further
preconditions the audience’s reception of his pupil’s stubbornness: Gālava is,
in a sense, learning to be like his teacher in challenging social laws that good
people are discouraged from challenging.
Viśvāmitra’s Brahmanhood, one might say, also takes on a “metamythical”
capacity in the Gālavacarita, gluing an older Indo-European narrative relic into
a more contemporary mythological tradition.65 Apart from his famous change
of caste, epic audiences would certainly have known of Viśvāmitra’s being one
the seven preeminent Vedic seers (saptarṣi), his prominence in the genealogies
of kings and sages, and his involvement in Hariścandra’s aborted human sacri-
fice of Śunaḥśepa.66 At this point in the epic’s narration, audiences would
already have heard of Viśvāmitra’s rivalry with Vasiṣṭha (Mahābhārata 1.165-7),
his irascibility, and his potential for unleashing violent curses (1.65-66). They
would have also known at least one version of the story of his miscegenated
birth and its connection to the śyāmakarṇa horses (115).67 Those with more
advanced Vedic training would have known Viśvāmitra had a son named
65 On Viśvāmitra legends as “metamyths” see Robert P. Goldman, Gods, Priests, and Warriors:
The Bhṛgus of the Mahābhārata (New York: Columbia University Press, 1977), 7-8, 104, 111.
66 On Viśvāmitra’s place within the Seven Seers, see John E. Mitchiner, Traditions of The
Seven Ṛṣis (Delhi: Motilal Banarsidass, 1982); for Viśvāmitra’s genealogical lines, see
V.G. Rahurkar, The Seers of the Ṛgveda (Poona: University of Poona, 1964) and Umesh
Chandra Sharma, The Viśvāmitras and the Vasiṣṭhas: An Exhaustive Historical Study, Vedic
and Post-Vedic (Aligarh: Viveka Publications, 1975); on Śunaḥśepa, see H.L. Hariyappa,
Ṛgvedic Legends through the Ages (Poona: Deccan College Postgraduate Research Insti-
tute, 1953), Hermann Lommel, “Die Śunaḥśepa-Legende,” Zeitschrift der Deutschen Mor-
genländischen Gesellschaft 114 (1964): 122–61, David Gordon White, “Śunahśepa Unbound,”
Revue de l’histoire des religions 203 (1986): 227–62.
67 These horses, as is revealed at the end of the Gālavacarita but already discussed in the
Āraṇyakaparvan (Mahābhārata 3.115.15-18), miraculously emerged from the Ganges at a
site called the “Horse Ford” (aśvatīrtha) due to a boon given to Ṛcīka, a Bhārgava Brahman
who had married Viśvāmitra’s sister Satyavatī. Her father Gādhi had demanded as a bride
price from Ṛcīka a thousand śyāmakarṇa horses, which did not exist prior to their mirac-
ulous emergence from the river (115.12). Afterwards, these divine horses were distributed
to various kings, but, we learn, four hundred of them drowned in the Vitastā river. Since
only six hundred now remained, it turns out that Gālava had been given an impossible
task. On the epic and purāṇic Satyavatī legend, see Robert P. Goldman “Akṛtavraṇa vs.
Śrīkṛṣṇa as Narrators of the Legend of Bhārgava Rāma: A Propos Some Observations of Dr.
V.S. Sukthankar,” Annals of the Bhandarkar Oriental Research Institute 53 (1970): 161–73;
Sathaye, “How to Become a Brahman” and Sathaye, “The Other Kind of Brahman.”
Pride and Prostitution 253
68 Aṣṭaka is named as the seer of Ṛgveda 10.104, while the Brāhmaṇas state that he was the
son of Viśvāmitra (Aitareya Brāhmaṇa 7.17; Śāṅkhāyana Śrautasūtra 15.26) and installed
into kingship (rājyam) by his father (Jaiminīya Brāhmaṇa § 145). For an exhaustive analy-
sis of Aṣṭaka in Vedic literature, see Umesh Chandra Sharma, “Aṣṭaka Viśvāmitra: A Study,”
Journal of the Oriental Institute of Baroda 23 (1975): 169–74.
69 Harivaṃśa 9.97-100; Brahmāṇḍa Purāṇa 63.86-89; Vāyu Purāṇa 88.88-90; Śiva Umāsaṃhitā
37.56-59. The Mahābhārata knows of Triśaṅku’s ascension to heaven while being cursed
as an outcaste (1.65.34), but not the events of the Satyavrata legend. F.E. Pargiter posits
that the legend dates from the sixth or seventh century bce (1913: 902), while other schol-
ars more realistically place the purāṇa-pañcalakṣaṇa corpus, in which the legend is
found, in the fifth or sixth centuries CE (Hans T. Bakker, “Early Mythology Relating to
Vārāṇasī,” in Banāras (Vārāṇasī): Cosmic Order, Sacred City, Hindu Traditions (Festschrift to
Prof. R.L. Singh), ed. Rana P.B. Singh, Cultural Traditions of India 5 [Varanasi: Tara Book
Agency, 1993], 21).
70 Danielle Feller, The Sanskrit Epics: Representation of Vedic Myths (New Delhi: Motilal
Banarsidass, 2004), 15–17.
254 Sathaye
upon this curious narrative relic: one of religious significance and the other of
early Indian political geography.
The names of nearly all of the characters in this story—Brahman and
Kṣatriya—are found in the Vedic corpus (Mādhavī does not appear in Vedic
literature). The Brahmans Viśvāmitra, Aṣṭaka, and Gālava, as we have seen,
were well-known Vedic seers. As for the Kṣatriyas, Yayāti is mentioned twice in
the Ṛgveda Saṃhitā (1.31.17, 10.63.1), while Mādhavī’s second husband Divodāsa
appears throughout Vedic literature (e.g., ṚV 6.16, 6.61, 9.61) and was “one of the
leading princes of the early Vedic Age.”71 Pratardana is found in the Yajurveda
(Kāṭhaka Saṃhitā 21.10; Kauṣītaki Brāhmaṇa 26.2.5) and is said to be Divodāsa’s
son (Kauśītaki Upaniṣad 3.1). Mādhavī’s third husband, Uśīnara, appears in the
Aitareya Brāhmaṇa (8.14), while Śibi, son of Uśīnara, is mentioned in the
Baudhāyana Śrautasūtra (21.18). The only kings not having explicit Vedic ante-
cedents are Mādhavī’s first husband Haryaśva, king of Ayodhyā, and his son
Vasumanas. However, even these two appear in an important passage within
one Vedic ancillary text in which nearly all of these kings are unified.
The Ṛgvedasarvānukramaṇi of Kātyāyana (c. second century, bce) attri-
butes the seership of a single, three-line Ṛgveda verse (ṚV 10.179) jointly to Śibi
Auśīnara (“the son of Uśīnara”), Pratardana Kāśīrāja (“the king of Kāśī”), and
Vasumanas Rauhidaśva (“the son of Rohidaśva”).72 This correspondence is no
coincidence, and the epic composers may have been grafting the Mādhavī story
onto the older Vedic attribution, perhaps as a way of filling out the background
details. The concluding scene of the story, in which the three half-brothers
(together with Aṣṭaka) conduct a sacrifice for their grandfather, thus offers
an epic narrativization of the original Vedic mantra in question: “Your friends
with their treasures are sitting around you waiting like tribal leaders for their
wandering king.”73 This is precisely what takes place at the story’s end as Yayāti
71 Arthur Anthony Macdonell and Arthur Berriedale Keith, Vedic Index of Names and Sub-
jects, vol. 1 (Delhi: Motilal Banarsidass, 1995 [1912]), 363.
72 The situation is further clarified by the commentary of Ṣaḍguruśiṣya (late twelfth century
CE) who explains that “each one saw each verse in sequence” (ekaikasyā ṛcaḥ krameṇa
draṣṭāraḥ | ; Vedārthadīpikā 10.179). “Rohidaśva” here must be a variant for Haryaśva, as
both have the same meaning of “possessing bay horses.” Also, while Macdonell and Keith
state that Pratardana “is not in Vedic literature a king of Kāśī,” he clearly has become one
by Kātyāyana’s time. Arthur Anthony Macdonell and Arthur Berriedale Keith, Vedic Index
of Names and Subjects, vol. 2 (Delhi: Motilal Banarsidass, 1995 [1912]), 29.
73 pári tvāsate nidhíbhiḥ sákhāyaḥ kulapā́ ná vrājápatiṃ cárantam | (ṚV 10.179.2); Ralph
Griffith translates this line as: “Friends with their stores are sitting round thee waiting like
lords of clans for the tribe’s wandering chieftain.” Ralph T.H. Griffith, The Hymns of the
Rigveda (New Delhi: Motilal Banarsidass, 1973 [1889]).
Pride and Prostitution 255
tumbles from heaven, his merit depleted due to an excess of pride, along the
smoke trail generated by his grandsons’ sacrificial rite. The Gālavacarita thus
absorbs two generations of otherwise obscure Vedic kings into the dynastic
lines of a better-known progenitor: Yayāti.
These rulers of Vedic pedigree are all also tied further to specific northern
Indian cities, a geocultural mapping that constitutes a second type of “lighting”
cast upon the Gālavacarita. Divodāsa and his son Pratardana are placed at Kāśī
(modern-day Vārāṇasī), Haryaśva and Vasumanas are at Ayodhyā, while
Uśīnara and Śibi are in the city of Bhojapura. And while Aṣṭaka’s capital is
unstated, the epics consistently locate Viśvāmitra as a ruler of Kānyakubja, or
modern-day Kanauj.
What are we to make of these geographical references? Vedic, Buddhist,
and Jain sources make it clear that Kāśī was one of the sixteen Great States
(mahājanapadas) that constituted the political landscape of North India in the
sixth century bce and that it was “at one time, one of the most powerful states
of north India.”74 Vārāṇasī was Kāśī’s capital but was perhaps more important
as a commercial center, owing to its strategic location at the confluence of the
Gaṅgā, Varaṇā, and Asī rivers.75 And while nearby Sarnath was a major cen-
ter of Buddhist activity, Vārāṇasī appears not to have grown in significance for
Hindu practice until the end of the third century CE when it began a several-
centuries-long transformation into the premier site for Śaiva worship.76 Hans
Bakker links this to the emergence of four cycles of Śaiva myths, in which
Divodāsa plays a key role in the founding of the city, its abandonment due to
a curse, and the arrival of Śiva to build his home there.77 Divodāsa, in other
words, was central to the cultural imagination of Vārāṇasī as a site of political
and religious power.
Ayodhyā was located in the janapada of Kosala, which, like Kāśī, was a well-
known North Indian polity in the sixth century bce. But, like Vārāṇasī, the
74 Upinder Singh, A History of Ancient and Early Medieval India: From the Stone Age to the
12th Century (New Delhi, Upper Saddle River, NJ: Pearson Education, 2008), 262; cf. Par-
manand Gupta, Geography from Ancient Indian Coins and Seals (New Delhi: Concept Pub-
lishing Company, 1989), 118–9.
75 Dilip Kumar Chakrabarti, The Oxford Companion to Indian Archaeology: The Archaeologi-
cal Foundations of Ancient India, Stone Age to ad 13th Century (New Delhi: Oxford Univer-
sity Press, 2006), 331.
76 Hans T. Bakker, “Construction and Reconstruction of Sacred Space in Vārāṇasī,” Numen 43
(1996): 33.
77 Bakker, “Early Mythology Relating to Vārāṇasī,” 23–28. Bakker notes that Divodāsa’s
founding of the city is already mentioned in the Mahābhārata (13.31), as is the connection
of Vārāṇasī to Śaiva worship (82.69, 13.26.14).
256 Sathaye
town of Ayodhyā itself perhaps was still a place on the rise. Then called “Sāketa,”
it was a center of trade, while Śrāvastī to the north was Kosala’s capital.78 The
city is a place of importance in early Buddhist and Jain literature and both
epics do connect it to Rāma’s Ayodhyā.79 Still, it seems not to have risen as a
center of power until Kumāragupta made it the Guptas’ imperial seat in the
late fifth century.80
Unlike Divodāsa and Kāśi, the connections of Haryaśva and Vasumanas to
Ayodhyā remain obscure. Clearer, however, are Ayodhyā’s commonalities with
Vārāṇasī—both were important commercial centers at North Indian cross-
roads before the Mahābhārata’s composition and both were important sites of
Buddhist and Jain activity. While the two classical janapadas in which they
were located—the fierce rivals Kosala and Kāśī—were annexed by the
Mauryans, Ayodhyā and Vārāṇasī rose to lasting cultural and political promi-
nence in the centuries after the empire’s disintegration.
The home of Mādhavī’s third son, Śibi Auśīnara, is somewhat more difficult
to determine, and several possibilities exist. The most likely candidate is the
Panjab, as both Pāṇinī (Aṣṭādhyāyī 2.4.20) and Patañjalī (Mahābhāṣya 9.2.118)
place the Uśīnaras or Śibis in the Beas river valley, while the Macedonians
under Alexander encountered a tribe called the Sibae at the confluence of the
Jhelum (Gr. Hydaspes) and Chenab (Gr. Akesines) rivers.81 But they don’t
appear to have stayed there exclusively: coins of the Śibis dated from c. 150 bce
have been found near Chittor in Rajasthan, and so Parmanand Gupta argues
that the Śibis migrated from the Panjab into Rajasthan shortly before this
time.82 Other scholars have placed the Śibis in the upper Gangetic valley near
Dehradun or Haridwar.83 Because of the city’s name, Feller suggests that Śibi
78 Ayodhyā’s commercial prominence was likely due to its being at a crossroads—with high-
ways going east to Vārāṇasī, south to Pratiṣṭhāna, and west towards Taxila (Hans T. Bakker,
Ayodhyā, Groningen Oriental Studies, vol. 1 [Groningen: Egbert Forsten, 1986], 13).
79 Alf Hiltebeitel, “Empire, Invasion, and India's National Epics,” International Journal of
Hindu Studies 2 (1998): 402.
80 Bakker, Ayodhyā, 30.
81 Gupta, Geography from Ancient Indian Coins and Seals, 42; Singh, A History of Ancient and
Early Medieval India, 274. The Jātakas also place King Śibi in the Shorkot region of the
Panjab (Debarchana Sarkar, Geography of Ancient India in Buddhist Literature, [Kolkata:
Sanskrit Pustak Bhandar, 2003], 213–14).
82 Gupta, Geography from Ancient Indian Coins and Seals, 43. Sarkar (Geography of Ancient
India in Buddhist Literature, 213) provides corroborative evidence of this migration from
Buddhist literature.
83 Gupta, Geography from Ancient Indian Coins and Seals, 44; Bhagwan Singh Suryavan-
shi, Geography of the Mahabharata (New Delhi: Ramanand Vidya Bhawan, 1986), 162;
Pride and Prostitution 257
Auśīnara belonged to the Bhoja tribe, which the Mahābhārata locates in the
Vindhyas.84 On the other hand, the exclusively southern location of the Bhojas
is not so clear, since the Sabhā Parvan, in describing the fanning out of the vari-
ous branches of the Lunar Dynasty, speaks of a vast dispersal of the descendants
of Yayāti and the Bhojas in all the cardinal directions (Mahābhārata 2.13.6).85
This is perhaps a mythic echo of the historical migrations of the Śibis to
Rajasthan, the upper Ganges, Sindh, and Vidarbha. Another possibility arises,
I suggest, if we take Bhojapura to be a variant of “Bhojanagara” or “Bhoganagara,”
a town that Buddhist sources locate on the road between Vaiśālī and
Kapilavastu, and which scholars have identified with the village of Badaraon in
Bihar.86
The exact location of Śibi’s Bhojapura must ultimately remain a mystery.
However, the identity of the king is not at all mysterious—this is clearly the
same Śibi of Buddhist fame, who in Jātaka literature selflessly cuts off pieces of
his flesh to ransom a dove from a predatory hawk, and pulls out his own eyes to
give to a blind man. The fourth-century Chinese traveller Faxian (as well as
Xuanzang two hundred years later) speaks of two stupas in northwest India
where Śibi’s acts of self-sacrifice were commemorated, and the Buddhist nar-
rative tradition features a number of stories of Śibi’s generosity and virtue as a
bodhisattva.87 As Reiko Ohnuma points out, Śibi exemplifies in these texts the
supremacy of the universal Buddhist dharma over what is termed “nīti-
dharma,” a context-based ethics of conduct.88 In the Mahābhārata on the other
hand, Śibi represents the “moral essence of ksatriya dharma.”89 In Śibi, then,
Vijayendra Kumar Mathur, Aitihasik Sthanavali (New Delhi: Ministry of Education, Stand-
ing Committee on Scientific and Technical Terminologies, 1969), 102.
84 Feller, “The Strange Story of Princess Mādhavī,” 13; Sarkar Geography of Ancient India in
Buddhist Literature, 235. There is also Buddhist textual evidence as well as the mention of
a “Bhojakaṭa” in the Barhut inscription that would place the Bhojas in Vidarbha (Sarkar,
Geography of Ancient India in Buddhist Literature, 235).
85 Suryavanshi, following K.D. Bajpai (The Geographical Encyclopaedia of Ancient and
Medieval India, Based on Vedic, Puranic, Tantric, Jain, Buddhistic Literature and Historical
Records [Varanasi: Indic Academy, 1967], 69), claims Mathurā to be the capital of the Bho-
jas (Suryavanshi, Geography of the Mahabharata, 23), but also corroborates Sarkar’s sug-
gestion by noting that a “Bhojakaṭanagara” has been mentioned as the capital of Vidarbha
in the Mahābhārata (ibid., 166).
86 Sarkar Geography of Ancient India in Buddhist Literature, 235.
87 Reiko Ohnuma, Head, Eyes, Flesh, And Blood: Giving Away the Body in Indian Buddhist Lit-
erature (New York: Columbia University Press, 2007), 2–3, 275–76.
88 Ibid., 114.
89 Arti Dhand, “The Politics and the Dharma of Conversion: Reflections from the Ma-
habharata,” Journal of Hindu-Christian Studies 15 (2002): 9, col. 2. The only complete
258 Sathaye
telling of the Śibi story in the Mahābhārata has been relegated to an appendix of the
Critical Edition (Mahābhārata 3, App. 1, no. 21). For a comparative analysis of Śibi in the
Mahābhārata and Buddhist texts, see Edith Parlier, “La légende du roi des Śibi: Du sacri-
fice brahmanique au don du corps bouddhique,” Bulletin d’É tudes Indiennes 9 (1991): 133–
60.
90 Rama Shankar Tripathi, History of Kanauj to the Moslem Conquest, 2nd ed. (Delhi: Motilal
Banarsidass, 1959), 15; Pargiter, Ancient Indian Historical Tradition, 267–70.
91 On Patañjali, see Tripathi, History of Kanauj to the Moslem Conquest, 16; for numismatic
evidence, see Gupta, Geography from Ancient Indian Coins and Seals, 100.
92 On the Mahāvastu, see Sarkar, Geography of Ancient India in Buddhist Literature, 247; for
Faxian’s reports on Kanauj, see Tripathi, History of Kanauj to the Moslem Conquest, 18, and
Baij Nath Puri, Cities of Ancient India (Meerut: Meenakshi Prakashan, 1966), 27.
93 George Erdosy, “The Origin of Cities in the Ganges Valley,” Journal of the Economic and
Social History of the Orient 28 (1985): 84.
Pride and Prostitution 259
Vedic pedigree. Just as a single verse links them together in Vedic ancillary lit-
erature, their genealogical lines are here tied together by a single mother,
Mādhavī, who serves in a matrilocal capacity for one of the most famous Lunar
Dynasty kings. That is to say that behind the geocultural mappings there is a
genealogical power-play—one that may be better appreciated by looking more
closely at the capital of Yayāti, the fifth and most important king in our story.
Yayāti’s prominence in the Mahābhārata is especially due to pivotal role
he plays in the genealogy of the Lunar Dynasty.94 The site of the king’s capi-
tal, however, is generally not named and does not really form an integral
part of his career. In the Gālavacarita, Yayāti is called “the Lord of Vatsa and
Kāśī” (vatsakāśīrāja), and his capital is placed at Pratiṣṭhāna, a city that the
epic locates near the confluence of the Gaṅgā and Yamunā, near modern-day
Allahabad (Mahābhārata 5.118.1; 3.83.72). Scholars have identified it with the
town of Jhusi and as part of a hierarchical network of urban settlements in
the area headed by Kauśāmbi.95 I find it tempting to conjecture that the name
of Yayāti’s capital might also covertly invoke another Pratiṣṭhāna, located fur-
ther to the south in Vidarbha on the banks of the Godāvarī river, which came
to be the capital of the Sātavāhana rulers in the post-Mauryan period. Like
the other cities in the Gālavacarita, this Pratiṣṭhāna, identified with the mod-
ern Maharashtrian town of Paithan, was a noted commercial center in the
mid-first millennium bce, and Buddhist texts identify it as the capital of the
janapada of Aśmaka (“Assaka”).96 Like Vārāṇasī, Ayodhyā, and Kanauj, this
Pratiṣṭhāna rose in prestige in the centuries after the fall of the Mauryas. The
Sātavāhanas appear to have been singularly responsible for its fame—for at
this site they fostered both material uniformity and interregional communica-
tion, leading to the flourishing of industry, agriculture, and classical religion
across the Deccan.97
In a Nashik cave inscription dated c. 141 bce, the Sātavāhana ruler Pulumāyin
Vasiṣṭhīputra compares himself to a list of epic heroes that includes Yayāti,
94 Brodbeck, The Mahābhārata Patriline, 103–117; see also Michel Defourny, Le Mythe de
Yayāti dans la Littérature Épique et Purānique, Bibliothèque de la Faculté de Philosophie
et Lettres de l’Université de Liège, 221 (Paris: Les Belles Lettres, 1978).
95 Singh, A History of Ancient and Early Medieval India, 285.
96 Sarkar, Geography of Ancient India in Buddhist Literature, 260; Puri, Cities of Ancient India,
61; Alexander Cunningham, Ancient Geography of India, edited and with introduction and
notes by Surendranath Majumdar Sastri (Calcutta: Chuckervertty, Chatterjee & Co, 1924),
746.
97 Aloka Parasher-Sen, “Urban Centres—Deccan,” in Archaeology of Early Historic South
Asia, ed. Gautam Sengupta and Sharmi Chakraborty (New Delhi: Pragati Publications,
2008), 328.
260 Sathaye
Viṣṇu, the Lord of all three worlds—that Kṛṣṇa, who is the highest path
for mortal creatures. He pervades all the other gods and demons, and
98 EI 8.61. See Defourny, Le Mythe de Yayāti dans la Littérature Épique et Purānique, 11.The
connection with the upstart Sātavāhanas might perhaps explain why Yayāti informs
Garuḍa, “I am not as rich as you used to know me, my friend, for I have lost my wealth”
(yathā jānāsi māṃ purā | na tathā vittavān asmi kṣīṇaṃ vittaṃ hi me sakhe ||; Mahābhā
rata 5.113.6bcd).
99 Along these lines, we should note that Yayāti appears to have been a favorite name among
the Lunar Dynasty kings of early medieval Orissa (see Defourny, Le Mythe de Yayāti dans
la Littérature Épique et Purānique, 11).
Pride and Prostitution 261
It is then that Garuḍa, with Viṣṇu’s permission, rushes to take Gālava on his
adventure.
Viṣṇu appears no further in the Gālavacarita, but he does play a larger
role within the two legends that precede it. The Mātali story reinforces
Viṣṇu’s supremacy over his servant Garuḍa as well as Indra, his older Vedic
brother, and thereby shows the folly of challenging Viṣṇu’s power. The
Dambhodbhava legend presents an even more lucid vision of Viṣṇu’s divin-
ity. The Brahman sage Rāma Jāmadagnya—better known in popular Hinduism
as “Paraśurāma”—tells Duryodhana the “Subnarrative of Dambhodbhava”
(dambhodbhavopākhyāna) in order to discourage the prince from going to war
with the Pāṇḍavas.101 Believing himself invincible and “intoxicated with great
pride” (darpeṇa mahatā mattaḥ; Mahābhārata 5.94.8c), this universal king
(sārvabhauma) boastfully asks a group of Brahmans if anyone is more powerful
than him. They tell him that there is in fact a pair of sages, Nara and Nārāyaṇa,
to whom he could never be equal. Dambhodbhava brashly gathers his armies,
marches to their hermitage and attacks the two mystical beings. To repel his
onslaught, Nara grabs a fistful of reeds and launches them at Dambhodbhava’s
forces. These magically pierce the soldiers and the king is soundly defeated.
Dambhodbhava falls at Nara’s feet seeking refuge, and Nara blesses him in
terms quite similar to what Śāṇḍilī and Viṣṇu use to rebuke Garuḍa: “Be good
to brahmans, keep a righteous nature, and don’t ever act like this again!”102
Rāma Jāmadagnya then explains that Nara is now Arjuna, while the even more
powerful Nārāyaṇa has become Kṛṣna. This is why Duryodhana should ensure
his own welfare (svārtham; Mahābhārata 5.94.45d) and make peace with the
Pāṇḍavas.
The “lighting” of classical Vaiṣṇava theology, which starkly illuminates the
Dambhodbhava legend, grows progressively more diffuse in the subsequent
stories. Viṣṇu intervenes only at the very end of the Mātali legend, while,
The voice of the epic’s composers is most clearly heard through the third fea-
ture of our museological analogy—the dialogues that frame each epic
subnarrative and serve as explanatory “text panels” for the story-exhibit.
Indeed, if we want to know why Nārada tells the Gālavacarita, we ought to
consider his own explanation in its opening and closing dialogues. Like the
information that curators place next to artifacts in a museum, these dialogical
frames were clearly designed to regulate the audience experience of the narra-
tive relic to which they are juxtaposed.
Nārada begins his narration by cautioning Duryodhana against behaving
like Gālava: “You shouldn’t be obstinate, because obstinacy [nirbandha] is very
dangerous. And in this regard, there is an old chronicle illustrating how Gālava
found himself defeated through his obstinacy.”104 But this explanation, how-
ever nice it sounds, does not actually fit the story. For upon closer scrutiny, we
can see that while Gālava’s troubles are no doubt initiated by his stubbornness
to give his guru a dakṣiṇā, it is not really his main problem. Not only does it
pale in comparison to Viśvāmitra’s extraordinary perseverance in becoming a
Brahman, obstinacy turns out not to lead Gālava to ruin but to success. Doesn’t
Gālava actually satisfy his teacher’s demands? Don’t his efforts lead to the suc-
cessful continuation of four separate royal lines and doesn’t it help save Yayāti
when he falls from heaven in disgrace? If everything eventually works out for
the best, what then is the point of criticizing Gālava’s obstinacy? Why then
shouldn’t Duryodhana behave like him?
As if anticipating this very criticism, Nārada reworks his message in his clos-
ing remarks. After describing how Yayāti had fallen from heaven, only to be
restored by the merits of his grandsons, his daughter, and Gālava, the sage
explains:
103 See James W. Laine, Visions of God: Narratives of Theophany in the Mahābhārata (Vienna:
Institut für Indologie der Universität Wien, 1989).
104 na kartavyaś ca nirbandho nirbandho hi sudāruṇaḥ ||
atrāpy udāharantīmam itihāsaṃ purātanam |
yathā nirbandhataḥ prāpto gālavena parājayaḥ || (Mahābhārata 5.104.6cd–7)
Pride and Prostitution 263
It was with pride [abhimāna] that Yayāti had been corrupted, long ago,
and from extreme obstinacy [nirbandha] that Gālava had been as well,
Lord. You should listen to friends who have good intentions, and those
who wish you well. And do not be obstinate, because obstinacy only
brings ruin. And so, Duryodhana, you yourself should let go of your pride
and anger. Make peace with the Pāṇḍavas, and forget about war, mighty
King.105
Nārada’s takeaway message synthesizes the key issues of the story’s three plot
incidents: Gālava’s stubborn nature (in the Viśvāmitra incident), the problem-
atic friendship of Garuḍa (in the Śāṇḍilī episode), and Yayāti’s prideful fall
from grace (in the Mādhavī subplot). Obstinacy may have been one reason for
Gālava’s “corruption” (doṣa), but it has here been subordinated to pride—
abhimāna. In doing so, this “text panel” refocuses critical attention onto the
pressing anxieties of kings and princes in the post-Mauryan political milieu:
the perils of pride, disrespect, and not listening to friends—or more precisely,
listening to the wrong kind of friends, who lack moral compass, and who, like
Garuḍa, may not always show respect to those who deserve it, be they high or
low. It was a point already made in the Dambhodbhava legend when Nara had
urged the repentant king to be “righteous” (dharmātman):
Don’t ever get puffed up with pride [darpa] and persecute others, King,
whether they are important or inferior—that is the best plan for you. Be
wise, lose your greed, be selfless but self-aware. Be forgiving, forbearing,
gentle, and comforting, and take care of your people, King. Take our
blessings and go—but don’t ever behave this way again.106
108 Compare this with the Śānti, where these are fifteen and twelve, respectively, or the
Anuśāsana, where dharmārtha is used in the sense of “for the sake of dharma” a total of
thirteen times.
109 Van Buitenen makes a similar suggestion in noting that the Udyoga reflects upon “the
knowledge of the futility of this war, and an endeavor—it is there now in prospect; later
when the war is over, in retrospect—to deal with it as a moral lesson. “Van Buitenen,
Introduction [to Book 5. The Book of the Effort],” in The Mahābhārata: 4. The Book of
Virāṭa; 5. The Book of the Effort (Chicago: Chicago University Press, 1978), 181.
Pride and Prostitution 265
explains, “while that of the wicked appears as the opposite.”110 He urges the
prince to eschew immorality and failure (adharma and anartha) by following
a certain set of moral principles that he then proceeds to outline. It is a passage
that breaks out of the epic frame and appears to speak directly to the epic’s
realworld audiences.
Kṛṣṇa’s monologue argues for a hierarchy within the three aims of life
(puruṣārthas). “The undertakings of the enlightened,” he explains, “are filled
with the three aims of life. But when it is impossible to have all three, then
men (should) adhere to dharma and artha.”111 Not only is dharma the supreme
life-aim (Mahābhārata 5.122.33), the other two should be pursued only in sub
ordination to it:
A man who, carried away by his senses, abandons dharma due to his
desires and begins to strive for kāma and artha through illicit means—he
is ruined. One should still strive for kāma and artha, but one should first
pursue dharma—for artha or kāma never take precedence over dharma.
They say that dharma itself is the means for getting the three aims of life,
King, and anyone who pursues it therefore grows stronger, like a fire in
dry forest.112
This is to say that all aspirations towards pleasure or power (kāma or artha)
must be tempered by moral responsibilities. As we might glean from the
Arthaśāstra and Buddhist sources, politics in Mauryan and pre-Mauryan North
India was driven by a mandate to achieve and preserve power by any means
necessary.113 While the edicts of Aśoka do indicate a public policy based on
universalized Buddhist ethics, there is little indication that Aśoka’s strict regu-
lations on interpersonal relations had sustained effect on inter-state politics
after the disintegration of the Mauryan imperial formation. We may therefore
historicize the Mahābhārata’s Brahmanical discourse on dharmārtha as speak-
ing to the subsequent period of social and political flux in North India.
References
Primary Sources
Aitareya Brāhmaṇa. Edited by R. Anantakrishna Shastri, P.K. Narayana Pillai, and
S.K. Pillai. Trivandrum: University of Travancore, 1942–55.
Baudhāyanaśrautasūtra. Edited by C.G. Kashikar. New Delhi: Indira Gandhi National
Centre for the Arts, 2003.
Bhāgavata Purāṇa (with commentary of Gaṅgasahāya). Delhi: Nag Publishers, 1987.
(Original edition, Bombay: Ksemaraja Srikrsnadasa, 1928-32).
Brahmāṇḍa Purāṇa. Edited by K.V. Sarma. Kṛṣṇadāsa Sanskrit Series; 41. Varanasi:
Kṛṣṇadāsa Academy, 2000.
Caitanyacaritāmṛta of Kṛṣṇadāsa Kabirāja. Edited by Radhikanath Gosvami and
Nityasvarupa Brahmacari. Calcutta: Sriguru Library, 1995.
Harivaṁśa. Critical edition. Edited by Parashuram Lakshman Vaidya. Pune: Bhandarkar
Oriental Research Institute, 1969–71.
Jaiminīya Brāhmaṇa (in Auswahl). Edited by Wilhelm Caland. Amsterdam: Johannes
Moller, 1919.
Kāṭhaka Saṃhitā of the Yajurveda. Edited by Leopold von Schroeder. Leipzig:
F.A. Brockhaus, 1900.
Kauśītakī Brāhmaṇa. Edited by E.R. Sreekrishna Sarma. Verzeichniss der Orientalischen
Handschriften in Deutschland, Supp. 9, 1. Wiesbaden: Franz Steiner Verlag.
Kauśītakī Upaniṣad. Edited by Patrick Olivelle. South Asia Research. New York: Oxford
University Press, 1998.
Liṅga Purāṇa. Bombay, Sri Venkatesvara Steam Press. 1963 [Original edition, 1906].
The Mahābhārata for the First Time Critically Edited. Edited by Vishnu S. Sukthankar,
Shripad K. Belvalkar, Parashuram L. Vaidya, et al. 19 vols. Poona: Bhandarkar Oriental
Research Institute, 1933–66.
The Mahābhārata, vol. 3. 4. The Book of Virāṭa; 5. The Book of the Effort. Translated by
J.A.B. van Buitenen. Chicago: University of Chicago Press, 1978.
Rāmāyaṇa of Vālmīki. Critically edited by J.M. Mehta, et al. Baroda: Oriental Institute,
1960-75.
Ṛgveda Samhitā (with the commentary of Sāyaṇācārya). Pune: Vaidik Samshodhan
Mandal, 1933.
Ṛgvedasarvānukramaṇi of Kātyāyana (with the Vedārthadīpikā commentary of Ṣaḍgu
ruśiṣya). Edited by Arthur Anthony Macdonell. Anecdota Oxoniensa, Aryan Series
1, part 4. Oxford: Clarendon Press, 1886.
Śāṅkhāyanaśrautasūtra. Edited by Alfred Hillebrandt. Calcutta: Baptist Mission Press,
1888.
Śiva Purāṇa. Delhi: Nag Publishers, 1996. (Original edition, Bombay: Srivenkatesvara
Steam Press, 1906).
Vāyu Purāṇa. Pune: Ānandāśrama Press, 1983. Ānandāśrama Sanskrit Series; 49.
Pride and Prostitution 269
Viṣṇu Purāṇa (with the commentaries of Periyālvār and Śrīdharasvāmin). Delhi: Nag
Publishers, 1985. (Original edition, Bombay: Sri Venkateshvara Steam Press, 1910).
Secondary Sources
Adluri, Vishwa. “Frame Narratives and Forked Beginnings: Or, How to Read the
Ādiparvan.” Journal of Vaishnava Studies 19, no. 2 (Spring 2011): 143–210.
Adluri, Vishwa, and Joydeep Bagchee. The Nay Science: A History of German Indology.
New York: Oxford University Press, 2014.
Allen, Nicholas J. “Hinduism, Structuralism and Dumézil.” In Miscellanea Indo-Europea,
edited by Edgar C. Polomé, 241–60. Journal of Indo-European Studies Monographs 33.
Washington, D.C.: Institute for the Study of Man, 1999.
Bajpai, Krishna Datta, ed. The Geographical Encyclopaedia of Ancient and Medieval India,
Based on Vedic, Puranic, Tantric, Jain, Buddhistic Literature and Historical Records.
Varanasi: Indic Academy, 1967.
Bakker, Hans T. Ayodhyā. Groningen Oriental Studies, vol. 1. Groningen: Egbert Forsten,
1986.
———. “Construction and Reconstruction of Sacred Space in Vārāṇasī.” Numen 43
(1996): 32–55.
———. “Early Mythology Relating to Vārāṇasī.” In Banāras (Vārāṇasī): Cosmic Order,
Sacred City, Hindu Traditions (Festschrift to Prof. R.L. Singh), edited by Rana P.B. Singh,
21–28. Cultural Traditions of India 5. Varanasi: Tara Book Agency, 1993.
Bhattacharji, Sukumari. “Economic Rights of Ancient Indian Women.” Economic and
Political Weekly 26 (1991): 507–12.
———. “Prostitution in Ancient India.” Social Scientist 15 (1987): 32–61.
Bhattacharya, Pradip. “Desire under the Kalpataru.” Journal of South Asian Literature 28
(1993): 315–35.
———. “Five Holy Virgins, Five Sacred Myths: A Quest for Meaning.” Manushi 141 (July
2004): 4–12.
———. “Living by Their Own Norms: Unique Powers of the Panchkanyas,” Manushi 145
(February 2005): 30–37.
———. “Of Kunti and Satyawati: Sexually Assertive Women of the Mahabharata.”
Manushi 142 (August 2004): 21–25.
———. “‘One-in-Herself’: Why Kunti Remains a Kanya.” Manushi 143 (October 2004):
25–33.
———. “Pañcakanyā: Women of Substance.” Journal of South Asian Literature 35 (2000):
13–56.
———. “Pancha Kanya, A Quest in Search of Meaning—Part I.” Journal of Human Values
12 (2006): 1–29.
———. “Pancha Kanya, A Quest in Search of Meaning—Part II.” Journal of Human
Values 12 (2006): 107–36.
270 Sathaye
———. “The Riddle of the ‘Pancha Kanyā’ (Five Maidens).” South Asia: Journal of South
Asian Studies 32 (2009): 3–45.
———. “She Who Must Be Obeyed: Draupadi, the Ill-Fated One.” Manushi 144 (December
2004): 19–30.
———. “Women in Ancient India.” Economic and Political Weekly 26, no. 24 (1991): 1426.
Blackburn, Stuart. “The Brahmin and the Mongoose: The Narrative Context of a Well-
Travelled Tale.” Bulletin of the School of Oriental and African Studies 59 (1996):
494–507.
Bowles, Adam. Dharma, Disorder and the Political in Ancient India: the Āpaddharmaparvan
of the Mahābhārata. Leiden: E.J. Brill, 2007.
Brockington, Mary. “The Art of Backwards Composition: Some Narrative Techniques in
Vālmīki’s Rāmāyaṇa.” In Composing a Tradition: Concepts, Techniques and Relation
ships, Proceedings of the First Dubrovnik International Conference on the Sanskrit Epics
and Purāṇas, August 1997, edited by Mary Brockington and Peter Schreiner, 99–110.
Zagreb: Croatian Academy of Arts and Sciences, 1999.
Brodbeck, Simon. The Mahābhārata Patriline: Gender, Culture, and the Royal Hereditary.
Aldershot: Ashgate Publishing, 2009.
———. Review of Pancha-Kanya: The Five Virgins of Indian Epics. A Quest in Search of
Meaning, by Pradip Bhattacharya. South Asia Research 26 (2006): 101–5.
Chakrabarti, Dilip Kumar. The Oxford Companion to Indian Archaeology: The Archaeolog
ical Foundations of Ancient India, Stone Age to ad 13th Century. New Delhi: Oxford
University Press, 2006.
Cunningham, Alexander. Ancient Geography of India, edited and with introduction and
notes by Surendranath Majumdar Sastri. Calcutta: Chuckervertty, Chatterjee & Co.,
1924.
Dasgupta, Madhusraba. “Usable Women: The Tales of Ambā and Mādhavī.” In Faces of
the Feminine in Ancient, Medieval, and Modern India, edited by Mandakranta Bose,
48–55. New York: Oxford University Press, 2000.
Defourny, Michel. Le Mythe de Yayāti dans la Littérature Épique et Purānique. Bibliothèque
de la Faculté de Philosophie et Lettres de l’Université de Liège, 221. Paris: Les Belles
Lettres, 1978.
Dhand, Arti. “The Politics and the Dharma of Conversion: Reflections from the
Mahabharata.” Journal of Hindu-Christian Studies 15 (2002): 6–12.
Doniger, Wendy. “Fluid and Fixed Texts in India.” In Boundaries of the Text: Epic
Performances in South and Southeast Asia, edited by Joyce Burkhalter Flueckiger and
Laurie J. Sears, 31–42. Michigan Papers on South and Southeast Asia 35. Ann Arbor,
MI: Center for South and Southeast Asian Studies, University of Michigan, 1991.
Edgerton, Franklin. “Introduction.” In The Sabhāparvan, being the Second Book of the
Mahābhārata, The Great Epic of India, critically edited by Franklin Edgerton, ix–l.
Poona: Bhandarkar Oriental Research Institute, 1944.
Pride and Prostitution 271
Erdosy, George. “The Origin of Cities in the Ganges Valley.” Journal of the Economic and
Social History of the Orient 28 (1985): 81–109.
Feller Danielle. The Sanskrit Epics: Representation of Vedic Myths. New Delhi: Motilal
Banarsidass, 2004.
———. “The Strange Story of Princess Mādhavī.” Paper presented at the Fifth Dubrovnik
International Conference on the Sanskrit Epics and Puranas, August 11–16, 2008.
Fitzgerald, James L. “The Many Voices of the Mahābhārata.” Review of Rethinking the
Mahābhārata: A Reader’s Guide to the Education of the Dharma King, by Alf Hiltebeitel,
Journal of the American Oriental Society 123, no. 4 (2003): 803–18.
———. “The Great Epic of India as Religious Rhetoric: A Fresh Look at the ‘Mahābhā
rata.’” Journal of the American Academy of Religion 51 (1983): 611–30.
Goldman, Robert P. “Akṛtavraṇa vs. Śrīkṛṣṇa as Narrators of the Legend of Bhārgava
Rāma: A Propos Some Observations of Dr. V.S. Sukthankar.” Annals of the Bhandarkar
Oriental Research Institute 53 (1970): 161–73.
———. “Fathers, Sons and Gurus: Oedipal Conflict in the Sanskrit Epics.” Journal of
Indian Philosophy 6 (1978): 325–92.
———. “Gods in Hiding: The Mahābhārata’s Virāṭa Parvan and the Divinity of the Indian
Epic Hero.” In Modern Evaluation of the Mahābhārata (R.K. Sharma Felicitation
Volume), edited by S.P. Narang, 73–100. Delhi: Nag Publishers, 1995.
———. Gods, Priests, and Warriors: The Bhṛgus of the Mahābhārata. New York: Columbia
University Press, 1977.
———. “Matricide, Renunciation, and Compensation in the Legends of Two Warrior-
Heroes of the Sanskrit Epics.” Indologica Taurinensia 10 (1982): 117–31.
Griffith, Ralph T.H. The Hymns of the Rigveda. New Delhi: Motilal Banarsidass, 1973
[1889].
Gupta, Parmanand. Geography from Ancient Indian Coins and Seals. New Delhi: Concept
Publishing Company, 1989.
Hardy, Friedhelm. “Information and Transformation: Two Faces of the Purāṇas.” In
Purāṇa Perennis: Reciprocity and Transformation in Hindu and Jaina Texts, edited by
Wendy Doniger, 159–81. Albany: State University of New York Press, 1993.
Hariyappa, H.L. Ṛgvedic Legends through the Ages. Poona: Deccan College Postgraduate
Research Institute, 1953.
Hegarty, James M. “Encompassing the Sacrifice: On the Narrative Construction of the
Significant Past in the Sanskrit Mahābhārata.” Acta Orientalia Vilnensia 7 (2006):
77–118.
———. Religion, Narrative, and Public Imagination in South Asia: Past and Place in the
Sanskrit Mahabharata. London: Routledge, 2011.
Hiltebeitel, Alf. “Comparing Indo-European ‘Epics.’” History of Religions 15 (1975):
90–100.
———. “Dumézil and Indian Studies.” Journal of Asian Studies 34 (1974): 129–37.
272 Sathaye
———. “Empire, Invasion, and India’s National Epics.” International Journal of Hindu
Studies 2 (1998): 387–421.
———. “More Rethinking the Mahābhārata: Toward a Politics of Bhakti.” Indo-Iranian
Journal 47 (2004): 203–27.
———. “Not Without Subtales: Telling Laws and Truths in the Sanskrit Epics.” Journal
of Indian Philosophy 33, no. 4 (2005): 455–511.
———. Rethinking the Mahābhārata: A Reader’s Guide to the Education of the Dharma
King. Chicago: University of Chicago Press, 2001.
———. The Ritual of Battle: Krishna in the Mahābhārata. Albany: State University of
New York Press, [1976] 1990.
Laine, James W. Visions of God: Narratives of Theophany in the Mahābhārata. Vienna:
Institut für Indologie der Universität Wien, 1989.
Lath, Mukund. “The Concept of Ānṛśaṃsya in the Mahābhārata.” In The Mahābhārata
Revisited, edited by R.N. Dandekar, 113–19. New Delhi: Sahitya Akademi, 1990.
Lommel, Hermann. “Die Śunaḥśepa-Legende.” Zeitschrift der Deutschen Morgenlän
dischen Gesellschaft 114 (1964): 122–61.
Lyle, Emily B. “Dumézil’s Three Functions and Indo-European Cosmic Structure.” History
of Religions 22 (1982): 25–44.
Macdonell, Arthur Anthony and Arthur Berriedale Keith. Vedic Index of Names and
Subjects, 2 vols. Delhi: Motilal Banarsidass Publishers, 1912.
Mathur, Vijayendra Kumar. Aitihasik Sthanavali. New Delhi: Ministry of Education,
Standing Committee on Scientific and Technical Terminologies, 1969.
Mehta, Mahesh. “The Problem of the Double Introduction to the Mahābhārata.” Journal
of the American Oriental Society 93 (1973): 547–50.
Minkowski, Christopher Z. “The Interrupted Sacrifice and the Sanskrit Epics.” Journal
of Indian Philosophy 29 (2001): 169–86.
———. “Janamejaya’s Sattra and Ritual Structure.” Journal of the American Oriental
Society 109, no. 3 (1989): 401–20.
———. “Snakes, Sattras, and the Mahābhārata.” In Essays on the Mahābhārata, edited
by Arvind Sharma, 384–400. Leiden: E.J. Brill, 1991.
Mitchiner, John E. Traditions of the Seven Ṛṣis. Delhi: Motilal Banarsidass Publishers,
1982.
O’Flaherty, Wendy Doniger. Women, Androgynes, and other Mythical Beasts. Chicago:
University of Chicago Press, 1980.
Ohnuma, Reiko. Head, Eyes, Flesh, And Blood: Giving Away the Body in Indian Buddhist
Literature. New York: Columbia University Press, 1980.
Pargiter, Frederick Eden. “Viśvāmitra and Vasiṣṭha.” Journal of the Royal Asiatic Society
of Great Britain and Ireland 45 (1913): 885–904.
———. “Viśvāmitra, Vasiṣṭha, Hariścandra, and Śunaḥśepa.” Journal of the Royal Asiatic
Society of Great Britain and Ireland 49 (1917): 37–67.
Pride and Prostitution 273
———. Ancient Indian Historical Tradition. London: Oxford University Press, 1922.
Parasher-Sen, Aloka. “Urban Centres—Deccan.” In Archaeology of Early Historic South
Asia, edited by Gautam Sengupta and Sharmi Chakraborty, 313–35. New Delhi: Pragati
Publications, 2008.
Parlier, Edith. “La légende du roi des Śibi: Du sacrifice brahmanique au don du corps
bouddhique.” Bulletin d’É tudes Indiennes 9 (1991): 133–60.
Puri, Baij Nath. Cities of Ancient India. Meerut: Meenakshi Prakashan, 1966.
Raheja, Gloria Goodwin, and Ann Grodzins Gold. Listen to the Heron’s Words: Reimagining
Gender and Kinship in North India. Berkeley: University of California Press, 1994.
Rahurkar, V.G. The Seers of the Ṛgveda. Poona: University of Poona, 1964.
Ramanujan, A.K. “Two Realms of Kannada Folklore.” In Another Harmony: New Essays
on the Folklore of India, edited by Stuart H. Blackburn and A.K. Ramanujan, 41–75.
Berkeley, CA: University of California Press, 1986.
Ramaswamy, Vijaya. “Gender and Transcendence in Early India.” In A Social History of
Early India, edited by Brajadulal D. Chattopadhyaya. Vol. 2, part 5 of A History of
Science, Philosophy, and Culture in Indian Civilization, 233–47. New Delhi: Center for
Studies in Civilizations, 2009.
Sarkar, Debarchana. Geography of Ancient India in Buddhist Literature. Kolkata: Sanskrit
Pustak Bhandar, 2003.
Sathaye, Adheesh. “How to Become a Brahman: The Construction of Varṇa as Social
Place in the Mahābhārata’s Legends of Viśvāmitra.” Acta Orientalia Vilnensia 8 (2007):
41–67.
———. Crossing the Lines of Caste: Viśvāmitra and the Construction of Brahmin Power
in Hindu Mythology. New York: Oxford University Press, 2015.
———. “The Other Kind of Brahman: Rāma Jāmadagnya and the Psychosocial
Construction of Brahman Power in the Mahābhārata.” In Epic and Argument in
Sanskrit Literary History: Essays in Honor of Robert P. Goldman, edited by Sheldon
Pollock, 185–207. New Delhi: Manohar, 2010.
———. “Was the Mahābhārata an Encyclopedia? Orality, Narrative Structure, and the
Legends of Viśvāmitra in the Sanskrit Epic.” Paper presented at the 217th Meeting of
the American Oriental Society. Chicago, IL. March 2008.
Sharma, Umesh Chandra. The Viśvāmitras and the Vasiṣṭhas: An Exhaustive Historical
Study, Vedic and Post-Vedic. Aligarh: Viveka Publications, 1975.
———. “Aṣṭaka Viśvāmitra: A Study.” Journal of the Oriental Institute of Baroda 23 (1975):
169–74.
Singh, Upinder. A History of Ancient and Early Medieval India: From the Stone Age to the
12th Century. New Delhi, Upper Saddle River, NJ: Pearson Education, 2008.
Suryavanshi, Bhagwan Singh. Geography of the Mahabharata. New Delhi: Ramanand
Vidya Bhawan, 1986.
274 Sathaye
Chapter 9
Introduction
The Ambā-Upākhyāna, which occurs in the fifth major book of the epic, the
Udyogaparvan,2 throws some light on the intellectual poverty of upholding the
process of “interpolation” over and above compositional genius. Although
superficially this subtale appears to be a circumscribed insertion, its semantic
roots, in fact, hold together the overall architecture of the epic. Standing in
opposition to the patriarch Bhīṣma pitāmaha, Ambā is ultimately instrumen-
tal in bringing him down, and with this, decisively resolving the fratricidal
conflict that climaxes in war.
Ambā is mentioned briefly in the Ādiparvan in connection with her abduc-
tion by Bhīṣma along with her two sisters Ambikā and Ambālikā. Ambā reveals
to Bhīṣma that she loves king Śālva and Bhīṣma permits her to go. Surprisingly,
Ambā is not referred to again until the epic’s fifth book when Bhīṣma narrates
how, thirsting for revenge,3 she obtained a boon from Śiva that she shall be
reborn a man and kill him. Scholars have hitherto considered Ambā’s story an
etiological narrative (explaining how the invincible warrior could have been
1 This chapter draws on two papers on the Ambā-Upākhyāna, presented on different occasions:
Vishwa Adluri, “Inscribing the Ineffable: The Ambā Narrative in the Mahābhārata,” paper
presented at The Third Australasian Sanskrit Conference, Sydney, Australia, July 22, 2012 and
Vishwa Adluri, “Bringing it All to a Close: The Ambā Upākhyāna of the Mahābhārata,” paper
presented at The Forty-First Annual Conference on South Asia, Madison, WI, October 13, 2012.
Wherever references are found to those articles, they should be understood as referring to this
chapter, as most of their materials have been absorbed into it.
2 Mahābhārata 5.170–193, preceded by a brief reference at 5.169.16–21.
3 We learn in the fifth book that Ambā, after returning to Śālva, was rejected by him as she had
been another man’s. Ambā therefore regards Bhīṣma as the cause of her ruin.
defeated4), but there is more to this story than meets the eye. Ambā, whose
name means “mother,”5 specifically evokes the Goddess, while the story of her
transgendering recalls not only Śiva’s own gender ambiguity, but also the
Puruṣa-Prakṛti dyad. Her evolution shows a marked affinity to Umā: she evolves
from a bridesmaid to a stigmatized outsider, transforming herself through
tapas and becoming first a half-woman (paralleling ardhanārīśvara) and ulti-
mately and explicitly the androgyne Śikhaṇḍin.6 Her full manifestation
unleashes the power of Śiva, who, going ahead of Arjuna, destroys the Kurus
(yas tu te so ’grato yāti yuddhe saṁpraty upasthite | taṁ viddhi rudraṁ
kaunteya devadevaṁ kapardinam ||; Mahābhārata 12.330.69), and by present-
ing weapons to both Arjuna (paśupata; 3.41.13) and to Aśvatthāman (khaḍgam
uttamam; 10.7.64), brings the yuga to a close. “By protecting the Pāñcālas,
I have honored him [Viṣṇu],” says Śiva to Aśvatthāman in the Sauptikaparvan
(kṛtas tasyaiṣa saṁmānaḥ pāñcālān rakṣatā mayā; 10.7.63), highlighting both
their complementarity at the level of continuity and Śiva’s ultimate role in
closing out the cosmic and textual cycle.
In this chapter, I show how the Ambā narrative—by evoking a constellation
of themes central to the theology of the Goddess such as the divine androgyne,
the Puruṣa-Prakṛti dyad, and natality and mortality—is a vital element of the
epic’s philosophical architecture. The theology of the Mahābhārata, it thus
seems, is less concerned with sectarian glorifications and more concerned
with the textual project of a very specific kind of ontology: one that takes both
the cycles of Becoming and the unity of Being seriously. This ontological aspect
has mainly been identified with the male deity, either Kṛṣṇa Vāsudeva, or Śiva,
4 For a discussion of some of these theories (and some wider issues pertaining to Mahābhārata
scholarship) see Vishwa Adluri and Joydeep Bagchee, The Nay Science: A History of German
Indology (New York: Oxford University Press, 2014), 284–88 and see especially 288, n. 584.
5 Apte, s. v. “ambā.”
6 When Ambā embarks upon severe austerities to propitiate Śiva, the goddess Gaṅgā appears
and attempts to dissuade her. But when she refuses to be swayed, the goddess curses her to
become a crooked river (nadī bhaviṣyasi śubhe kuṭilā; Mahābhārata 5.187.34) and the maiden
indeed becomes a river with half her body, while with the other half remaining a woman (cf.
Mahābhārata 5.187.40: sā kanyā tapasā tena bhāgārdhena vyajāyata | nadī ca rājan vatseṣu
kanyā caivābhavat tadā ||). But the word for “half a river” is ardhanadī, which recalls ardhanāri,
half-woman, and thus Śiva as the divine androgyne Ardhanārīśvara. Further, ultimately it is
the full power of the Rudra-Goddess pair that is required to bring down Bhīṣma: Arjuna, as
both Scheuer and Hiltebeitel have shown, is a Rudra figure. Like Ambā, he too features a
complex gendering, as Biardeau and Hiltebeitel have noted. Finally, it is not only Arjuna as
Rudra/Nara behind Śikhaṇḍin, but also Śiva as Kapardin ahead of the pair, who is required to
bring down the patriarch. The entire episode is thus couched in terms of pralaya motifs.
The Divine Androgyne 277
or Brahmā. 7 But the Goddess plays a significant role in this philosophical liter-
ary edifice as well, whether as Draupadī,8 or, as I argue here, as Ambā.
In her book Splitting the Difference,9 Wendy Doniger provides an analysis of the
story of Ambā/Śikhaṇḍin, focusing on the theme of gender enantiomor-
phism. Ambā, a princess of Kāśi rejected by the Kuru patriarch Bhīṣma, is
forced to change gender (and be reborn as Śikhaṇḍin) before she can exact
vengeance on the man who humiliated her. Doniger explores a number of
other gendered reversals that mark her narrative:
[S]he is reborn as the child of Drupada, which makes her the sister of the
polyandrous Draupadi, whose hypersexuality stands in dramatic contrast
with the reborn Amba’s ambiguous sexuality. The liminal Shikhandin/
Shikhandini rejects her bride, who is humiliated as Amba has been, and
unsexes (and humiliates) a helpful goblin. His/her sexual ambivalence is
itself ambivalent, or at least doubled: s/he is a female first masquerading
as a male and then transformed into a male.10
7 See V.S. Sukthankar, On the Meaning of the Mahābhārata (Bombay: Asiatic Society, 1957),
focusing on Kṛṣṇa; Jacques Scheuer has written about the roles of Śiva and the Goddess in
the epic (see his Śiva dans le Mahābhārata [Paris: Presses Universitaires de France, 1982])
and Bruce M. Sullivan has done the same for Brahmā (Bruce M. Sullivan, Kṛṣṇa Dvaipāyana
Vyāsa and the Mahābhārata: a New Interpretation [Leiden: E.J. Brill, 1990] and “The Reli-
gious Authority of the Mahābhārata: Vyāsa and Brahmā in the Hindu Scriptural Tradi-
tion,” Journal of the American Academy of Religion 62, no. 2 [1994]: 377–401).
8 See Alf Hiltebeitel, “Śiva, the Goddess, and the Disguises of the Pāṇḍavas and Draupadī,”
History of Religions 20 (1980): 147–74.
9 Wendy Doniger, Splitting the Difference: Gender and Myth in Ancient Greece and India
(Chicago: University of Chicago Press, 1999).
10 Ibid., 285.
11 This seems to be refuted by the fact of other strong feminine roles such as Draupadī,
Kuntī, and Gāndhārī. But none of these women take part in the combat, which remains
essentially the preserve of male heroes.
278 Adluri
notions of divinity, especially the ambiguous power of the Goddess. Two points
from Doniger’s analysis are crucial to my argument. First, “Goddesses, always
dangerous, become even more dangerous when they become male.”12 Second,
that “Shikhandin does not seem to remember that s/he was Amba, even though
Shiva expressly promises her that she will remember … Shikhandin knows he
was Shikhandini, but apparently not that he was Amba.”13 These two points
show that Śiva’s promise does not contain a simple promise of memory of her
former identity as a woman, that is, ontic memory, but rather, ultimately of
memory of her true identity, that is, of her role in the Creation as the Goddess.
This is ontological memory in contrast to ontic memory: that she will “remem-
ber” to bring down the Brahmā figure Bhīṣma. From Ambā’s perspective, she
will remember that she is not a hapless princess but the divine androgyne;
from Śikhaṇḍin’s perspective, she will remember to become the divine andro-
gyne; from Bhīṣma’s perspective, he will remember that the play of Vasus is
finished when the avatāra Vāsudeva himself drives the double androgynes of
Ambā/Śikhaṇḍin and Arjuna/Bṛhannaḍā (with Kapardin in the front) on the
divine chariot14 that Śiva himself used in the destruction of Tripura.15 Raised
in Brahmaloka, educated by sage Paraśurāma and knowing the true identities
of Kṛṣṇa and Arjuna as Nara and Nārāyaṇa,16 Bhīṣma knows better than to take
on Śikhaṇḍin. The raṇa-yajña is now ready to consume not only the vaṁśas of
all characters, but also śṛṣṭi itself.17
12 Ibid., 286.
13 Ibid., 284.
14 On the origins of this chariot, see Hiltebeitel’s account in “The Two Kṛṣṇas on One Char-
iot: Upaniṣadic Imagery and Epic Mythology,” History of Religions 24, no. 1 (1984): 1–26.
15 See Madeleine Biardeau (Études de mythologie hindoue I. Cosmogonies purāṇiques [Paris:
École Française d’Extrême Orient, 1981] and Études de mythologie hindoue II. Bhakti et
avatāra [Paris: École Française d’Extrême Orient, 1994]) on the intersection of the myths
of pralaya and avatāra.
16 See Mahābhārata 5.48.1–28.
17 As I have argued elsewhere (see Vishwa Adluri, “Authenticity and the Problem of the
Beginning in the Mahābhārata” [PhD diss., Philipps-Universität Marburg, 2013]), the pre-
sentation of Becoming or bhavābhavau in the epic is articulated according to the four
paradigms of sacrifice, cosmology, genealogy, and war. Here the conflagration is complete
and the epic re-begins only with the inceptive gesture of Rudra turning to the Goddess in
the Umā-Maheśvara-Saṁvāda. This takes place in Book 13, after both the destructive and
the salvific paths are mapped out in Books 10–11 and Book 12.
The Divine Androgyne 279
In his article “Not Without Subtales: Telling Laws and Truths in the Sanskrit
Epics,”18 Hiltebeitel provides a list of sixty-seven upākhāyanas in the Mahābhā
rata.19 Of the narrators of these subtales, Bhīṣma narrates the most upākhyānas
(twenty-three)20 and Yudhiṣṭhira is the auditor of most of them (forty-nine).21
18 Alf Hiltebeitel, “Not Without Subtales: Telling Laws and Truths in the Sanskrit Epics,” Jour-
nal of Indian Philosophy 33, no. 4 (2005): 455–511.
19 Hiltebeitel arrives at this count as follows: “This number is reached by including all units
that are mentioned to be upākhyānas either in passing in the text, cited as upākhyānas in
the PS [Parvasaṁgraha], or called upākhyānas in the colophons and/or the running heads
for units in the Pune Critical Edition. In assessing instances mentioned only in the colo-
phons, I err toward generosity and count anything as an upākhyāna that appears to be
called such as the prominent title in either the Northern (N) or Southern (S) Recension.
In treating this number for special attention, it should thus be clear it is not a boundaried
group without overlap with other ‘ancillary story’ material (see Gombach, 2000). Rather, I
wish to take the 67 and the reverberations between them as a kind of sonar with which to
plumb the epic’s depths.” Ibid., 469–70.
20 Hiltebeitel writes: “As to upākhyāna narrators, Vaiśampāyana addresses ten to Janamejaya
(1–4, 11, 32, 39–40, 66–67); Bhīṣma narrates twenty-three: 21 (44–52, 54–65) to Yudhiṣṭhira
and the Pāṇḍavas and two (35–36) to Duryodhana; Lomaśa Ṛṣi tells eight of nine (from
numbers 14–22) to the Pāṇḍavas, Mārkaṇḍeya Ṛsi also tells them eight (23–28 and 30–31),
and Kṛṣṇa four (12, 41–42); Citraratha narrates three (6–8) to Arjuna and the Pāṇḍavas;
Śalya tells two: one (33) to Yudhiṣṭhira, the other (38) to Karṇa and Duryodhana; Vyāsa
tells one to Draupadī’s father Drupada (9) and another to the Pāṇḍavas (29); and six are
told by single-time speakers: Kuntī to Pāṇḍu (5, the only upākhyāna spoken by a woman),
Nārada to the Pāṇḍavas (10); Bṛhadaśva to the Pāṇḍavas (13); Akṛtavraṇa to the Pāṇḍavas
(15, interrupting Lomaśa’s skein); Rāma Jāmadagnya to the Kauravas (34); and Duryo
dhana to Karṇa and Śalya (47).” Ibid., 472.
21 Hiltebeitel, again, writes: “As to auditors, of the 56 that are addressed to main characters,
49 are told primarily to Yudhiṣṭhira, 48 of these to him and his Pāṇḍava brothers, and 44
of these also to their wife Draupadī (all of these told once the Pāṇḍavas and Draupadī are
in the forest). On the Kaurava side, three are addressed to Duryodhana and two to Karṇa.
Adding the 10 told to Janamejaya and one narrated to Pāṇḍu by Kuntī, one finds that 65 of
the 67 upākhyānas are addressed directly to members of the larger Kaurava household to
which all these listeners belong, and of which Yudhiṣṭhira is clearly the chief listener. Not
irrelevant to this pattern is the one in which King Drupada hears upākhyāna 9 as an expla-
nation of how his daughter can marry into that household. And likewise not irrelevant
would be the last upākhyāna in this tally, the anomalous number 53 known in S colo-
phons … as the Nārāyaṇīye Hayaśira-Upākhyāna. Here the primary narrator is Ugraśravas,
who answers a question by Śaunaka (speaking for the Naimiṣa Forest Ṛṣis) about the
Horse’s Head, a form of Viṣṇu, by quoting what Vyāsa told Janamejaya about that subject.”
Ibid., 472.
280 Adluri
In a way, one is tempted to see Bhīṣma and the upākhyānas as parallels on the
genealogical and textual levels of the text. This is especially so if we see what
functions these two serve in relation to the “main” Kuru narrative. Bhīṣma
stands in a complex relation to the Kuru family. He is both the wise patriarch
and cut away from the kingly business of succession and actual rule. For all his
insuperable military prowess and heroism, he remains primarily a resource of
wise narratives22 and, as Hiltebeitel has shown, puzzles about dharma.23 So
also the upākhyānas: they appear somewhat extraneous to the Kuru narrative,
which is mistakenly taken to be the “main” narrative, characterized by a
straightforward tale of heroes. The upākhyānas function as powers function in
mathematics: Ku. They so completely change the text that it would be hasty to
think they can be nothing more than random interpolations. The main narra-
tive depends on these texts to multiply its meanings. Hiltebeitel focuses mainly
on what these subtales do with questions of dharma24; I agree with him on this
point, but argue that the significance of the upākhyānas extends beyond even
dharma to the basic problem of the epic: the tension between pravṛtti and
nivṛtti and, ultimately, the relation of time to eternity.25
If we look at Bhīṣma’s role in the upākhyānas in the context of the signifi-
cant role he plays in the philosophical and pedagogic tracts of the Mahābhārata,
The Ambā-Upākhyāna occurs in the fifth major book of the epic, the Udyoga
parvan. It recounts the story of Ambā, the princess of Kāśi abducted by Bhīṣma
and rejected by Śālva. But already in the Ādiparvan (1.96), we hear of how
Bhīṣma, seeing his half-brother Vicitravīrya attain manhood, sets off for the
king of Kāśi’s daughters’ bridegroom choice. Bhīṣma abducts the princesses
Ambā, Ambikā, and Ambālikā,27 and returns to Hāstinapura. However, it
26 See Sullivan, Kṛṣṇa Dvaipāyana Vyāsa and the Mahābhārata and see also his “The Reli-
gious Authority of the Mahābhārata,” both cited in earlier in note 7.
27 The names of the three princesses links them with Rudra, addressed in Vedic literature as
“Tryambaka” (Vājasaneyi Saṁhitā 3.58, Śatapatha Brāhmaṇa 2.6.29, and a possible refer-
ence at Ṛgveda 7.59). Macdonell derives the epithet from “one who has three mothers”
(Ambikā = “mother”), which he takes as an allusion to the threefold division of the uni-
verse (GRV 1.555). He also notes that “Ambikā, a post-Vedic name of Śiva’s wife, is men-
tioned for the first time in VS. 3,5” (although it is used here not for his wife, but for his
sister), and that the more common names of Śiva’s wife Umā and Pārvati, “seem first to
282 Adluri
transpires that Ambā, the eldest, was promised to King Śālva and Bhīṣma
grants her permission to return to her lover.28 Ambikā and Ambālika are mar-
ried off to Vicitravīrya. This “prince of colorful virility,”29 however, dies before
producing any children. Anxious that her line should not die out, the dowager
queen Satyavatī urges Bhīṣma to beget offspring on the pair. When Bhīṣma, cit-
ing his vow of celibacy, refuses, she turns to another son, begotten before her
marriage to king Śaṁtanu with the sage Parāśara: the seer Vyāsa. With Bhīṣma
stepping aside for Satyavatī’s doomed sons and Satyavatī replacing Śaṁtanu’s
dead sons with her own, the true origins of the Kuru genealogy now lie outside
Kṣatriya paternity. Thus, supplemented by Gaṅgā and Draupadī, the Kṣatriya
core seems more infected with feminist inversions than Brahmanic interpola-
tions. On the one hand, Satyavatī displaces Gaṅgā as Śaṁtanu’s wife and the
legitimate heir Bhīṣma with her own sons. On the other, she also effectively
ends Śaṁtanu’s blood line: the line of the Kurus is now fully constituted by her
son Vyāsa and the princesses of Kāśi. The narrative of the descent of the
heaven-sent and heaven-linked Gaṅgā, with her engendering and liberating
the Vasus (see Mahābhārata 1.96), thus comes to an abrupt end, leaving behind
only Bhīṣma. The cosmological narrative that began in Brahmaloka with the
occur in the TA [Taittirīya Āraṇyaka] and the Keṇa Upaniṣad.” A.A. Macdonell, Vedic
Mythology (Delhi: Motilal Banarsidass, 2002 [Strassburg: Karl J. Trübner, 1897]), 74 (all ref-
erences are to 2002 edition).
28 The further story of Ambā, which is not recounted here, is as follows: in Book 5, Bhīṣma
tells Duryodhana that he will kill all of the warriors of the Pāṇḍavas, except Śikhaṇḍin,
whom he will not fight. Duryodhana asks him why he will not fight with Śikhaṇḍin and
Bhīṣma explains that he has sworn never to fight any woman or any man who was for-
merly a woman. Then he explains that in a former life Śikhaṇḍin was Ambā. Rejected by
her lover Śālva for having belonged to another man and thirsting for revenge, Ambā
sought refuge with sage Rāma Jāmādagnya. The sage battled Bhīṣma but was unable to
defeat him. Learning that no hero capable of defeating Bhīṣma existed, Ambā herself
undertook terrible austerities, ultimately winning Śiva’s grace. Śiva promised her that in
her next life she would be reborn a man and would defeat Bhīṣma. Ambā was reborn as
the girl-child Śikhaṇḍinī but true to Śiva’s word, she became a man after exchanging her
sex with a Yakṣa named Sthūṇākarṇa. For this reason, because of his vow that he will
“shoot no arrows at a woman, a former woman, one with the name of a woman, and an
apparent woman” (striyāṁ strīpūrvake cāpi strīnāmni strīsvarūpiṇi | na muñceyam ahaṁ
bāṇān iti kauravanandana ||; Mahābhārata 5.193.63), Bhīṣma says he will not kill Ambā/
Śikhaṇḍin when ambushed in battle.
29 See J.A.B. van Buitenen, “Introduction,” in The Mahābhārata: I. The Book of the Beginning
(Chicago: University of Chicago Press, 1971), xviii. Monier-Williams translates “of marvel-
lous heroism” which is linguistically possible, but false in context: the epic does not men-
tion any heroic deeds by Vicitravīrya in contrast to his brother Citrāngada. However, his
sexual exploits are mentioned (see Mahābhārata 1.96.53–56).
The Divine Androgyne 283
30 These parallels are spread throughout the narrative, not only at the level of the first
generation. Karṇa’s untimely birth recalls Vinatā’s breaking open of Aruṇi’s egg (Mahā
bhārata 1.14.15–20), whereas the birth of the five Pāṇḍavas recalls Garuḍa’s birth in great
splendor (1.20.4–15). The supernumerary Kauravas, further, recall the countless snakes, all
but a remnant of which perish in Janamejaya’s snake sacrifice. Finally, just as Garuḍa
becomes the vehicle of Nārāyaṇa (1.29.16), so also Kṛṣṇa Vāsudeva becomes the refuge of
the Pāṇḍavas and the charioteer of Arjuna’s vehicle.
31 Sacrifice is an organizing principle in the epic for understanding pravṛtti. It is co-original
with the creation of the universe (cf. Bhagavadgītā 3.10: sahayajñāḥ prajāḥ sṛṣṭvā purovāca
prajāpatiḥ | anena prasaviṣyadhvam eṣa vo ’stv iṣṭakāmadhuk ||). It is also co-constitutive
of reality (Bhagavadgītā 3.14: annād bhavanti bhūtāni parjanyād annasaṁbhavaḥ | yajñād
bhavati parjanyo yajñaḥ karmasamudbhavaḥ). Thus, to show that the two princesses are
embedded in a sacrificial ritual suffices to tie them to pravṛtti.
32 See Stephanie W. Jamison, Sacrificed Wife/Sacrificer’s Wife: Women, Ritual, and Hospitality
in Ancient India (New York: Oxford University Press, 1996), 243. This sacrificial meaning
does not oppose the theological meaning discussed earlier, because Jamison also notes
that the three vocatives “taken together” are “variants on affectionate words for mother”;
“together[,] they add up to the ‘three Ambikās’ of Rudra Tryambaka and its vṛddhi deriva-
tive, the Traiyambakahoma.” Ibid.
284 Adluri
horse.33 “The year-long vow [Vyāsa asks the two queens to obey] replicates the
Aśvamedha requirement that the queen remain abstinent during the year the
horse wanders”34; “the pair Mitra and Varuṇa [in whose likeness Vyāsa prom-
ises to beget sons on the queens] … has a Vedic ring. Vedavyāsa could be
alluding to ways that the Aśvamedha identifies the king with dharma, and also
that the Rājasūya invokes Mitra as ‘lord of truth’ and Varuṇa as ‘lord of dharma’
in announcing the newly consecrated Bharata king (MS 2.6.6; TS 1.8.10.1-2).”35
Hiltebeitel also notes that the epic refers (via Satyavatī) to Ambikā and
Ambālikā as “the two mahiṣīs” (mahiṣyau; Mahābhārata 1.97.9a). This incon-
gruous usage (the only time the dual is used in “either epic”) may be intended
to emphasize Ambā’s “unavailability as a Mother” (itself a “complement” to
“Bhīṣma’s [unavailability] as a Father”36), and to point out that, though unusual,
each of the two mahiṣīs now has an “equal chance to become the mother of the
one desired heir.”37
However, the greatest evidence for an epic “aśvamedha” comes from Vyāsa
himself, specifically from his parallels with the horse of the Vedic sacrifice: in
the Nārāyaṇīya, Vyāsa reveals to Janamejaya the existence of a form of Nārāyaṇa
called Harimedhas, who bears the head of a horse (hence, also “Hayaśiras”).
This form is identified in two ways with the Vedas: when at the dawn of the eon
the demons Madhu and Kaiṭabha steal the Vedas and conceal them in the
great ocean (i.e., the Milky Ocean), Nārāyaṇa assumes the horse-headed form,
“repository of the Vedas” (vedānām ālayaṁ; 12.335.44) to recover the Veda.
Seizing them from their resting place and returning them to Brahmā,
Nārāyaṇīya establishes the horse’s head in the northeast of the great ocean as
the “repository of the Vedas” (vedānām ālayas; 12.335.54) and returns to his
sleeping Aniruddha form. Now the Mahābhārata tells us that Nārāyaṇa is Vyāsa
and Vyāsa is Nārāyaṇa (kṛṣṇadvaipāyanaṁ vyāsaṁ viddhi nārāyaṇaṁ prabhum;
12.334.9), and Vyāsa himself reveals that he was originally born of Harimedhas
(the “sacrificial sap [medhas] of Hari”; cf. 12.337.54). Further, Vyāsa is referred
to as the “receptacle of the Veda” (kṛṣṇadvaipāyanaṁ vyāsamṛṣiṁ vedanidhiṁ;
App. 1, 32.14).38 As Hiltebeitel has noted, all this strengthens Vyāsa’s claim to
33 Alf Hiltebeitel, “Epic Aśvamedhas,” in Reading the Fifth Veda: Studies on the Mahābhārata,
Essays by Alf Hiltebeitel, vol. 1, ed. Vishwa Adluri and Joydeep Bagchee (Leiden: E.J. Brill,
2011), 259–78.
34 Ibid., 271.
35 Ibid., 271–72.
36 Ibid., 273.
37 Ibid., 274.
38 This line is moved in the Critical Edition text to an appendix by the editor S. K. Belvalkar,
following the reading of the minority Malayālam manuscripts which exclude it here.
The Divine Androgyne 285
39 Ibid., 272.
40 Ibid., 272 (emphasis in original).
41 See Uma Chakravarti, Of Meta-Narratives and ‘Master’ Paradigms: Sexuality and the Reifi-
cation of Women in Early India (New Delhi: Centre for Women’s Development Studies,
2009).
42 After she is rejected by king Śālva, Ambā makes her way to the hermitage of sage
Jāmadagnya. Rāma urges Bhīṣma to take the maiden back, since no man will want her
after she has been abducted. Bhīṣma refuses, citing kṣatradharma (Mahābhārata 5.178.11c),
and Jāmadagnya fights him. If Bhīṣma had followed his preceptor’s orders and obtained
offspring (albeit in violation of his promise to Satyavatī, though note that Satyavatī herself
is later willing to revoke this stipulation), his offspring would have been the legitimate
heir to the throne, avoiding the paternity and precessional conflicts that lead to the Kuru
war. The textual universe of the Mahābhārata thus in a sense hangs from the rejection of
Ambā’s desire for union: her expulsion creates the space within which the epic drama
unfolds and this space rapidly folds in on itself, once she returns in the fifth and sixth
books. Ambā is therefore the ultimate cipher for the laya motif of the epic.
43 Hiltebeitel, “Epic Aśvamedhas,” 272.
44 Note also that when Satyavatī, acting on Vyāsa’s advice, chooses exile to the forest to prac-
tice austerities, she places blame unambiguously on Ambikā’s side. “Ambikā, the mis-
guided policies of your son, so we hear,” she says, “will destroy the Bhāratas and their
followers and their grandsons” (Mahābhārata 1.119.9).
286 Adluri
of Śiva (as Ardhanārīśvara), this return initiates the laya cycle of the Mahā
bhārata.45 Two of the three mothers in the narrative initiate the pravṛtti cycle
that includes the Kuru narrative; the third, Ambā, by bringing down Bhīṣma,
terminates it.
Along with the Drupada-Droṇa rivalry, the opposition between Bhīṣma and
Ambā is one of the central conflicts in the Kuru narrative. The patriarch of the
Kuru dynasty and the most important warrior on the Kaurava side is a polyva-
lent figure, whose multiple identities refer to various levels of meaning of the
narrative. Here I will focus on two of his identities: as the grandfather, and as a
Vasu. Sullivan notes the significance of the term “pitāmaha” or “grandfather,”
an appellation given to two characters in the Mahābhārata: Vyāsa and Bhīṣma.46
Sullivan argues that “Vyāsa represents Grandfather Brahmā much more effec-
tively and often than does Bhīṣma.”47 Sullivan’s overall thesis is to show that
“The MBh, presenting itself as the ‘fifth Veda’ was created by pitāmaha Vyāsa as
the quintessence and completion of the four Vedas created by the pitāmaha
Brahmā. In many subtly suggestive expressions, the MBh reveals continuities
between Vyāsa and Brahmā.”48 While this thesis is essentially correct, I argue
that there are important ways in which Bhīṣma functions as Brahmā. Sullivan
himself notes that the “brahmin Vyāsa and the Kṣatriya Bhīṣma are comple-
mentary figures, who in certain respects are mirror images of each other. Both
remained unmarried and became the Bhārata family’s elders, vigilant guard-
ians of their family’s fortunes.”49
To throw light on Bhīṣma’s role as a patriarch, we need to look closely at the
qualifiers of the Kurukṣetra battle. The earthly battle is a foil for another battle:
the battle of gods and demons.50 Sullivan claims that,
45 On the Mahābhārata as a myth of laya, see Biardeau, Études de mythologie hindoue I and
II.
46 Bruce M. Sullivan, “The Epic’s Two Grandfathers, Bhīṣma and Vyāsa,” in Essays on the
Mahābhārata, ed. Arvind Sharma (Delhi: Motilal Banarsidass, 2007), 204–11.
47 Ibid., 207.
48 Ibid., 211.
49 Ibid., 205.
50 “The war between the Bhārata cousins is seen as another battle in the eternal struggle
between the gods and demons, this time fought on earth … as in III.92.; VI.20.5.; XII.34.13–
21.; and in the descriptions of the conflict as a ‘sacrifice of battle’ (e.g., V. 139.) taking place
on the altar of Brahmā (III.81.177–78 and III.129.22.).” Ibid., 207. Sullivan rightly notes the
The Divine Androgyne 287
Looking at the epic’s explicit citations of a divine identity for Bhīṣma, one
sees that he is depicted as an incarnation of Dyaus, or Dyaus Pitṛ, the Sky
Father of the Ṛg Veda. Certainly this is the deity who has appropriately
fatherly attributes to be the divine alter ego of Bhīṣma. Beyond this, how-
ever, the connection between Bhīṣma and Dyaus is tenuous and not
particularly meaningful. There is little mythology about Dyaus the Vasu
with which the character of Bhīṣma might interact, and the fact that the
motif remains undeveloped in the text indicates that the epic poets were
uninterested in it. The identification of Bhīṣma with this superseded,
retired god Dyaus Pitṛ seems to be an afterthought, as if Bhīṣma were
thought to need a divine figure with which to be associated because each
of the other important epic characters already was associated with a deva
or asura.51
Some comments are in order here. The idea of prādurbhāva, avatāra meaning
descent, incarnation, or aṁśa meaning a ray or particle of a god as the origin of
the characters of the epic is as important as the specific god who serves as this
original.52 Thus, not only Bhīṣma, but also many of the principal characters
have more than one “source” and some gods such as Dharma have more than
one “effect.” Mahābhārata 1.189 tells us that the five Pāṇḍavas are, in fact, five
Indras, while we know that Indra specifically sires only Arjuna (1.114.20–35),
the other four Pāṇḍavas being sired by other gods (e.g., Yudhiṣṭhira by the god
significance of this conflict for Indian literature in general, and for his analysis of Brahmā
in particular. Gonda calls this myth of the struggle between gods and demons for sover-
eignty “the central myth of Indian civilization.” Jan Gonda, Aspects of Early Viṣṇuism
(Utrecht: Oosthoek, 1954).
51 Sullivan, “The Epic’s Two Grandfathers, Bhīṣma and Vyāsa,” 207–8.
52 See André Couture, “From Viṣṇu’s Deeds to Viṣṇu’s Play, or Observations on the Word
avatāra as a Designation for the Manifestations of Viṣṇu,” Journal of Indian Philosophy 29,
no. 3–4 (2001): 313–26. Couture’s analysis suffers from his reliance upon Hacker (Paul
Hacker, “Zur Entwicklung der Avatāralehre,” Wiener Zeitschrift für die Kunde Südasiens 4
[1960]: 47–70), whose analysis is motivated in a theological anxiety over avatāra offering
a real rival to Christ’s incarnation. I cannot agree with Couture that “Hacker is right when
he says that the word avatāra gradually replaced prādurbhāva as the designation for
Viṣṇu’s manifestations.” Couture, “From Viṣṇu’s Deeds to Viṣṇu’s Play,” 313. Hacker’s analy-
sis is dependent upon assigning dates to different parts of the Mahābhārata, something
that I do not believe can be done. Couture also overlooks the fact that Hacker is con-
cerned to demonstrate the lateness of avatāra, which he interprets, in contrast to
bhārāvataraṇa, to mean a salvific incarnation, a notion that threatens his sense of Chris-
tian exclusivism. Thus, his analysis is based on the a priori assignation of terms to layers
and layers to dates, based on the presumed lateness of certain terms (i.e., it is circular).
288 Adluri
other characters, themes, and literary projects within and outside the epic nar-
ration.57 More to our point here, Bhīṣma too recalls Brahmā in many ways, and
we should not easily set aside enquiry along this line. I will return to the signifi-
cant way in which Bhīṣma recalls Brahmā in a later section, but here I would
like to point out three ways in which Bhīṣma functions as Brahmā:
1. As the epithet pitāmaha suggests, he is not only the ancestor (in some
way) of the Kuru heroes, equally disposed to both sets of cousins, but he is
exactly so as Brahmā is to the devas and asuras. In this Bhīṣma represents a
certain ancestral point: not quite the origin, but closely related to it. Brahmā
too is a pitāmaha only through the functions of Prajāpati and other interven-
ing and complicating factors.
2. Brahmā is given the Veda by Nārāyaṇa, of which the pitāmaha is the stu-
dent and custodian. Likewise, in Brahmaloka, Bhīṣma is the student of the
diverse modes of knowledge,58 and thus of Veda in a metaphorical sense.59 Not
only in the Śānti and Anuśāsana parvans, but elsewhere as well, he is a reposi-
tory of wisdom ordinarily hidden from the eyes of mortals.60
3. Finally, even as Brahmā can create human kind through some device
other than procreation, so also Bhīṣma, although he wins the brides in the
bride contest, hands over procreation to his brother Vicitravīrya first and then
ultimately to the Nārāyaṇa figure, Vyāsa. In these ways, then, Bhīṣma is a
Brahmā figure.
What enrichments does the Dyaus connection bring? I have previously
argued that the epic is extremely aware of the matutinal character of the begin-
ning of the epic cycle, populated by the figures of dawn such as Saramā
(Mahābhārata 1.3), the Aśvins (1.3.55–70), etc., and such themes as awakening
(Āruṇi; 1.2.25–30), opening of eyes (Upamanyu; 1.3.55–60), and the tale of fire
examines how he works these things out in collaboration with the other ‘dark’ figures: not
only Kṛṣṇa, Draupadī, and Arjuna, but his mother Kālī [another word for ‘Black’] Satya
vatī.” Alf Hiltebeitel, “Two Kṛṣṇas, Three Kṛṣṇas, Four Kṛṣṇas, More Kṛṣṇas: Dark Interac-
tions in the Mahābhārata,” in Essays on the Mahābhārata, ed. Arvind Sharma (Delhi:
Motilal Banarsidass, 2007), 101–2.
57 On Vyāsa’s literary functions, see Alf Hiltebeitel, Rethinking the Mahābhārata: A Reader’s
Guide to the Education of the Dharma King (Chicago: University of Chicago Press, 2001).
58 Like Brahmā, Bhīṣma is a student of Viṣṇu’s, though of his Paraśurāma form rather than
Hāyaśiras.
59 Alf Hiltebeitel, “Bhīṣma’s Sources,” 261–78.
60 Bhīṣma is the one who discloses the identity of Arjuna and Kṛṣṇa as Nara-Nārāyaṇa in the
Udyogaparvan (see Mahābhārata 5.48.1–29).
290 Adluri
61 See Vishwa Adluri, “The Perils of Textual Transmission: Decapitation and Recapitulation,”
Seminar 608, The Enduring Epic: A Symposium on Some Concerns Raised in the Mahābhā
rata (April 2010): 48–54.
62 See Śatapatha Brāhmaṇa 7.1.2.6, 10 and 6.5.2.3. In the Mahābhārata, the Vasus are listed as
Āpa, Dhruva, Soma, Dhara or Dhava, Anila, Anala, Pratyuṣa, Prabhāsa (dharo dhruvaś ca
somaś ca ahaś caivānilo ’nalaḥ | pratyūṣaś ca prabhāsaś ca vasavo ’ṣṭāv iti smṛtāḥ || ;
Mahābhārata 1.60.17), that is, Water, the Pole-Star, the Moon, Earth, Wind, Fire, the Dawn,
and Light.
63 A. Hillebrandt, Vedic Mythology, vol. 2, trans. S.R. Sarma (Delhi: Motilal Banarsidass, 1990),
413, n. 294.
The Divine Androgyne 291
being connected with Agni, the Rudras with Indra, the Ādityas with Varuṇa,
the Maruts with Soma, and the Sādhyas with Brahmā (cp. RV. 10, 9).”64 In the Ṛg
Veda, the Vasus appear with some frequency. The occurrences divide into two
types of uses: (1) “Vasu” in the sense of a good, kind, Good Lord occurs about
twenty-eight times and (2) “Vasus” as a class of Gods, occurs nearly sixty times.
In the Upaniṣads, the Vasus are most significant in the Bṛhadāraṇyaka
Upaniṣad:
Yājñvalkya said, ‘These are but the manifestations of them, but there are
only thirty-three gods.’ ‘Which are those thirty-three?’ ‘The eight Vasus,
the eleven Rudras and the twelve Ādityas—these are thirty-one, and
Indra and Prajāpati make up the thirty-three.’ (Bṛhadāraṇyaka Up. 3.9.2;
Mādhavānanda trans.)
‘Which are the Vasus?’ ‘Fire, the earth, air, the sky, the sun, heaven, the
moon and the stars—these are the Vasus, for in these all this is placed;
therefore they are called the Vasus.’ (Bṛhadāraṇyaka Up. 3.9.3; Mādhav
ānanda trans.)
Yājñvalkya said, ‘These, the three hundred and three etc., are but the man-
ifestations of them, the thirty-three gods. But really there are only
thirty-three gods.’ ‘Which are those thirty-three?’ The reply is being given:
‘The eight Vasus, the eleven Rudras and the twelve Ādityas—these are
thirty-one, and Indra and Prajāpati make up the thirty-three.’ (Com
mentary on Bṛhadāraṇyaka Up. 3.9.2; Gambhīrānanda trans.)
‘Which are the Vasus?’ The identity of each group of the gods is being
asked. ‘Fire, the earth, etc.,—from fire up to the stars are the Vasus.
Transforming themselves into the bodies and organs of all beings,
which serve as the support for their work and its fruition, as also
into their dwelling-places, these gods help every being to live, and they
themselves live too. Because they help others to live (Vas), therefore
they are called Vasus.’ (Commentary on Bṛhadāraṇyaka Up. 3.9.3; Gam
bhīrānanda trans.)
To live, dwell, pass the night, with derivatives meaning “to be.” (Oldest
form *ə2wes-.) 1. O-grade (perfect tense) form *wos-. was, from Old
English wæs, was, from Germanic *was-. 2. Lengthened-grade form *wēs-.
were, from Old English wæ̅ re (subjunctive), wæ̅ ron (plural), were, from
Germanic *wēz-. 3. wassail, from Old Norse vesa, vera, to be, from
Germanic *wesan. 4. Perhaps suffixed form *wes-tā-. Vesta, from Latin
Vesta, household goddess. 5. Possibly suffixed variant form *was-tu-.
astute, from Latin astus, skill, craft (practiced in a town), from Greek
astu, town (< “place where one dwells”). 6. Suffixed form *wes-eno-.
divan, from Old Persian vahanam, house. [Pokorny 1. ṵes- 1170.]
To clothe. Extension of eu-. 1. Suffixed o-grade (causative) form *wos-eyo-.
wear, from Old English werian, to wear, carry, from Germanic *wazjan.
2. Suffixed form *wes-ti-. vest; devest, invest, revet, travesty, from
Latin vestis, garment. 3. Suffixed form *wes-nu-. himation, from Greek
hennunai, to clothe, with nominal derivative heima, hīma (< *wes-mn̥ ),
garment. [Pokorny 5. ṵes- 1172.]65
three worlds as the Left-handed Archer, has been ordained to be the death of
Bhīṣma in due time.”66 Bhīṣma’s connections to the Vasus, at any rate, should
not be dismissed as an insignificant detail, especially when considering his
relationship to Ambā. In the Nārāyaṇīya, the text refers to the seven Citra
śikhaṇḍins, who compose the seven prakṛtis (Mahābhārata 12.322.26–28). Now
the names of these “colorful sages” recalls the name of Ambā upon her rebirth
as a man, and in Sāṁkhya cosmology animating Prakṛti is the name for the
female principle, the complement of Puruṣa. Further, in later theistic tradi-
tions, the Goddess is identified precisely with animating prakṛti. Thus, not
only as a Brahmā figure, but also as a Vasu, Bhīṣma stands in a complex and
never quite stable relationship to the Goddess. This relationship operates at
several levels, of which their personal antagonism is perhaps the least signifi-
cant. Thus, not only genealogically, but in terms of their dharma, kāma, svarga,
the two figures mark out two very different terrains, ones we may identify as
the terrains of pravṛtti and nivṛtti respectively. I discuss in greater detail how
these two terrains relate to each other in the next but one section, but before
I turn to this section, I would like to note one further point. In the Bhīṣmaparvan,
at the precise moment of the fateful encounter between these two opposed
forces, Bhīṣma says to Ambā: “You are still that Śikhaṇḍin that the Ordainer
made you, therefore I will not fight. Knowing this, do what you will” (kāmam
abhyasa vā mā vā na tvāṁ yotsye kathaṁ cana | yaiva hi tvaṁ kṛtā dhātrā saiva
hi tvaṁ śikhaṇḍinī ||; Mahābhārata 6.104.41). These are the last words he utters
to her, and they are significant because they clearly demonstrate that it is upon
the pravṛtti perspective that Bhīṣma ultimately stands, and falls.
dharma. But the beginning is never simple, and inception is always a struggle
for emergence. Thus to understand Bhīṣma’s role as a pravṛtti figure, we also
need to understand the resistance he will undergo. Here we must relate Bhīṣma
to that anti-pravṛtti force, Rudra. It is my argument that the Ambā-Upākhyāna
is a coded commentary on what Stella Kramrisch calls the “primordial drama”
of existence:68 the opposition of Brahmā and Rudra. I will do this in two stages:
(1) a rough sketch of the Rudra-Brahmā or Rudra-Nārāyaṇa conflict in the
Nārāyaṇīya; (2) a detailed interpretation of the story of Ambā as the story of
Rudra-Śiva’s counterpoint to Bhīṣma-Brahmā in the epic (in the next section
“This Sex Which is Not One”).
Rudra is mentioned in the Nārāyaṇīya at 322.36–38, 327.7–10, 327.30–32,
327.69–70, 328.11–12, 328.16–28, 328.32, 329.14d–15, 329.49, 330.42–49, 330.51–
66, 330.68–71, 336.15–17, 337.61–64, and 338.11–25. The largest group of these
references concerns the birth of Rudra from Nārāyaṇa.69 The birth of Rudra is
almost always mentioned in the same context as the birth of Brahmā,70 with a
smaller group referring to the birth of Rudra from Brahmā.71 A fourth group
refers to the conflict between Nārāyaṇa and Rudra,72 while another empha-
sizes the complementarity of the gods Nārāyaṇa and Rudra.73 Finally, a
reference (Mahābhārata 12.338.11–25) recounts the story of how Rudra received
instruction from Brahmā on Nārāyaṇa. I first present the relevant passages:
68 Stella Kramrisch, The Presence of Śiva (Princeton, NJ: Princeton University Press, 1981), 58
and passim.
69 E.g., Mahābhārata 12.322.36–38, 12.327.30–32, 12.327.69–70, 12.328.11–12, 12.328.16–28,
12.330.51–66, 12.330.68–71.
70 E.g., Mahābhārata 12.322.36–38, 12.328.11–12, 12.330.51–66.
71 E.g., Mahābhārata 12.327.30–32, 12.327.69–70, 12.328.16–28.
72 E.g., Mahābhārata 12.329.14d–15, 12.330.42–49, 12.330.51–66.
73 See Mahābhārata 12.328.16-28 and 12.330.51–66.
74 Note the close proximity of the textual universe to the creative and destructive functions.
Creation of the text is cited alongside emergence of Rudra and Brahmā (cf. Ṛgveda
10.129.4: desire as the first seed of the mind)
The Divine Androgyne 295
friendship and equality with Rudra, Nara and Nārāyaṇa continue their
austerities.
330.68–71: Kṛṣṇa reveals to Arjuna that the figure that proceeded before
his chariot was Rudra/Kapardin. He is Time, born of Kṛṣṇa’s wrath, and
had already slain the enemies slain by Arjuna. He is to be worshipped as
Umāpati and as Hara.
This brief summary of the theology of the Trimūrti as it occurs in this one com-
pact section of the Mahābhārata thus provides us with rich perspectives to
study the Ambā narrative. I focus here on the most important points:
1. Rudra and Nārāyaṇa are identical in essence, but differ at the level of
function.
2. Rudra and Brahmā both emerge from Nārāyaṇa and represent different
aspects of the God.
3. Rudra and Brahmā are respectively associated with temporal functions
of laya and śṛṣṭi.
4. Rudra is Time; he is born of Nārāyaṇa’s wrath; his function is to end the
cosmic cycle.
5. This end takes the form of the destruction of the sacrifice (Dakṣa’s
sacrifice), but the energy of the destruction cannot be contained and it
speeds on to Badari āśrama.
6. Nārāyaṇa transcends both laya and śṛṣṭi; as immutable and unmanifest
Being, he does not participate in cosmology.
7. Nārāyaṇa in his form as one half of the Nara-Nārāyaṇa pair must oppose
Rudra and bring destruction back into a balance with the creative force
of Brahmā.
8. All three gods, that is, Brahmā, Rudra, and Nārāyaṇa of the Nara-
Nārāyaṇa pair, cooperate to maintain this balance; Rudra must be
restrained for śṛṣṭi to occur,75 but ultimately his is the force that brings
Brahmā’s cycle of creation to an end.
75 As the diremptive force, Rudra is coeval with the universe’s inception. Although this force
is checked in the beginning of the cycle, the diremption slowly rises as an undertow and
ends the universe.
The Divine Androgyne 297
Thus, the narrative of Rudra in this section of the epic is closely woven together
with the narrative of not only Brahmā but also Nārāyaṇa. Further, the text
underscores the antagonism between the two in all the passages where it refers
to their birth: Rudra is born of Nārāyaṇa’s wrath, Brahmā of his grace or, alter-
natively where the text refers to Rudra’s birth from Brahmā, he is born of
Brahmā’s wrath. This antagonism between the two comes to the fore in the
three references to conflict with Rudra, although, paradoxically, it is always
Nārāyaṇa whom Rudra battles. However, in the longest group of lines refer-
ring to this conflict (Mahābhārata 12.330.42–49, continued in 330.51–66) the
origins of the conflict are placed at Dakṣa’s sacrifice, with the conflict with
Nārāyaṇa presented as the aftermath or aftershock of Rudra’s destruction of
the sacrifice.
The interweaving of all three major gods of the Hindu trinity, especially in
the context of a narrative of conflict, has led many scholars to assume that the
Nārāyaṇīya is a sectarian text.76 Thus, scholars have sought to explain the dif-
ferent versions of Rudra’s birth (and hence variation in his hierarchy relative to
Brahmā) as evidence of a re-working of the narrative. Further, scholars argue
that the reconciliation with Nārāyaṇa and the eulogization of Nārāyaṇa as the
Supreme Being demonstrate that a Rudra theology was in the process of being
integrated and subsumed into a Nārāyaṇa theology.77 However, once we look at
the text itself and understand its project, these views prove to be naïve, ill-
considered, and pointless. For example, the text not only introduces other gods
such as Rudra, Brahmā, and Indra but also relates Nārāyaṇa himself to four
hypostases: Vāsudeva, Aniruddha, Pradyumna, and Saṁkarṣaṇa. Further, the
text itself contains explicit warnings against reading the epic as a sectarian
work. Thus, at Mahābhārata 12.328.19 we are told that Rudra is to be known as
Nārāyaṇa’s own self; he who worships Rudra worships Nārāyaṇa also. At
12.330.64 we are told that he who knows the one knows the other; there is no
difference between Nārāyaṇa and Rudra. Not only this, but at 12.329.15d we are
also told that Rudra obtained his blue throat when gripped by Nārāyaṇa. In
case one takes this conflict too literally, as two different and separate gods
fighting it out, then one is disappointed, because at 15c, we are told he obtained
his blue throat from the bite of a snake. The blue throat is a cipher. So is the
conflict. The unity and separation of Nārāyaṇa and Rudra are ways of speaking
about the process(es), the indivisible dyad, that holds together both singular
Being and plural Becoming.
Once we set aside the sectarian interpretation, light can also be shed on the
Rudra-Brahmā antagonism. Although it is true that this antagonism is less pro-
nounced than the Rudra-Nārāyaṇa antagonism, one must note that Nārāyaṇa
only intervenes as a restraining force on Rudra. The destructive force of Rudra
is initially directed against Dakṣa’s sacrifice, that is, the creation of Brahmā.
Nonetheless, because Dakṣa cannot defend himself and Rudra’s resorptive
force threatens to spin out of control and absorb even the hermitage of Badari,
Nārāyaṇa must battle Rudra. In this conflict, note that it is Brahmā who even-
tually brings about peace between the two, by revealing to Rudra the secret of
Nārāyaṇa. Nārāyaṇa is none other than Rudra’s self; both he (Brahmā) and
Rudra have emerged from Nārāyaṇa as opposed and opposing aspects of his
being. Note also that in absorbing the force of Rudra’s spear, which had reduced
Dakṣa’s sacrifice to ashes, Nārāyaṇa creates the conditions for Brahmā’s
renewed creation. Rudra’s destructive force thus represents a check on the cre-
ative, inceptive power of Brahmā. Now the word for “check” is nirodha, and one
of Nārāyaṇa’s names is Aniruddha or “one who removes [the source of]
model, which derives from Paul Hacker’s influential notion of “inclusivism,” is not a viable
model for thinking about either the epics or the Purāṇas, in spite of the immense produc-
tivity (in creating religious and textual “histories”) it has encouraged. For the historical
issues informing Hacker’s work, see Joydeep Bagchee and Vishwa Adluri, “The Passion of
Paul Hacker: Indology, Orientalism, and Evangelism,” in Transcultural Encounters between
Germany and India: Kindred Spirits in the Nineteenth and Twentieth Centuries, ed. Joanne
Miyang Cho, Eric Kurlander, and Douglas McGetchin (New York: Routledge, 2013), 215–29.
The Divine Androgyne 299
obstruction.” Brahmā is the only one who knows this form of Nārāyaṇa, and
thus it is appropriate that he is the one who reveals to Rudra the secret of his
origin from Nārāyaṇa.
Thus, one can see that what is being worked out in the Nārāyaṇīya is not the
story of a conflict between two sectarian gods, but fundamental ontological
processes. “Rudra” and “Nārāyaṇa” are ways of talking about the impulses and
inertias operative in the emanations and recoils of the cycles of the phenom-
enal universe. Thus, neither a theory of sectarianism78 nor one of syncretism79
will be able to do justice to the text: at stake is not the combination of different
or contradictory beliefs into a more or less cohesive system, but the articula-
tion of a dynamic conception of ontological processes. As such, the text is not
just cohesive or coherent, but has to be cogent. Śṛṣṭi and pralaya are both
events, but the conditions for them are ever present. We may call these the
traits of natality and mortality, whose interaction is responsible for the phe-
nomenal presencing of all things. Thus śṛṣṭi must address the undertow of laya
and vice versa. Rudra-Śiva and Brahmā represent these forces. Hence in śṛṣṭi
Rudra-Śiva must be appeased or restrained as part of Aniruddha’s removal of
nirodhana. Thus in the Nārāyaṇīya one sees Nārāyaṇa fighting Rudra. But
Rudra remains an undertow and dominates in laya scenes, once again making
it possible for Nārāyaṇa to destroy the universe as we see in the Mārkaṇḍeya
episode (Mahābhārata 3.186–87). The Nārāyaṇīya, which maps out a cosmog-
ony of the relation of the One to the many, shows—or rather ends with—Rudra
“asking” Brahmā how many puruṣas there are, that is, allowing Brahmā to
determine and guide plural existence, or rather allowing him to create
(12.338.22–23).
The narrative of Rudra and Nārāyaṇa’s conflict itself occurs in a chapter on
the secret import of nirukti. Brahmā himself, that is, the Creator or (by associa-
tion) the creation, is described as niruktagaḥ, “one who is grasped by nirukti”
or, if we expand on the implicit meaning, “one whose secret is grasped (or
penetrated) by (grasping) nirukti.” Thus, the text itself is giving us a clue: to
understand the universe is to understand the secret identities (simultaneously
concealed and disclosed in the names) of its agents, and this means, ultimately,
to understand the agents: (1) In their trifunctionality as creator, preserver,
78 Here one can cite almost any German Mahābhārata scholar writing in the last two centu-
ries. For a criticism of the “sectarian” hypothesis, especially as applied to the Nārāyaṇīya
in the work of Oberlies, Schreiner, and Malinar, see Vishwa Adluri, “Philosophical Aspects
of Bhakti in the Nārāyaṇīya,” in The Churning of the Epics and Purāṇas at the Fifteenth
World Sanskrit Conference, ed. Simon Brodbeck, Alf Hiltebeitel, and Adam Bowles (New
Delhi: Rashtriya Sanskrit Sansthan and D.K. Printworld, in press), 127–54.
79 This term is used by Nicholas Sutton, Religious Doctrines in the Mahābhārata (Delhi: Moti-
lal Banarsidass, 2000).
300 Adluri
80 I use the words “preserver” and “preservation” because they have become popular; actu-
ally, however, the function is not so much “preservation” but a kind of temporal deferral
and a bringing into balance of the Brahmā/Raudra aspects of Being (Nārāyaṇa) itself.
81 Bhīṣma remains celibate; he is a renowned warrior; he is the authority on dharma.
82 Recall that all the other Pāṇḍavas are aṁśas of gods; Bhīṣma is the only Kaurava figure
who is a living god, self-conscious of his status, able to choose his hour of death, and
raised in heaven.
83 Of course, by this I do not mean there was some core “war narrative” independent of the
theological narrative: as the Nārāyaṇīya passages should have made clear by now, the the-
ology makes use of conflict as an essential topos or, to put it another way, cosmology
unfolds via strife. War is the literary figure the epic poets exploit in order to work out this
theology.
The Divine Androgyne 301
With these points in mind, let us now return to the story of the abduction
of Ambā as related in the Ādiparvan and the Udyogaparvan. Van Buitenen
considers this episode to be based on a historical event (and hence part of
the “original” epic).85 A closer look, however, shows that the episode is itself
84 Both commit irrevocable self-sacrifice, both become empty and thus receptacles for
Śiva. But when Ambā commits immolation, she has already reduced herself through
tapas; moreover, Gaṅgā’s curse has already taken over half her body. When Śiva appears,
she asks to be able to kill Bhīṣma, but there is no evidence Śiva enters her as he does with
Aśvatthāman. So while Aśvatthāman is more completely Śiva; Ambā retains her feminin-
ity and thus her distinction from the masculine aspects of Śiva. She goes as far as ardhanārī
(in Śikhaṇḍin). But her masculinity will revert back to the Yakṣa, and her femininity
remains intact after her death. Bhīṣma, who is a Brahmā figure, remains in the role of a
son to Ambā. Thus, when Ambā/Śikhaṇḍin is able to be penetrated, it is through Śiva’s
sword and through the complete possession of Aśvatthāman. If we do not see these dis-
tinctions between Aśvatthāman and Śikhaṇḍin we will miss these interactions between
Śiva and the Devī.
85 Van Buitenen’s views are spelled out clearly in his introduction to the translation of
the Udyogaparvan. In the context of a critique of Madeleine Biardeau’s views of the
epic as the restatement, in mythic form, of the purāṇic cosmogony undergirded by the
yogic self-reabsorbtion, he defends a long period of compositional history for the Mahā-
bhārata. Since his views of the Ambā-Upākhyāna are characteristic of his understanding
of the epic’s history (and simultaneously also serve as a good summary of the historical
approach to decoding myth), they are restated here. After rejecting, mainly through cari-
cature, Biardeau’s view that one must resist any “hypothèses d’ordre historique” and that
one must understand the elements of its symbolism “dans les structures idéologiques
d’ensemble,” Van Buitenen lists four ways in which the expansion of the Mahābhārata
from an “epic” to a “purāṇa” might have taken place (these are “An Old Myth and Its
Sequel”; “New Myth”; “An Old Legend and Its Sequel”; and “New Legend—A Sequel”), with
the Ambā-Upākhyāna being an example of the last (i.e., a new legend, created as a sequel
to an old one). J.A.B. van Buitenen, “Introduction [to Book 5. The Book of the Effort],” in
The Mahābhārata: 4. The Book of Virāṭa; 5. The Book of the Effort (Chicago: Chicago Univer
sity Press, 1978), 142–73. He writes that the legend of Ambā, up to the story of Ambā’s
self-sacrifice and immolation, is “fairly straightforward.” Thereafter, however, “the story …
gets bogged down in talk” with the introduction of Rāma Jāmadagnya and the reference
to his “brother-Vasus” and it gets “positively droll” with the reference to the childless king
302 Adluri
rich in theological details. As Scheuer notes, the name “jyeṣṭhā” used of Ambā,
which van Buitenen translates simply as “eldest” or “oldest,” is a name of the
Goddess.86 Further, in a passage from the Udyogaparvan, where Bhīṣma clari-
fies why he could not take on the niyoga obligation, he refers to his vow of
celibacy as follows: ity uktaḥ prāñjalir bhūtvā duḥkhito bhṛśam āturaḥ | tebhyo
nyavedayaṁ putra pratijñāṁ pitṛgauravāt | ūrdhvaretā hy arājā ca kulasyārthe
Drupada and his sexually ambiguous son/daughter Śikhaṇḍin. Ibid., 175. After presenting
his “monologue intérieur” of what might have taken place inside the ingenious storyteller’s
mind (basically a conversation in which Van Buitenen makes up adventitious and uncon-
vincing reasons why a storyteller, especially one without literary genius or creative intent,
might have chosen to identify the ambiguously gendered warrior Śikhaṇḍin with Ambā),
he concludes: “The point I wish to make with this monologue intérieur of a storyteller try-
ing to tie up one loose end of the Mahābhārata is that within the half millenium of the
composition of the text, a minor element … could create a new legend, an instant tradi-
tion …. If we were to take this story seriously as simultaneous to the epic portions of the
Mahābhārata, we would ultimately have to lay the death of Bhīṣma at the fragrant door
of the Yakṣa Sthūṇakarṇa’s mansion in a wood off Kāmpilya. I, among the trees of differ-
ent ages find this view absurd.” Ibid., 177–78. The problem with this explanation is that
it explains nothing: it does not tell us why Śikhaṇḍin has to be invulnerable to Bhīṣma,
nor why this invulnerability could not be explained by other means (say, superior armor
or a boon from a god), nor why Ambā had to return at this juncture in the text. Nor does
it tell us why Rāma had to fail and Bhīṣma’s brothers, the eight Vasus, had to reappear at
just this juncture (Van Buitenen merely comments: “it all sounds epigonic”; ibid., 175). Van
Buitenen’s explanation only appears probable to him: it is neither undergirded histori-
cally nor based on philological investigations, and reflects more his lack of comprehen-
sion of—indeed, discomfort with—“Purāṇic Hinduism” than anything else (see ibid., 152
for his statement that “not all scholars are interested in Purāṇic Hinduism” and 144 for the
observation that “no doubt, Hindu symbolification of the epic started quite early and has
become an overlying part of the text itself. This process went on in the Purāṇas, and this
specious “continuity” between “epic” and the Purāṇas can easily lead one to lay a Purāṇa-
Hindū interpretation upon the Bhārata, if one resists all ‘hypothèses d’ordre historique’”).
But we must wonder: is it Biardeau who is laying a “Purāṇa-Hindū interpretation upon the
Bhārata” in the name of myth or is it Van Buitenen who is laying a “European/Indo-Euro-
pean” interpretation upon the Mahābhārata in the name of another myth—the heroic,
bardic oral epic—now raised to the level of “history”? At the very least, we should not
overlook the role of supersessionism in Van Buitenen’s interest in postulating a specious
discontinuity between epic and the Purāṇas, a discontinuity that makes it possible to
argue that it is Western “critical” scholarship and not Purāṇic Hinduism that is the true
fulfillment of and has correctly grasped the meaning of the original epic.
86 “Les textes précisent bien qu’Ambā est l’aînée des trois princesses de Kāśī; elle est ‘jyeṣṭhā’
(1.96.47, 51/102.60, 64). Or ‘jyeṣṭhā’ est une épithète de Durgā (voir Harivaṃśa no. 8.4/2.3.4)
considérée comme soeur aînée de Yama (jyeṣṭhā Yamasya bhaginī) …” Jacques Scheuer,
“Śiva dans le Mahābhārata: l’histoire d’Ambā/Śikhaṇḍin,” Puruṣārtha 2 (1975): 82.
The Divine Androgyne 303
punaḥ punaḥ ||.87 Now the word Van Buitenen translates as “celibate” is in
fact ūrdhvaretā, and this is a very unusual word for “celibacy.” Apte lists the
following words as much more common terms for “celibacy”: anudvāhaḥ,
adārparigraha, avivāhaḥ. In contrast, Monier-Williams and the Petersburg dic-
tionary both clarify ūrdhvaretas as “retaining the semen above,” and the term
is a frequent epithet of sages or ascetics. Although the term occurs as early as
Taittirīya Āraṇyaka (10.12), it is especially associated with Śiva.88 Commenting
on Śiva’s status as the ithyphallic deity, Doniger notes:
In the longer passage from which this verse is excerpted, Bhīṣma attempts to
convince his mother Satyavatī not to force him into this obligation. Again, his
words are significant: tato ’haṁ prāñjalir bhūtvā mātaraṁ saṁprasādayam |
nāmba śaṁtanunā jātaḥ kauravaṁ vaṁśam udvahan | pratijñāṁ vitathāṁ
kuryām iti rājan punaḥ punaḥ || viśeṣatas tvadarthaṁ ca dhuri mā māṁ
niyojaya | ahaṁ preṣyaś ca dāsaś ca tavāmba sutavatsale || (Mahābhārata
5.145.32–33). The use of the word amba (for ambā) for his mother twice in this
passage is significant, especially since he has just called her “Kālī” twice in the
preceding verses in his narration of the incident to Duryodhana (5.145.18, 29).
“Ambā” is of course his mother Satyavatī, but it also plays on the name of the
87 Mahābhārata 5.145.31; cf. also Ādiparvan, App. 1, no. 55, lines 12–13: aham apy ūrdhvaretā
vai nivṛtto dārakarmaṇi | na saṁbandhas tad āvābhyāṁ bhavitā vai kathaṁ cana ||. I owe
this citation to Scheuer.
88 See Mahābhārata 13.83.41–47 for the story of how Śiva acquired this name.
89 Wendy Doniger O’Flaherty, Śiva: The Erotic Ascetic (London: Oxford University Press,
1973), 9–10. The reference is to the birth of Skanda, recounted at Mahābhārata 3.213–14
and 9.43. Note, however, that the word for “seed” here is tejas and not retas (the only time
retas is used, 3.214.15b, the reference is to Agni). In 13.83, the gods, concerned that the off-
spring of Rudra and Umā will burn down the universe, beg him not to procreate. Śiva
draws up his semen, earning him the epithet ūrdhvaretāḥ (13.83.47c). Doniger misunder-
stands Śiva’s ability, as the primordial uncreated Being, to remain within himself, while
also lending his energy to the creation (cf. 13.83.52: a portion of Rudra’s tejas falls to the
ground).
304 Adluri
90 bhīma uvāca |
vadhārthaṁ yaḥ samutpannaḥ śikhaṇḍī drupadātmajaḥ |
vadanti siddhā rājendra ṛṣayaś ca samāgatāḥ ||
yasya saṁgrāmamadhyeṣu divyam astraṁ vikurvataḥ |
rūpaṁ drakṣyanti puruṣā rāmasyeva mahātmanaḥ ||
na taṁ yuddheṣu paśyāmi yo vibhindyāc chikhaṇḍinam |
śastreṇa samare rājan saṁnaddhaṁ syandane sthitam ||
dvairathe viṣahen nānyo bhīṣmaṁ rājan mahāvratam |
śikhaṇḍinam ṛte vīraṁ sa me senāpatir mataḥ || (Mahābhārata 5.149.29–32)
91 Apte, sv “bhid.”
92 duryodhana uvāca |
kathaṁ śikhaṇḍī gāṅgeya kanyā bhūtvā satī tadā |
puruṣo ’bhavad yudhi śreṣṭha tan me brūhi pitāmaha || (Mahābhārata 5.189.1)
93 tapaś cacāra yā ghoraṁ kāśikanyā purā satī |
bhīṣmasya vadham icchantī pretyāpi bharatarṣabha || (Mahābhārata 5.49.31)
The Divine Androgyne 305
his wives (Dakṣa’s daughters) are all mentioned,94 there are no references to
Rudra-Śiva as Dakṣa’s son-in-law. According to Annemarie Mertens, the story
first appears in the Brahmāṇda and Vāyu Purāṇas, while the Bhāgavatam is the
first to merge the death of Satī into the story of Dakṣa’s sacrifice.95 The longest
94 See Mahābhārata 7.173, 10.18, 12.274, and 13.145–146. Śiva is also frequently addressed with
epithets referring to his role in the destruction of the sacrifice (so, for instance, at 1.103.9c,
where he is called “the one who took Bhaga’s eyes,” bhaganetraharaṁ; see also 3.40.57a
[bhaganetranipātana] and 3.41.12a [bhaganetrahan]). In spite of the variations between
the versions, I do not think a chronological development such as that Mertens posits
(Annemarie Mertens, Der Dakṣamythus in der episch-purāṇischen Literatur: Beobachtun-
gen zur religionsgeschichtlichen Entwicklung des Gottes Rudra-Śiva im Hinduismus [Wies-
baden: Harrassowitz, 1998]) can be justified. Mertens argues for a four-stage chronological
development, beginning with the “earliest Mahābhārata version of the Dakṣa myth”
(Mahābhārata 7.173 and 13.145–46, the latter a “copy” of ch. 173 “albeit in abridged form”)
through the “expanded version” (10.18) and “two developed Śaiva versions” (12.274 and
App. 1, no. 28, the latter “the latest version of the myth within the Mahābhārata”). Ibid.,
16, 21, 25, 30, and 53. Mertens also posits an internal development within each version,
according to which each version was “redacted” more than once, sometimes in a “Śivaite
interest,” sometimes as the result of a “Viṣṇuite revision” (the latter again in two ways, i.e.,
either as the result of an “intensive” or a “superficial” Viṣṇuitization [Viṣṇuitisierung]). But
although Mertens speaks of a “Śivaite interest,” it turns out that these putative “Śivaite”
redactors were following different and sometimes competing interests. At one time it is
“a small or a young following of the rising outsider-god Rudra [that] struggles against the
‘orthodox’ Brahmandom and for the [its? his?] recognition in cult and society (ibid., 21;
apparently, this development, which Mertens detects in the text, “reflects the historical
situation”); at another, it is “to degrade the competing Viṣṇuism” (ibid., 24); at yet others,
it is either to “highlight a milder character [of Rudra] propagated (by the followers in
later texts)” (ibid., 27) or to “legitimate Rudra’s sacrilegious action and his new status in
the gods’ sacrifice resulting therefrom” (ibid., 28). Then again, it is not clear how these
putative “interests” or “motivations” permit Mertens to “bring the multiple versions of
the Dakṣa myth … into a chronological sequence,” since surely she has to presume the
sequence in order to identify the motivation? That is, whether we think version x was
revised in interest y or vice versa will depend on whether we think version x preceded
version y or the other way around. Mertens’s analysis suffers from an extremely simplistic
understanding of the way texts are composed. Even if it was a valid procedure to explain
the text only in terms of the social realities governing its production, Mertens’s work illus-
trates the exhaustion and destitution of historical criticism.
95 I quote the “scholarly” opinion, though I am skeptical of Mertens’s attempt to date the
Purāṇas; often, the dates ascribed to the texts rest on no more than her perception of
which version came “earlier” or “later,” the more developed form always being presumed
to have come later. Mertens follows in the tradition of Hacker’s “text-historical” scholar-
ship; she gratefully acknowledges his influence in her introduction with the words “With
what difficulties a text-historical [textgeschichtliche] investigation within the anony-
mous, composite [gewachsen] Purāṇa literature is associated Paul Hacker has formulated
306 Adluri
treatment of the Dakṣa myth comes in the Śiva Purāṇa. Viṣṇu Purāṇa, thought
to be one of the earliest, contains some elements of the story as well. And yet,
the Ambā story contains many of these elements, albeit in incipient form.
Ambā’s career begins at an aśvamedha sacrifice (Vyāsa’s niyoga ritual) and
ends in the apocalyptic raṇa yajña. This is a direct textual parallel to the later
Satī’s career, which begins with her being insulted at Dakṣa’s sacrifice and ends
in the cataclysmic destruction of that sacrifice, run over by the martial aspect
of a wrathful Śiva. The Mahābhārata knows that Umā was upset and Śiva
destroyed the sacrifice; the Mahābhārata also knows that Śiva oversaw both
the Kurukṣetra raṇa yajña and the Sauptika massacre.96 It is only in the attempt
to graft the Ambā narrative onto this Umā narrative in the Mahābhārata that
the Satī narrative becomes necessary in the Purāṇas. Thus, the goddess cycle in
the Mahābhārata is composed of the trio Ambā-ardhanārī-Umā. Elsewhere it
is Satī-Pārvati.
I make this point to show how wrongheaded it is to do a textual history
where we keep assuming interpolations into the Mahābhārata from the empir-
ical prejudices based on later textual traditions. Any proper reading of the epic
must make room for not only its originality but also its originary power.
Consequences derive from such prejudiced, unscientific ways of reading these
texts that are disastrous. If one grants the Mahābhārata its true power as a strī-
śūdra-veda as it purports to be, clearly the practice of Satī may at best be based
on Mādrī’s choice to join king Pāṇḍu on the pyre rather than on the Goddess
who lends her name to the practice. Because in the case of the Goddess, it is a
virginal form of the Goddess as Ambā who complicates and throws out of joint
the very institutions of gender and marriage, and works with Śiva to restore the
breach of creation through absorbtion or laya. But even so, she represents
pravṛtti’s mortal power. As a river filled with crocodiles she represents the epic
story itself, where Kṛṣṇa and Balarāma appear as dead crocodiles to Arjuna
towards the end of the epic (Mahābhārata 16.6.10).
The theology of the Mahābhārata, which has carefully cast Bhīṣma as a
pitāmaha or a Brahmā figure, is working with all of these elements. Indeed, as
we discover in the Nārāyaṇīya, Brahmā is the only one to have seen the
Aniruddha form of Nārāyaṇa97 and thus Bhīṣma provides a proper link between
the pravṛtti Kuru narrative by serving as the pitāmaha and the nivṛtti narrative,
which he provides to Yudhiṣṭhira in the form of an instruction. I have shown
grasps. Moreover, the thesis that Umā’s role in Mahābhārata 12.274 and App. 1, no. 28 rep-
resents necessary changes induced by the fact that “their desecration [i.e., the desecration
of the Brahmanic rituals] is a sacrilege that, on the one hand, expresses the protest of
Rudra’s following, [and,] on the other, cannot be carried out by the god who is [in the
process of] establishing himself” (ibid., 50) is dependent upon accepting the sequence of
versions she posits. Unfortunately, the Critical Edition provides support for only one of
her claims, namely, that the version in the Śāntiparvan, App. 1, no. 28, is late. Other than
this it does not support the kind of internal historical sequence she posits. And finally,
some of her claims are simply bizarre, as when she claims that “in this place [i.e., preced-
ing 7.173.42] the reading of the northern recensions [which mention Dakṣa “as a sacrificer
who sacrifices according to the prescription”] is to be preferred since the Mahābhārata 13
version, which is almost word-for-word identical with 7.173 and is dependent on it (see
below), names Dakṣa in all recensions and he is the sacrificer in all following versions.”
Ibid., 19.
97 Brahmā is thus the only one who knows the secret of Nārāyaṇa’s and Rudra’s identity. He
knows that Rudra is none other than the resorbtive force, whose other face is he himself,
the inceptive or creative force. Both he and Rudra have emerged from Nārāyaṇa as
opposed and opposing aspects of his being.
308 Adluri
98 Angelika Malinar seeks to demonstrate continuity between the Udyogaparvan and the
Gītā by focusing on the debates concerning kingship as enunciated by Duryodhana in
that book. I have discussed the problems with her argument elsewhere (see Vishwa
Adluri, Review of The Bhagavadgītā: Doctrines and Contexts, by Angelika Malinar, History
of Religions 50, no. 1 [2010]: 102–7). But even with her problematic interpretation of the
Gītā as a text on proper kingship, which requires various distortions, she cannot explain
the nivṛtti elements either in the Udyogaparvan, chiefly the Sanatsujātīya, as well as such
The Divine Androgyne 309
elements in the Gītā. She thus resorts to analyzing “layers” in these texts, which allow her
to choose those elements in the texts that match her argument.
99 Personal communication.
310 Adluri
taneously being instituted in the niyoga ritual with the other two mothers
begins here. It grows slowly by problematizing not merely marriage but even-
tually gender itself. This progressive involution reaches its zenith when the
androgyne Śikhaṇḍin helps the androgyne Arjuna-Bṛhannaḍā bring down
Bhīṣma.100 Then the failure of Paraśurāma demonstrates that she is genuinely
under the protection of Śiva and, in a sense, Śiva alone is her savior and bene-
factor. Her askesis is an important motif. Two points need to be made about the
vignette about Gaṅgā’s curse that Ambā becomes half a river, but also half a
woman: ardhanārī. As a river, she is possibly also the daughter of a mountain.
Her full apotheosis occurs as ardhanārī on the chariot of Arjuna—an image, as
Hiltebeitel has shown, of the chariot of the tripurāntaka Rudra.
100 I will not elaborate the Arjuna–Rudra identification, because Hiltebeitel has clearly
brought it out elsewhere. Let me provide the following quotes, which ought to suffice:
“With Uttara at the reins, Arjuna defeats the Kauravas first singly and then collectively in
a battle that prefigures his triumph at Kurukṣetra. It is here that the symbolism of the
dance first moves from the seraglio to the battlefield: says the poet, ‘it was as if Arjuna was
dancing in battle [pranṛtyadiva saṃgrāme]’ (4.57.9) This would seem a reminder of the
destructive dance of Rudra-Śiva, a suggestion reinforced by more direct allusions: ‘Thus
having caused [the Kauravas] to see his raudra self, he of the might of Rudra, held in
check for thirteen years, Pārtha, the son of Pāṇḍu, roamed about releasing the terrible fire
of his wrath on the sons of Dhṛtarāṣṭra.’ Arjuna’s raudra (“Rudra-like”) fire, held in check
for the microcycle of twelve plus one years, can only evoke Kālāgnirudra, Rudra-Śiva as
the Fire of Time, lord of the pralaya or ‘dissolution of the universe.’ One must also wonder
at the following: ‘While Pārtha was releasing his arrows, shooting with the right and left
hands, [his bow] Gāṇḍīva became, O king, like a whirling wheel of fire [agnicakramiva]’
(4.59.12) …. Then, on return to Indra’s abode, Arjuna and Mātali see another asura city
called Hiraṇyapura, identical in name with one of the three cities which Śiva destroys in
his conquest of Tripura. Endangered by this enemy, Arjuna bows to ‘the god of gods
Rudra’ (170.38) and fixes the Raudra weapon on his bow. He perceives the weapon as a
‘three-headed, nine-eyed man [puruṣa] with three mouths and six arms, hair alight with
blazing flames, his head surrounded by serpents darting their tongues’ (170.39). Before
releasing it, he bows again to the ‘three-eyed Śarva’ (170.41)—a name evoking Rudra-Śiva
as hunter (from śaru, “arrow”), and finally bows a third time ‘to the god who destroyed
Tripura’ (170.50). There can be no doubt that this pair of battles reinforces the dual iden-
tification of Arjuna with Indra and Śiva.” Hiltebeitel, “Śiva, the Goddess, and the Disguises
of the Pāṇḍavas and Draupadī,” 157, 158–59. Arjuna as Nara, Arjuna as the androgyne, and
Arjuna the hypermasculine hero maps out an ontology of gender that is not the thesis of
this paper. Moreover, this paper is restricted to Ambā and thus I am interested here only
in tracing this one character’s development. However, matters are more complicated and
exciting than I have shown here, for there seems to be a Raudra development in Arjuna as
well. These two paths cross precisely on the chariot when Bhīṣma is brought down by
Arjuna/Śikhaṇḍin.
The Divine Androgyne 311
I have argued in an earlier work that the ontological program of the Mahā
bhārata drives its architecture, beginning with the first books of the Ādiparvan
which are “initiatory” in nature and finally reaching a full manifestation of it
towards the end of the Mokṣadharmaparvan. The beginnings of this program
are beset with a certain set of difficulties; chief among them is the problem of
how to read the epic as it wishes to be read. This problem is overcome by inte-
grating into the Ādiparvan a set of hermeneutic, mnemonic, and pedagogical
tools.101 The end of this program in Book 12 generates a different set of prob-
lems; chief among them is the problem of how to bring the narrative back
down from the soteriological event. This event is the liberation of Śuka, son of
the “author,” Vyāsa. I have also argued that the Nārāyaṇīya inaugurates a
descent in the form of Nārāyaṇa’s four vyūhas, his several incarnations, and
the cosmological function which he delegates to Brahmā.102 Many sophisti-
cated literary strategies seal the rupture between Śuka’s one-way and upward
nivṛtti path and the epic’s need to continue on pravṛtti’s circular path:
Note also that we are told that the Mahābhārata is pañcama veda, that it is an
exhaustive text that deals with both pravṛtti and nivṛtti, and that it is an “Āstīka”
101 See Vishwa Adluri, “Frame Narratives and Forked Beginnings: Or, How to Read the Ādipar
van,” Journal of Vaishnava Studies 19, no. 2 (2011): 143–210 and see also Vishwa Adluri,
“Hermeneutics and Narrative Architecture in the Mahābhārata,” in Ways and Reasons for
Thinking about the Mahābhārata as a Whole, ed. Vishwa Adluri (Pune: Bhandarkar Orien-
tal Institute, 2013), 1–27.
102 Vishwa Adluri, “Plotinus and the Orient: Aoristos Dyas,” in The Routledge Handbook of
Neoplatonism, ed. Pauliina Remes and Svetla Slaveva-Griffin (New York: Routledge, 2014),
77–99, see especially 94, n. 13.
103 This is not found in the Critical Edition, because of S. K. Belvalkar’s decision to retain the
narration at the Vaiśaṁpāyana–Janamejaya dialogical level. However, the majority of
manuscripts do feature it, and it is surely significant.
312 Adluri
ontological text, that is, an Upaniṣad (see Mahābhārata 1.1.191). The text, fur-
ther, is all encompassing (nothing that is not here exists; 1.56.33cd and 18.5.
38cd), and presents reality as a mimetic cycle: the main characters descend
from a more true and more enduring dimension to act out the “reality” on the
ground. This is the devarahasya, the secret of the epic.104 Vyāsa holds this
secret by simultaneously engendering the text and its characters. Nature, as
presented in the text, is essentially fictional—and not just in the superficial
sense that Vyāsa is the author of this fiction. Rather the world of phenomena is
itself “fictional” in that phenomena lack any permanent or enduring being.
Thus Vyāsa, in the fifth Veda, describes all that the four Vedas do, but beyond
that, engenders the fifth: mokṣadharma śāstra and Śuka, the “fifth” disciple.
Both are Vyāsa’s creations. One gets only half of the Mahābhārata’s meaning
(and only the trivial half) if one neglects that Vyāsa is also the father of Śuka,
and that Śuka is engendered through sacrifice, without a mother. Begetting the
Kuru descendants on the widows of Satyavatī’s son is not Vyāsa’s only (pro)-
creative act. Fifth Veda, fifth son, also has a pravṛtti significance. On the way
“down” from the one to the many, Nārāyaṇa explicitly stops at the fourth vyūha,
Aniruddha. Henceforth, it is Brahmā, who has both four and five heads, who
continues the cosmological, descending, function. In Sāṁkhya terms, it is the
arising of the five “evolutes,” which become the five mahatattvas, and later, the
five “elements.”
With this background of “five” as a numerically significant cipher, let us
return to the architecture of the Mahābhārata. I have argued elsewhere that
the epic suffers a sudden interruption in the Mokṣadharmaparvan. The Anuśā
sanaparvan, Book 13, must thus “return” to the circular narrative, working
towards the beginning in the Ādiparvan. The circular narrative of eighteen
books is thus divided into three segments of six books each. The first seg-
ment leads to the fall of the Kuru pitāmaha Bhīṣma, the symbol of pravṛtti,
incarnation of a Vasu. The Vasus, I have argued, are the elemental forces of
104 This rahasya is itself twofold. The basic problem the epic is concerned with is “how
does the One become the many?” and gender is one of the most fundamental ways of
understanding this duality. Rudra works through a language of gender. As Hiltebeitel
has correctly divined, there are Śaiva motifs in Arjuna’s becoming the hermaphrodite
Bṛhannaḍā and in the fact that he becomes a dance teacher (Śiva is Naṭarāja, Lord of
Dance). He has also correctly seen that Arjuna is to Kṛṣṇa on the chariot as Rudra is to
Brahmā in the myth of the destruction of Tripura. Śiva is one-half of the composite divin-
ity Ardhanārīśvara, and this motif is always in the background of the Kuru war. By the
time Śikhaṇḍin stands next to Arjuna on the chariot, Rudra is standing next to Kṛṣṇa
as Nara. Thus, the Mahābhārata is going even beyond the devarahasya, the secret of the
double identity of all beings as simultaneously mortal and immortal, to an even higher
rahasya—this is the Nārāyaṇa rahasya or the ontological rahasya that the many is also
simultaneously, in essence and always, the One.
The Divine Androgyne 313
pravṛtti. The Bhagavadgītā marks the destination of the first segment. The sec-
ond segment leads to the opposite (i.e., not the fall of a pravṛtti figure) but
the ascent of a nivṛtti figure: Śuka. The Nārāyaṇīya marks the destination of
the second segment. Vyāsa, Yudhiṣṭhira, and Śaunaka lead the narrative back
from the one Being, Nārāyaṇa to the many, from Being to Becoming, from the
war front (where Bhīṣma lays) back to the kingdom, where a new rule is being
instituted, and from death to the genealogical remainder: Arjuna’s grandson
and Janamejaya’s father. But the “turn of the text” to the universe is stunningly
described in Vyāsa’s turn to the lovely nymphs in the Mandākinī valley, who
cover themselves, blushing, at the sight of Vyāsa. Further Vyāsa cries. What are
we to make of his tears? Are they simply the tears of a father’s dejection? Vyāsa,
whose narrative as a whole is a repudiation of Dhṛtarāṣṭra’s paternal attach-
ment to progeny, surely knows better than to weep for his son, who has not
merely gone to svarga, having fallen in battle, but to mokṣa! The answer is given
by the lord of tears, Rudra. He appears to pacify Vyāsa. But how? By turning
Vyāsa’s gaze to an image, a representation, a shadow of his son (Mahābhārata
12.32.37–38). From the argument above, it follows that both the son and the
text are images, and that Śuka now stands with Rudra, and Vyāsa ought to
stand by the other representation: the Mahābhārata. Thus the text continues
along its circular path.
Yudhiṣṭhira also contributes to the circular narrative’s rejuvenation. Not
only is he the new king but also his questions lead the narrative: questions,
especially, of how the One becomes the many (Nārāyaṇīya) and his brilliantly
pregnant question that holds the secret of desire being the cosmological-
genealogical-sacrificial-conflictual force (whether it is the man or woman who
derives greater pleasure from sexual intercourse; see Mahābhārata 13.12.1).
The Anuśāsanaparvan is thus the beginning of the last segment of six books,
and it will thus repeat, symbolically, certain central motifs such as the story of
Nahuṣa, which is the story of both a fall, as well as of a genealogical beginning.
Here I am concerned with one role Rudra plays in it. The Nārāyaṇīya contains
explicit warnings against reading the epic as a sectarian work. In it, Nārāyaṇa
says that he is Rudra, and Rudra is his own self. Not only this, but within the
Nārāyaṇiya there is also the story of a conflict between Rudra and Nārāyaṇa
whereby Rudra obtains his blue throat from the grip of Nārāyaṇa. In case one
takes this conflict too literally, as two different and separate gods fighting it
out, then one is disappointed. Rudra, in the same text, achieves his blue throat
from the bites of a snake. The blue throat is a cipher. So is the conflict. The
unity and separation of Nārāyaṇa and Rudra are ways of speaking about the
process(es), the indivisible dyad, that holds together both singular Being and
plural Becoming. Biardeau and Hiltebeitel more recently have been arguing for
understanding this process: the creative cosmological process, in its inception,
314 Adluri
as “the god turning towards the goddess.” This is twofold. It is Nārāyaṇa turning
first to Rudra. And/or, Rudra turning to Umā. Only after this first, high turning
does Brahmā appear as the first of beings. And only after Brahmā does Indra’s
function as the “renewer” appear. Thus Brahmā’s time contains many renewals,
that is, Indras. Thus towards the end of the Anuśāsanaparvan, Vāsudeva turns
to Rudra, narrating the deeds and the plural deeds of Rudra. Rudra in turn
becomes the very exposition of nativity.
Let me tie all the above elements together now. In the fifth Veda, after the
fifth son achieves mokṣa, the five elements are to be recreated and the fifth
Veda’s circular integrity must be maintained. This juncture betrays a “turn-
ing” and not a break. The turning is disclosed as the turning of the god. This
turning is the symbol of the power of nativity, which is crucial. Unlike life-
denying philosophies, the Mahābhārata pays homage to not one, but two
phenomenological traits: mortality and nativity. Rudra is the cipher not only of
mortality: Kapardin, but also of nativity. He is the infant in the lap of Umā as the
Pañcaśikha Śisya. Indra is “jealous” of the boy, and attempts to strike him with
his thunderbolt. Jealousy (asūya, a term that occurs quite frequently in the lat-
ter half of Book 13), implies a common desire. Indra wishes to be the “renewer.”
Is he, after all, not the one who slays Vala and lets the cows of dawn free? Is he
not the slayer of Vṛtra? Is he not the smasher of Uśa’s chariot? Is he not the god
who renews the cycle of the Mahābhārata by plotting the fall of Nahuṣa and
installing Vasu as the regent on earth? Is he not the god of renewal who gives
Vasu the pole of yearly renewal? This “ignorance” of Indra, a king among the
many beings, is ontological ignorance. It is useful to read the Kena Upaniṣad
here, where Umā appears as Brahma-Vidyā to dispel Indra Māghavan’s onto-
logical ignorance. There he is “humiliated” by almost being shown that even a
blade of grass, ontologically, is superior to the gods’ ontic power (Kena Up.
3.4–11). Umā proceeds to teach him about Brahman (Kena Up. 4.1).
But that is the Kena Upaniṣad. In the Upaniṣad of Vyāsa, which is the text
we are discussing, in the Anuśāsanaparvan, Indra’s arm is paralyzed when he
attempts to throw his thunderbolt at the boy in Umā’s lap. Brahmā must inter-
vene, as it is only to Brahmā that the Aniruddha form of Nārāyaṇa and the
truth of Rudra are revealed. It is this revelation that enables Brahmā to create.
Indra must await for the universe to unfold, before he can take up his function.
What could that knowledge be, that comes from Rudra to Brahmā? For this,
one must look at the Pañcaśikha brāhmaṇa who defeats the nāstikas in the
court of Janaka (see the Pañcaśikha-Janaka-Saṁvāda, Mahābhārata 12.307).
Janaka learns from him, and teaches the secret link between the one Being
and the many to Śuka. But that was the way up. In the Anuśāsanaparvan, we
are talking about the way down, and we are thus introduced to the Pañcaśikha
The Divine Androgyne 315
wisdom as the boy Rudra, in its trait of nativity, not the trait of dissolution. Let
us read the relevant text from Book 13 now:
Vāsudeva said: O Bhārata, the asuras with their (triple) cities were burnt
by Rudra with that arrow which had solar effulgence and whose energy
resembled that fire which appears at the end of the cycle (to consume all
beings).
Seeing that child with five locks of hair lying in her lap, Umā, desiring
to know, said: “who is this?”
Overcome by jealousy, when Śakra was about to strike (the child) with
his thunderbolt, however, his stout arm was paralyzed with the thunder-
bolt in it.
All the gods along with the Prajāpati became confounded regarding
who that mighty one was. They did not recognize that that one was the
Lord of the universe.
Then, having reflected, the effulgent lord Brahmā, having understood
that (the child) of immeasurable prowess is the Supreme One, paid obei-
sance to the Lord of Umā.
Then the gods appeased Umā and Rudra. The arm of the slayer of Vala
then became as it was before. That powerful one (i.e., Rudra) became the
Brāhmaṇa by the name Durvāsas. He lived for a long time in my house in
Dvāravatī.105
And thus Vāsudeva turns to Rudra, or Rudra turns to Umā. The turning of the
God is the cosmological event, it is the secret of its inexhaustible nativity. What
is gained by the “turning” metaphor? Why not simply create? Why have Brahmā
as the creator and reserve for the one Being merely the metaphor of “turning”?
Those who know the secret of the One and the many know that the One can
never “become” many. To such readers, no explanation is necessary. To those
who do not understand these matters, no explanation is possible.
References
Primary Sources
The Mahābhārata for the First Time Critically Edited. Edited by Vishnu S. Sukthankar,
Shripad K. Belvalkar, Parashuram L. Vaidya, et al. 19 vols. Poona: Bhandarkar Oriental
Research Institute, 1933–66.
Secondary Sources
Adluri, Vishwa. “Authenticity and the Problem of the Beginning in the Mahābhārata.”
PhD diss., Philipps-Universität Marburg, 2013.
———. Review of The Bhagavadgītā: Doctrines and Contexts, by Angelika Malinar.
History of Religions 50, no. 1 (2010): 102–7.
———. “Frame Narratives and Forked Beginnings: Or, How to Read the Ādiparvan.”
Journal of Vaishnava Studies 19, no. 2 (2011): 143–210.
———. “Hermeneutics and Narrative Architecture in the Mahābhārata.” In Ways and
Reasons for Thinking about the Mahābhārata as a Whole, edited by Vishwa Adluri,
1–27. Pune: Bhandarkar Oriental Research Institute, 2013.
———. “Literary Violence and Literal Salvation: Śaunaka Interprets the Mahābhārata.”
Exemplar 1, no. 2 (2012): 45–68.
———. “The Perils of Textual Transmission: Decapitation and Recapitulation.” Seminar
608, The Enduring Epic: A Symposium on Some Concerns Raised in the Mahābhārata
(2010): 48–54.
———. “Philosophical Aspects of Bhakti in the Nārāyaṇīya.” In The Churning of the Epics
and Purāṇas at the 15th World Sanskrit Conference, edited by Simon Brodbeck, Alf
Hiltebeitel, Adam Bowles, 127–54. New Delhi: Rashtriya Sanskrit Sansthan and
D.K. Printworld, in press.
———. “Plotinus and the Orient: Aoristos Dyas.” In The Routledge Handbook of
Neoplatonism, edited by Pauliina Remes and Svetla Slaveva-Griffin, 77–99. New York:
Routledge, 2014.
———. “The Vasu Narratives of the Ādiparvan and the Nārāyaṇīya: Some Lexical and
Textual Issues.” In Papers of the Sixth Dubrovnik International Conference on the
Sanskrit Epics and Purāṇas. Zagreb: Croatian Academy of Arts and Sciences,
forthcoming.
Adluri, Vishwa, and Joydeep Bagchee. The Nay Science: A History of German Indology.
New York: Oxford University Press, 2014.
———. Philology and Criticism: A Guide to Mahābhārata Textual Criticism. London:
Anthem, 2016.
Bagchee, Joydeep, and Vishwa Adluri. “The Passion of Paul Hacker: Indology, Orientalism,
and Evangelism.” In Transcultural Encounters between Germany and India: Kindred
Spirits in the Nineteenth and Twentieth Centuries, edited by Joanne Miyang Cho, Eric
Kurlander, and Douglas McGetchin, 215–29. New York: Routledge, 2013.
Bailey, Greg. The Mythology of Brahmā. Delhi: Oxford University Press, 1983.
Biardeau, Madeleine. Études de mythologie hindoue I. Cosmogonies purāṇiques. Paris:
École Française d’Extrême-Orient, 1981.
———. Études de mythologie hindoue II. Bhakti et avatāra. Paris: École Française
d’Extrême-Orient, 1994.
Chakravarti, Uma. Of Meta-Narratives and ‘Master’ Paradigms: Sexuality and the
Reification of Women in Early India. New Delhi: Centre for Women’s Development
Studies, 2009.
Couture, André. “From Viṣṇu’s Deeds to Viṣṇu’s Play: Or Observations on the Word
avatāra as a Designation for the Manifestations of Viṣṇu.” Journal of Indian Philosophy
29, no. 3–4 (2001): 313–26.
Doniger, Wendy. Splitting the Difference: Gender and Myth in Ancient Greece and India.
Chicago: University of Chicago Press, 1999.
Gonda, Jan Aspects of Early Viṣṇuism. Utrecht: Oosthoek 1954.
Hacker, Paul. “Zur Entwicklung der Avatāralehre.” Wiener Zeitschrift für die Kunde
Südasiens 4 (1960): 47–70.
Hillebrandt, Alfred. Vedic Mythology, vol. 2. Translated by S.R. Sarma. Delhi: Motilal
Banarsidass, 1990.
Hiltebeitel, Alf. “Bhīṣma’s Sources.” In Vidyārṇavavandanam: Essays in Honor of Asko
Parpola, edited by Klaus Karttunen and Petteri Koskikallio, 261–78. Studia Orientalia
94. Helsinki: Finnish Oriental Society, 2001.
318 Adluri
———. Dharma: Its Early History in Law, Religion, and Narrative. New York: Oxford
University Press, 2011.
———. “Epic Aśvamedhas.” In Reading the Fifth Veda: Studies on the Mahābhārata,
Essays by Alf Hiltebeitel, vol. 1, edited by Vishwa Adluri and Joydeep Bagchee, 259–78.
Leiden: E.J. Brill, 2011.
———. “Not Without Subtales: Telling Laws and Truths in the Sanskrit Epics.” Journal
of Indian Philosophy 33, no. 4 (2005): 455–511.
———. Rethinking the Mahābhārata: A Reader’s Guide to the Education of the Dharma
King. Chicago: University of Chicago Press, 2001.
———. “Śiva, the Goddess and the Disguises of the Pāṇḍavas and Draupadī.” History of
Religions 20 (1985): 147–74.
———. “The Two Kṛṣṇas on One Chariot: Upaniṣadic Imagery and Epic Mythology.”
History of Religions 24 (1984): 1–26.
———. “Two Kṛṣṇas, Three Kṛṣṇas, Four Kṛṣṇas, More Kṛṣṇas: Dark Interactions in the
Mahābhārata.” In Essays on the Mahābhārata, edited by Arvind Sharma, 101–9. Delhi:
Motilal Banarsidass, 2007.
Jamison, Stephanie. Sacrificed Wife/Sacrificer’s Wife: Women, Ritual and Hospitality in
Ancient India. New York: Oxford University Press, 1996.
Kramrisch, Stella. The Presence of Śiva. Princeton, NJ: Princeton University Press, 1981.
Macdonell, A.A. Vedic Mythology. Delhi: Motilal Banarsidass, 2002. [Strassburg: Trübner,
1897.]
Mertens, Annemarie. Der Dakṣamythus in der episch-purāṇischen Literatur:
Beobachtungen zur religionsgeschichtlichen Entwicklung des Gottes Rudra-Śiva im
Hinduismus. Wiesbaden: Harrassowitz, 1998.
O’Flaherty, Wendy Doniger. Śiva: The Erotic Ascetic. London: Oxford University Press,
1973.
Scheuer, Jacques. Śiva dans le Mahābhārata. Paris: Presses universitaires de France, 1982.
———. “Śiva dans le Mahābhārata: l’histoire d’Ambā/Śikhaṇḍin,” Puruṣārtha 2 (1975):
67–86.
Schreiner, Peter, ed. Nārāyaṇīya-Studien. Wiesbaden: Harrassowitz, 1996.
Sullivan, Bruce M. “The Epic’s Two Grandfathers, Bhīṣma and Vyāsa.” In Essays on the
Mahābhārata, edited by Arvind Sharma, 204–11. Delhi: Motilal Banarsidass, 2007.
———. Kṛṣṇa Dvaipāyana Vyāsa and the Mahābhārata: A New Interpretation. Leiden:
E.J. Brill, 1990.
———. “The Religious Authority of the Mahābhārata: Vyāsa and Brahmā in the Hindu
Scriptural Tradition.” Journal of the American Academy of Religion 62, no. 2 (1994):
377–401.
Sukthankar, V.S. On the Meaning of the Mahābhārata. Bombay: Asiatic Society, 1957.
Sutton, Nicholas. Religious Doctrines in the Mahābhārata. Delhi: Motilal Banarsidass,
2000.
The Divine Androgyne 319
Van Buitenen, J.A.B. “Introduction [to Book 5. The Book of the Effort].” In The Mahāb
hārata: 4. The Book of Virāṭa; 5. The Book of the Effort. Chicago: Chicago University
Press, 1978.
Watkins, Calvert. The American Heritage Dictionary of Indo-European Roots. Boston:
Houghton Mifflin Harcourt, 2000.
320 Bowles
Reflections on the Upākhyānas in the āpaddharmaparvan
Chapter 10
Introduction
This chapter will consider some thematic and functional aspects of the rela-
tionships three upākhyānas (“sub-tales” or “episodes”) of the Āpaddharmapar
van (Ādhp) of the Mahābhārata share with the immediate textual environs of
the upaparvan and parvan of which they form a part, as well as with the broader
Mahābhārata itself. In terms of modern scholarship on the Mahābhārata, this
exercise therefore presents a double challenge, since not merely is it con-
cerned with textual elements (i.e., upākhyānas) oft-times seen as peripheral to
the concerns and overall design of the putative “core” of the Mahābhārata,1 but
it is doing so in the context of a section of the Mahābhārata that has—at least
until a number of notable recent exceptions2—been dismissed as secondary
to the epic ever since Hopkins’ famous designation of the so-called “didactic”
books (of which the Ādhp is one) as “pseudo-epic.”3 In addition, unlike some
upākhyānas found elsewhere in the Mahābhārata (e.g., the Śakuntalopākhyā
nam), those in the Ādhp do not in any ordinary sense narrate back-stories to
the lives of the Mahābhārata’s heroes or their ancestors.
The Āpaddharmaparvan, “The Book on Laws for Times of Distress,” is the
second of the three sub-sections (upaparvans) within the longest book of the
1 See, e.g., Moriz Winternitz, ‘The Mahabharata,’ The Visva-Bharati Quarterly 1 (1924): 344;
J.A.B. van Buitenen, “Introduction,” in The Mahābhārata, vol. 1. 1. The Book of the Beginning
(Chicago: University of Chicago Press, 1973), xiii–xxviii. No more evident than in attempts to
establish a prior shorter version of the Mahābhārata on the basis of Mahābhārata 1.1.61; see
Alf Hiltebeitel, “Not Without Subtales: Telling Laws and Truths in the Sanskrit Epics,” Journal
of Indian Philosophy 33, no. 4 (2005): 457.
2 See James L. Fitzgerald, trans., The Mahābhārata, vol. 7. 11. The Book of Women. 12. The Book of
Peace, Part 1 (Chicago: University of Chicago Press, 2004); Adam Bowles, Dharma, Disorder
and the Political in Ancient India: The Ā paddharmaparvan of the Mahābhārata (Leiden:
E.J. Brill, 2007).
3 Edward W. Hopkins, The Great Epic of India: Its Character and Origin (New York: Charles
Scribner’s Sons, 1901), 381.
7 In the 1961 edition of the Śāntiparvan ce that I own, this Register is found on pp. cxlv–clxiv.
These pages somewhat oddly begin immediately after some other forematter that ends on
p. vii. This anomalous situation is to be explained by the pages of the Register numerically
following the Concordance of the Śāntiparvan in vol. 14 of the ce (which contains the text of
the Ādhp). Hiltebeitel (Alf Hiltebeitel, Dharma: Its Early History in Law, Religion, and Narrative
[New York: Oxford University Press, 2011], 424 n. 29) indicates that the Topical Register in vol.
13 was compiled by V.M. Bedekar, rather than the Śāntiparvan’s editor S.K. Belvalkar.
8 “A laud to preparing for the future”; anāgatavidhānastutiḥ is also found on p. clxiv of the
Topical Register under the ‘stavaḥ, stutiḥ, or stotram’ list.
9 Recently I had the opportunity to attempt an inspection of the manuscripts K3, Da1, Da2, Dn3,
which the Critical Edition indicates were housed at the Bhandarkar Oriental Research
Institute (BORI) in Pune. Of these, K3, from the ‘Kamat Collection of the BORI’ was not avail-
able (it had reportedly been “returned”). Da1 and Da2 could not, unfortunately, be located in
the archive (though the Rājadharmaparvan equivalents could be). Dn3 was located, and its
colophon, it turns out, is more expansive than indicated in the Critical Edition, its text being
śākuṃtalaṃ mīnasaṃbaṃdhi upākhyānaṃ (sic). Leaving aside the still problematic
śākuṃtalaṃ and the defective grammar, the remainder would seem to mean simply “the
upākhyāna associated with fish.”
Reflections on the Upākhyānas in the āpaddharmapar 323
Theoretical Issues
Before discussing each of the narratives that are the focus of this paper, I shall
first outline some of the aspects of the interpretative agenda informing my
approach. This paper takes as its primary task considerations of how to make
sense of the three Ādhp upākhyānas introduced above within the context of
their narration in the Mahābhārata. It therefore follows that some time ought
to be spent reflecting on what might be meant by “context.” One might start
with the observation that the texts analysed here have multiple contexts, and
that, depending on the textual parameters one chooses, one might come to
different conclusions regarding each text’s contextual integration. At its sim-
plest, this could mean a choice between considering each text within the
context of its surrounding adhyāyas, or the upaparvan of which it forms a part
(in this case the Ādhp), or the parvan of which it forms a part (in this case the
Śāntiparvan), or of the entire Mahābhārata. In addition, it could be asked in
such cases whether the contextual significance is one of thematic congruence
or of syntactic function. Of course, none of these contexts are mutually exclu-
sive, and any one text may yield to more than one contextual analysis.
The contrast between contextual readings involving either thematic or syn-
tactic relations might profitably be viewed from the perspective of Ferdinand
de Saussure’s famous distinction between syntagmatic and associative
relations,13 later refined by Roman Jakobson,14 who substituted (and in the pro-
cess more strictly defined) the term “paradigmatic” for Saussure’s “associative.”
An analysis of this type, which became characteristic of the developing field
of semiotics, though its influence has been much broader than that, involves
distinguishing between the horizontal (or combinatorial) axis of the syntagm,
which accounts for the combination of elements that constitute the unit of
analysis, from the vertical (or substitutional) axis of the paradigm, which
accounts for the items selected for each slot available in the syntagm. In the
case of the Ādhp (and the Mahābhārata), the obvious candidates for items con-
stituting the vertical paradigmatic axis would be the various textual units
without saying that texts may yield to more than one thematic analysis and, in
addition, themes in themselves are not discrete entities, but regularly overlap
with and share in the concerns of other themes.
It remains now to briefly comment on two other matters related to the con-
texts of the texts being discussed here. First, at the most obvious level, each of
the texts of a collection like the Ādhp are integrated into their immediate con-
texts by the interlocutory framing system that provides the weft establishing the
surface structure of the corpus considered as a syntagm. There are many points
that could be made in relation to the work done by the interlocutory framing
system to fashion a coherent narration of sometimes-diverse subject matter.
For example, one could point to the ways in which the interlocutory framing
system implicates itself in the interpretation of the texts the framing system
embeds;18 or one could isolate the rhetorical elements that constitute the frame
and analyse their paradigmatic functions in creating coherent syntagms.19
Here, however, I shall especially reflect on the implications of the characters
engaged as interlocutors in the frame, in particular in respect to the praśnin
(and his colleagues), the “neophyte[s]” for whose benefit the texts of the Ādhp
are narrated by Bhīṣma within the unfolding narration of the Mahābhārata.
I take my lead in this instance from Biardeau20 whose contribution on this point
was brought to my attention by Hiltebeitel.21 Biardeau deploys the term “récit-
miroir,” “mirror story,” when discussing three upākhyānas in the Vanaparvan,
the Nalopākhyānam (Mahābhārata 3.50–78), Rāmopākhyānam (3.257–76),
and Sāvitryupākhyānam (3.277–83).22 Accordingly, each of these tales mirrors
the current concerns of the audience “in the poem,” that is to say, the Pāṇḍavas
and Draupadī. Both the Nalopākhyānam and Rāmopākhyānam are recounted
after Yudhiṣṭhira asks—first the sage Bṛhadaśva following his failure at dic-
ing in the sabhā and then Mārkaṇḍeya following Draupadī’s kidnapping by
Jayadratha and her subsequent rescue—if there is a man as unfortunate as he.23
18 I have explored this aspect in relation to “frame theory” in Bowles, Dharma, Disorder and
the Political in Ancient India, 159–63 (see also in relation to individual units of the Ādhp,
ibid., 190ff.).
19 See, e.g., Bowles, Dharma, Disorder and the Political in Ancient India, 172–89.
20 See Madeleine Biardeau, Le Mahābhārata: Un récit fondateur du brahmanisme et son
interprétation, vol. 1 (Paris: Éditions du Seuil, 2002), 412–13.
21 Hiltebeitel, “Not Without Subtales,” 476, 483; “Among Friends: Marriage, Women, and
Some Little Birds,” 112, 115f., 117.
22 For further discussion, see Biardeau, Le Mahābhārata, vol. 1, 497–502; 700–701 and 723–26;
727 and 737–42.
23 Mahābhārata 3.49.34 and 3.257.10. The wording in each instance is almost the same. Hilte-
beitel, “Among Friends: Marriage, Women, and Some Little Birds,” 115 connects the telling
Reflections on the Upākhyānas in the āpaddharmapar 327
of the Nalopākhyānam to Draupadī missing Arjuna while the latter visits Śiva and Indra
(the parallel being Damayantī’s lamenting for the absent Nala). While this could be
inferred from the broader context (perhaps especially in light of Draupadī’s lament at
3.79.11–15 after the Nalopākhyānam), the immediate frame of the Nalopākhyānam focuses
on Yudhiṣṭhira.
24 Mahābhārata 3.277.1–3. The wording of 3.277.3 reflects that of 3.49.34 and 3.257.10.
25 Biardeau, Le Mahābhārata, vol. 1, 412 suggests that as a narrative mirror the Nalopākhyānam
refers to “the entirety of the Mbh and helps to place stress on what truly counts.” In an
earlier study of Nala (Madeleine Biardeau, “Nala et Damayantī, Héros Épiques. Part 2,”
Indo-Iranian Journal 28, no. 1 [1985]: 32), she makes the general observation: “The ‘sub-
narrative’ [sous-récit] is one of the preferred means that the authors have given them-
selves to thereby wink to their audience.”
26 As Hiltebeitel, “Among Friends: Marriage, Women, and Some Little Birds,” 117 has noted,
Mahābhārata 13.57.42–44 “jolts us with the one confirmation” that Draupadī has been
“silently listening all along to Bhīṣma’s battlefield oration … .” See also Bowles, “Framing
Bhīṣma’s Royal Instructions,” 130 n. 22. On Draupadī’s listening in the Mahābhārata, see
also Brian Black, “Eavesdropping on the Epic: Female Listeners in the Mahābhārata,” in
Gender and Narrative in the Mahābhārata, ed. Simon Brodbeck and Brian Black (London
and New York: Routledge, 2007), 1–34.
27 Bowles, Dharma, Disorder and the Political in Ancient India, 23 n. 48.
28 Ibid., 304–5.
328 Bowles
29 For a nice overview of the global spread of the Pañcatantra, see McComas Taylor, The
Fall of the Indigo Jackal: The Discourse of Division and Pūrṇabhadra’s Pañcatantra (Albany:
State University of New York Press, 2007), 1–6. Perhaps ironically, the genre in a sense
“returns” to South Asia with the development of a significant body of Indo-Arabic “Mir-
rors for Princes” literature in the medieval period from the earlier Perso-Arabic tradi-
tion stemming from the “exported” Pañcatantra; of course, with the enduring popularity
of the Pañcatantra, Hitopadeśa, Kathāsaritsagara, and Bṛhatkathā (not to mention the
Mahābhārata), as well as other non-Sanskrit nītiśāstras (V. Narayana Rao and S. Subrah-
manyam “Notes on Political Thought in Medieval and Early Modern South India,” Modern
Asian Studies 43, no. 1 [2009]: 175–210), the genre never really left.
30 For the truth claims inherent in this auto-presentation, see Taylor, The Fall of the Indigo
Jackal, 132–34.
Reflections on the Upākhyānas in the āpaddharmapar 329
fables, these contexts are typically found in the nītiśāstras and the Buddhist
Jātakas. Indeed, two of the upākhyānas discussed here are found in versions of
the Pañcatantra (and the later derivative of the Pañcatantra, the Hitopadeśa),
and one of these two is further found in the Jātakas. These parallel versions
have tended to provoke inquiries of the nature of “source-criticism,”31 which
attempts to uncover the earliest version from which the others, it is then
argued, derive. Yet, this is not the only approach that one might adopt having
discovered parallel iterations of a fable. One could ask, for example, what is
unique about each narration of such a tale. This would attend not merely to
the context of its framing—who recites it? For whom? And why?—but, also, to
what variations it has that might be explained by its narrative context. In short,
this poses the question of how a particular iteration of a narrative has been
optimised for the particular context in which it has been found.
It remains now to investigate the three upākhyānas that are the focus of this
paper in light of the interpretative agenda outlined above. The procedure will
be to first provide a summary of the upākhyāna in question before proceeding
in each case to an analysis of the pertinent elements of each textual unit.
Three fish lived in a lake, the one “Knowing the time has come” (Prāpta
kālajña), one who was “Far-seeing” (Dīrghadarśin) and the “Procrastinator”
(Dīrghasūtra). One day some fisherman began to drain the lake in order
to catch the fish. Dīrghadarśin, seeing the danger, suggested to his friends
that they quickly leave. But Dīrghasūtra thought they should not be so
hasty, while Prāptakālajña said that they should wait for the proper time
before deciding what to do. The wise Dīrghadarśin escaped immediately.
The fisherman drained the lake and captured the other two. Prāptakālajña,
knowing the right time to act, feigned being bound and then escaped
while the other fish were being cleaned. But stupid Dīrghasūtra was
killed.
As is often the case with fables, the moral of this story is reflected in the names
of the story’s protagonists. In this case this moral is self-evident: if a crisis pres-
ents itself, do something about it before it is too late. The general thematic
31 Harry Falk, Quellen des Pañcatantra (Wiesbaden: Otto Harrassowitz, 1978); for cautionary
remarks see Patrick Olivelle, trans. The Pañcatantra: The Book of India’s Folk Wisdom
(Oxford and New York: Oxford University Press, 1997), xxxii–xxxiii.
330 Bowles
terrain of the upākhyāna is clearly a good fit for the Ādhp. In this respect,
three things are worthy of initial note. First, the prima facie appositeness of
the tale for the Ādhp is reinforced by the occurrence of the word āpad in 135.6,
when Dīrghadarśin (“Far-seeing”) appeals to his companions, “This danger
has arisen for all those living in the water” (iyam āpat samutpannā sarveṣāṃ
salilaukasām). This is noteworthy simply for the fact that the term āpad does
not occur in other versions of the story that I have consulted.32 Secondly, the
occurrence of the fable in the Ādhp reflects its nīti orientation, since as a col-
lection it is often especially concerned with what a king ought to do in a time of
crisis. In this respect, the Ā dhp often bears a close relationship to the concerns
of the Arthaśāstra, rather than the dharmaśāstras in which the earliest con-
ceptualizations of āpaddharma are found. And third, the upākhyāna’s frame
and its syntagmatic position would seem to point especially to the concept of
dīrghasūtra (being long-winded, dithering, procrastinating) as the focal point
of this story in the Ādhp.
The last of these points requires further explication in order to appreciate
the peculiar contextual significance of this fable in the Mahābhārata. That the
consequences of being dīrghasūtra is the focal point of the tale in the Ādhp is
emphasized in both the opening stanza (Mahābhārata 135.1) of the chapter in
which it is embedded, which has Bhīṣma announcing that Yudhiṣṭhira “must
listen to this unsurpassed tale having resorted to procrastinating in regard to
what should and shouldn’t be done” (śṛṇvākhyānam anuttamaṃ dīrghasūtraṃ
samāśritya kāryākāryaviniścaye), and the opening stanza of the next chapter
(136.1), in which Yudhiṣṭhira acknowledges the moral of the fable in terms of
“what hasn’t [yet] happened, what has happened, and destructive dithering”
(anāgatā tathotpannā dīrghasūtrā vināśinī). (We will have reason to return to
Yudhiṣṭhira’s wording shortly.) Dīrghasūtra is also prominent in Yudhiṣṭhira’s
praśna located at the very beginning of the Ādhp (129.1), which functions like
an anukramaṇikā for the Ādhp,33 and which lists conditions that entail crises
for kings. Considered syntagmatically, therefore, the Śākulopākhyānam fulfills
an expectation engendered by one element of the opening gambit of the col-
lection in addressing the problem of the king prone to dīrghasūtra.
If a horizontal syntagmatic analysis reveals that the Śākulopākhyānam has
been rhetorically and thematically well-grounded in the context of the Ādhp,
32 However, the word is found in the Hitopadeśa, but only in a stanza that introduces an
additional story embedded in the tale of the three fish; see Hitopadeśa 4.6; Judith Törzsök,
ed. and trans. “Friendly Advice” by Nārāyaṇa & “King Vikrama’s Adventures”, Clay Sanskrit
Library (New York: New York University Press and the JJC Foundation, 2007), 447–48.
33 Bowles, Dharma, Disorder and the Political in Ancient India, 211–12.
Reflections on the Upākhyānas in the āpaddharmapar 331
version, we shall now consider the names of the three protagonists across the
versions, since it is these that carry the rhetorical force of the fable in each
case.
Since it represents a markedly distinct set of names from the other two we can
quickly pass over the Jātaka example, apart from noting that these choices
indicate that the fables were evidently configured for specific cultural and nar-
rative contexts.38 The use of cintin perhaps reflects the Buddhist focus on the
volitional mind; in addition, no one dies in the Mitacintijātakam, with the first
fish managing to save the other two. However, the distinctions between the
Ādhp and Pañcatantra tradition are of greater interest. One may note at the
outset that it is the Ā dhp version that stands distinct from the other iterations.39
In addition, if we consider the language used to convey the underlying morals
for which each fish is a cipher, the Ādhp is very consistent in deploying the
word dīrghasūtra. On the other hand, the vocabulary (including the use of its
names) deployed to convey the moral imparted through Prāptakālajña varies
the most, with (Saṃ‑)pratipattimat or (Saṃ‑)pratipattijña substituting on
every occasion bar the first (Mahābhārata 135.9, 13, 15, 18). A more interesting
case is Dīrghadarśin, since, when explaining the moral conveyed by this fish,
the Ādhp occasionally deploys a language that is clearly cognate with the name
of the fish in the Pañcatantra iterations of the fable. Thus in 135.7, Dīrghadarśin
tries to convince the other two fish to depart by arguing that disaster (anartha)
is averted “before it happens” (anāgata) through good planning. Even more
overtly, in 135.19:
The first word of this stanza is clearly cognate with the name of the first fish
in the Pañcatantra versions.40 Furthermore, the opening stanza of the follow-
ing chapter referred to above, in which Yudhiṣṭhira acknowledges the basic
tenets of the Śākulopākhyānam in terms of “what hasn’t [yet] happened, what
has happened, and destructive dithering” (anāgatā tathotpannā dīrghasūtrā
vināśinī), reflects the fundamental point being made here. The terms anāgatā
and utpannā (in the feminine here because they modify buddhi in 136.1a)
are cognate with the names of the fish in the Pañcatantra, while the term
dīrghasūtrā, of course, maintains a consistency with the central import of the
story in the Mahābhārata.41 While Falk argues that such evidence suggests that
the Mahābhārata version of this fable is prior to (and indeed the model for)
the Pañcatantra version, it seems equally possible that the redactors of the
Mahābhārata responsible for its inclusion (and/or composition) had before
them other versions of the fable to consult; indeed, perhaps there were versions
of the fable prior to its appearance in both the Mahābhārata and Pañcatantra
of which each is derivative in some way.42 We shall probably never really
know. Whatever the case, the pertinent point here is that the Mahābhārata
version consistently reiterates the central moral of the fable through the term
dīrghasūtra. This is even the case in a stanza excised from the constituted text
of the Critical Edition that appears as *307 in the critical notes:43
40 In keeping with his general argument that the Mahābhārata version of the fable is the
model for that in the Pañcatantra, Falk, Quellen des Pañcatantra, 158–59 suggests that this
stanza inspired the shift from the Mahābhārata’s Dīrghadarśin to the Pañcatantra’s
Anāgatavidhātṛ.
41 Falk, Quellen des Pañcatantra does not discuss this stanza.
42 Olivelle, The Pañcatantra, xi, xxxii–xxxiii.
43 This occurs in the following manuscripts: K3–5 V1 B Da Dn1.n3 D2.3.5.8 G1.5 (Dn3 D2.3
repeat it after 135.19).
334 Bowles
This stanza appears in almost precisely the same form in the Pañcatantra ver-
sions, where (like here) it provides an epitome of the fable prior to its full
narration, with the only significant variation being that the Mahābhārata ver-
sion has substituted dīrghasūtrin for yadbhaviṣya, the usual name for the third
fish in the Pañcatantra tradition.44 Falk argues, I think probably correctly, that
this indicates the influence of the Pañcatantra on later Mahābhārata copyists.45
But what is notable here is that the only name cognate with those in the
Mahābhārata version is in fact dīrghasūtrin (“unkorrect,” according to Falk46),
suggesting again the conceptual primacy of this term in the Mahābhārata’s
iteration of the fable and the awareness of its primacy among the Mahābhārata’s
redactors.47 Accordingly, the emphasis placed on dīrghasūtra reflects both the
syntagmatic context of the fable in the Ādhp, and, more broadly, a theme asso-
ciated with discussions of royal behavior threaded through the Mahābhārata.
This takes on added significance in light of the fable being recited to the “dith-
ering” king Yudhiṣṭhira, and therefore being a “mirror” to his own behavior
through the epic.
A savage bird hunter wandered the earth killing and selling birds, oblivi-
ous to his evil conduct. One day, a massive storm hit the forest in which
he was hunting. Unable to find shelter, the hunter became terribly dis-
tressed by the cold, wind, and rain. Seeing a large tree, he sought shelter
44 Edgerton, The Panchatantra Reconstructed, 130 (stanza 127); Hertel, The Panchatantra, 86
(stanza 326); Kale, The Hitopadesa of Narayana, 87 (Hitopadeśa 4.5).
45 Falk, Quellen des Pañcatantra, 162. Note that the wording of pāda d in *307 reflects the
wording of Mahābhārata 12.135.17c, which contains the Ādhp’s summation of the moral
for which Dīrghasūtra is a cipher.
46 Though manuscripts K3.4 D2.3.3 have dīrghasūtra.
47 Falk, Quellen des Pañcatantra, 162 argues that Dīrghasūtrin is the only name that could be
easily replaced on account of syllable number (Silbenzahl) in the case of Dīrghadarśin,
and metrical difficulties (metrischen Schwierigkeiten) in the case of Pratyutpannamati.
However, one wonders if these would have been insurmountable problems for the
Mahābhārata’s redactors.
Reflections on the Upākhyānas in the āpaddharmapar 335
This is a complex fable that lends itself to multiple readings, especially through
its shifts in narrative perspective between the hunter, female dove, and male
dove. Indeed, in his summation of the fable in Mahābhārata 12.145.14–16,
Bhīṣma acknowledges its distinct moral concerns through each of the fable’s
three characters. As we shortly explore further, this fable has often been cited
in the Sanskrit tradition in discussions concerning the merits of a wife
336 Bowles
“following” her husband, in which the virtuous wife (satī) follows her husband
on to the funeral pyre (see, for example, Bhīṣma’s summation in 12.145.1548).
Fitzgerald, on the other hand, focuses especially on the hunter’s redemption
and how this reinforces the point, made elsewhere in the Ādhp, that even “bar-
barians can expiate their wicked way of life.”49 Accordingly, he considers this
fable to shift from the point of view of the king (which is common in earlier
Ādhp sections), though this is not necessarily the point of view of some ele-
ments of the Sanskrit tradition. Here, however, we shall again consider the
fable as a mirror text, first, for its “vocal” interlocutor the king Yudhiṣṭhira (and
therefore questioning Fitzgerald’s ready discarding of it reflecting a “king’s
point of view”), and, second, for its “silent” interlocutor Draupadī, who, as
already indicated, we can assume to be listening together with the Pāṇḍavas
(see above n. 26). In the latter case, emphasis will be placed not merely on the
fable’s implications for the practice of sahagamana (“going with” the husband
on to the funeral pyre), but also on the role of women in relation to dharma
and women’s roles in ensuring that men (or husbands) pursue their proper
dharma (articulated in this instance in relation to hospitality).
This fable’s position in the syntagm of the Ādhp can evidently be explained
by it being broadly about “distress” in a number of ways: the female dove’s cap-
ture by the hunter; the hunter’s being caught in a storm, leading him to seek
shelter; the consequences of the male bird’s onerous obligations towards his
“guest.”50 One could further add, too, Fitzgerald’s point regarding its intersec-
tion with a number of stories in the Ādhp that indicate that barbarians can
productively engage with dharma. Nevertheless, this text is arguably one of a
sequence that has looser ties to the thematic terrain of the Ādhp in compari-
son to earlier texts of the same collection.51 Yet, if the syntagmatic relations of
this text are relatively under-determined, its paradigmatic relations with
48 At Mahābhārata 12.144.8, the grieving female dove laments: “There’s no purpose for me,
lord, without you living. What good woman, bereft of a husband, could bear to remain
alive?” (na kāryam iha me nātha jīvitena tvayā vinā | patihīnāpi kā nārī satī jīvitum utsa-
het ||).
49 Fitzgerald, trans., The Mahābhārata, 161.
50 Of course, most stories deal with crises of one kind or another.
51 It is notable that the word āpad does not occur in this text, or in those that immediately
follow. In Bowles, Dharma, Disorder and the Political in Ancient India, 296–333 I include
this text together with Mahābhārata 12.146–48 (Indrotapārikṣitasaṃvādaḥ), 12.149
(Gṛdhragomāyusaṃvādaḥ) and 12.150–51 (Pavanaśālmalisaṃvādaḥ) in a section titled
“Diversions on a Theme” to reflect their looser thematic integration in the Ādhp. Despite
their looser thematic ties, that these texts are a sequence helps to lend a sense of “shape”
to the collection.
Reflections on the Upākhyānas in the āpaddharmapar 337
Jamison has shown that the rites of hospitality, which typically involved a pre-
scribed set of protocols, were granted an important place in the early Indian
tradition.52 The dharmaśāstras regard these rites to be among the central
duties of the householder (gṛhastha). Manu, for example, in describing
atithipūjana as a “sacrifice to humans” (nṛyajña), considers it (like our male
dove) one of the five “great sacrifices” (mahāyajña) a householder is obliged to
carry out daily.53 He therefore describes these rites in a broader account of the
householder’s duties. The idea of the śaraṇāgata, on the other hand, is typi-
cally encountered in the same literature in sections describing sins and their
penances, with the general points being that either associating with a mur-
derer of someone seeking refuge, or not looking after someone who has come
for refuge, are sins requiring absolution.54
In introducing this text to Yudhiṣṭhira, Bhīṣma offers the following epitome
of the fable in Mahābhārata 141.4:
Though this epitome would also seem to suggest the integration of these two
themes, this is not always how it has been used in other literary texts that cite
this stanza, where it has typically been deployed to emphasise the sanctity of
the śaraṇāgata and the obligations of (especially) kings to provide the shelter
(śaraṇa), even if the śaraṇāgata is an enemy. Typically this occurs in contexts
in which a king’s hawkish allies attempt to convince him to not observe his
obligations to a śaraṇāgata. This is evident, for example, where this epitome is
used in both the Rāmāyaṇa and Pañcatantra. Thus in the Rāmāyaṇa, this
stanza (6.12.11) is one of a number cited by Rāma in support of his decision to
offer shelter to Ravaṇa’s brother Vibhīṣaṇa, against the advice of some of his
attendants (though not Hanuman’s, and that of an initially skeptical Sugrīva).
Similarly, in the Pañcatantra it is cited by Ciraṃjīvin, the minister of the king
of crows Meghavarṇa, in his attempts to persuade the king of owls Arimardana
(“crusher of foes”), Meghavarṇa’s enemy, to provide shelter to his king.55
Yudhiṣṭhira and Bhīṣma, too, frame the text in terms of śaraṇāgata and royal
virtues. Yudhiṣṭhira’s opening question (141.1) asks Bhīṣma about the “dharma
of one maintaining a refuge” (śaraṇaṃ pālayānasya … dharmas …), to which
Bhīṣma responds (141.2), “Great is the dharma in respect to protecting some-
one who’s come for refuge” (mahān dharmo … śaraṇāgatapālane). Bhīṣma
subsequently evokes royal legacy to support this claim:
The deployment of the fable’s epitome in the Rāmāyaṇa and Pañcatantra, and
the particular ways in which Bhīṣma and Yudhiṣṭhira frame this story in the
Ādhp, would therefore suggest that the point of view of the king cannot be
55 The later Pūrṇabhadra Pañcatantra incorporates the entire fable, not just the epitome.
While it shares much in common with the Ādhp version, there are also significant diver-
sions. Hertel (The Panchatantra-Text of Purnabhadra: Critical Introduction and List of Vari-
ants, Harvard Oriental Series vol. 12 [Cambridge, MA, Harvard University Press, 1912], 52)
regards the Mahābhārata to have been Pūrṇabhadra’s source.
Reflections on the Upākhyānas in the āpaddharmapar 339
56 Thus, for example, in Mahābhārata 1.169.18, in the story Aurva told to Arjuna by the gan
dharva Citraratha; Vidura to Dhṛtarāṣṭra in 5.36.64 and 5.37.12; the Trigartas to Duryod-
hana when swearing vengeance on Arjuna (and thereby becoming saṃśaptakas,
“oath-bound warriors”) in 7.16.30; Karṇa to Arjuna in 8.66.62 when the former seeks the
latter’s forbearance on the battle field, Kṛpa to Aśvatthāman in 10.5.10 when the former
seeks to stop the latter slaying the Pāṇḍavas and Pañcālas in their sleep, and the bird
Pūjanī to King Brahmadatta in 12.137.16.
340 Bowles
57 In the preceding chapter the brahman expresses horror at the prospect of a brahman
(since the Pāṇḍavas are disguised as ascetics) and a guest (atithi) losing his life on his
behalf. In imploring Kuntī to not send Bhīma, he pleas (Mahābhārata 1.149.10): the “aban-
doning of someone seeking refuge who’s come in to your house, and the slaying of a sup-
plicant, is reckoned the greatest depravity” (āgatasya gṛhe tyāgas tathaiva śaraṇārthinaḥ
| yācamānasya ca vadho nṛśaṃsaṃ paramaṃ matam).
58 In the case of the Kapotopākhyānam, the hunter (lubdhaka, which means “greedy”) is the
counterpoint of the generous householder/king. Cf. Wilhelm, “Hospitality and the Caste
System.”
59 Mukund Lath, “The Concept of ānṛśaṃsya in the Mahābhārata,” in The Mahābhārata
Revisited, ed. R.N. Dandekar (New Delhi: Sahitya Akademi, 1990), 113–19; cf. Alf Hiltebeitel,
Rethinking the Mahābhārata: A Reader’s Guide to the Education of the Dharma King (Chi-
cago: University of Chicago Press, 2001), 202–9; Bowles, Dharma, Disorder and the Political
in Ancient India, 356–58; Hiltebeitel, Dharma: Its Early History in Law, Religion, and Narra-
tive, 450–52.
Reflections on the Upākhyānas in the āpaddharmapar 341
We shall now explore the role of the female dove (kapotī) in relation to
her husband, his reception of the hunter and aspects of dharma. In turn,
we shall also consider what reflections this leads to in respect to Draupadī.
In regards to the dharma of women, this tale has drawn attention because of
its depiction of a wife devoted to her husband, a pativratā, and its apparent
sanction for the practice of a wife following a husband into death by mount-
ing his funeral pyre (it therefore often being called sahagamana “going with,”
anugamana “going after,” or anumaraṇa “dying after”). In the early eighteenth
century Strīdharmapaddhati, for example, the Tanjore court minister Tryam-
bakayajvan gives an account of the story as part of his argument that a wom-
an’s primary duty is to obediently serve her husband, even to the extent of
“disregarding her own life” (prāṇānām avigaṇanayā).60 Curiously, though
in this context he does mention the female dove’s throwing herself on the
fire that consumed her husband, he does not refer to it in his earlier strong
advocacy for sahagamana.61 Others, however, do cite it when evaluating its
merits.62 For example, in the Mitākṣarā, Vijñāneśvara (late eleventh century)
quotes Mahābhārata 12.144.9cd, 10ab and 12 in support of the practice when
commenting on Yājñavalkyasmṛti 1.86,63 though Vijñāneśvara’s ambivalence
towards sahagamana would seem to be reflected in his advocacy of it only
for women seeking heaven (svarga) rather than mokṣa64; those seeking mokṣa
should choose a widow’s other option, ascetic celibacy (brahmacaryā). The
late fourteenth-century Madanapārijāta by Madanapāla also cites the story
when opposing the position that sahagamana is mandatory for a pativratā.65
In this respect, Madanapāla notes that the female dove in our story is actu-
ally a patrivratā before she enters the fire, and does not become so because
she entered the fire. Consequently sahagamana should be viewed as strictly
60 Julia Leslie, The Perfect Wife: The Orthodox Hindu Woman according to the Strīdharma
paddhati of Tryambakayajvan (New Delhi: Penguin, 1989), 305–7.
61 Ibid., 291–98.
62 As David Brick (“The Dharmaśāstric Debate on Widow-Burning,” Journal of the American
Oriental Society 130, no. 2 [2010]: 203–23) has shown, attitudes towards sahagamana in the
dharmaśāstra tradition varied, and arguments were prosecuted over its legitimacy,
whether it was for brahmans or only other classes, whether it ought to compulsory or
optional, and whether it should be on the same fire of the husband or another.
63 Setlur, ed., The Mitākshara with Visvarūpa, 59; J.R. Gharpure, trans., Yājñavalkya Smṛti with
the Commentaries of The Mitākṣarā by Vijñāneśvara Bhikṣu and the Vīramitrodaya by
Mitramiśra. Ācārādhyāya Chapters I-VII (Bombay: J.R. Gharpure, 1936), 227–28.
64 Cf. Brick, “The Dharmaśāstric Debate on Widow-Burning,” 213.
65 Ibid., 218–19.
342 Bowles
optional.66 Hertel has shown how the alterations of one copyist of a manu-
script identified as part of the Pūrṇabhadra recension of the Pañcatantra
reflect a wholesale rejection of sahagamana, in which case the fire (agni) is
identified metaphorically as tapas.67 It is for the focus on tapas, too, that this
tale attracted the attention of Kullūkabhaṭṭa. In commenting on Manusmṛti
11.240,68 which upholds the merits of tapas by suggesting that it enables all
manner of living creatures and inanimate things to go to heaven, Kullūkabhaṭṭa
refers to this story since it demonstrates that “even birds perform tapas, such as
entering into the fire.”69
If the dharmaśāstras are drawn to the story largely for its apparent advocacy
for sahagamana (or, less commonly, when arguing the merits of tapas), other
aspects of it are deserving of notice, too. Preeminent among these must be the
female dove’s expertise in dharma and, concomitantly, her role in ensuring that
her husband follows his dharma by offering shelter to the hunter and perform-
ing the hospitality rites. We have already seen in the above-cited Mahābhārata
12.142.25 that the male dove keenly expresses an awareness of a householder’s
obligations regarding hospitality. However, the male dove does this only after
his wife has reminded of these obligations, even while she remains captive in
the hunter’s cage. In so doing, the female dove demonstrates a point already
raised by her husband in the course of his bhāryāstuti (“Laud to the Wife”)
expressed in lamenting the absence of his wife: “in this world there is no com-
panion equal to a wife as a means for accomplishing dharma.”70
66 Brick (ibid., 205 n. 6) also indicates that Kamalākara Bhaṭṭa cites this story in his Nirṇaya
sindhu. I have been unable to consult this work.
67 Hertel, The Panchatantra-Text of Purnabhadra, 46, 51. Hertel also appears to suggest (ibid.,
52) that the copyist of the manuscript (referred to as A2) was Jain (Pūrṇabhadra, of course,
was known to be a Jain). However, of two other stanzas interpolated by the copyist (found
in column 2 of Hertel, The Panchatantra-Text of Purnabhadra, 48–49), one asserts that a
woman who offers her own body in a fire goes to a terrible hell, and the other supports
this position by referring to a rule forbidding suicide in the “smṛti, veda and śāstras”
(smṛtavedādiśāstreṣu (sic; one of the “copyist’s blunders” [Hertel, The Panchatantra-Text
of Purnabhadra, 51]). Apart from this indicating that the copyist was attempting to argue
from Vedic orthodoxy, Brick, “The Dharmaśāstric Debate on Widow-Burning” has shown
that similar arguments were also evoked in the dharmaśāstras. In addition, the same
copyist’s focus on the benefits of tapas reflects Kullūkabhaṭṭa’s interest in the story. Reli-
gious identity here is perhaps not a transparent matter.
68 11.241 in Olivelle, Manu’s Code of Law.
69 Shastri, Manusmṛti, 469: itihāsādau kapotopākhyānādiṣu pakṣiṇo ’py agnipraveśādikaṃ
tapas tapantīti śrūyate |.
70 Mahābhārata 12.142.10: nāsti bhāryāsamo loke sahāyo dharmasādhanaḥ ||.
Reflections on the Upākhyānas in the āpaddharmapar 343
The female dove’s exhortations to her husband are worth a closer look. In
succession, she reminds him that he must protect someone who has come for
refuge (śaraṇāgata; Mahābhārata 12.142.14), that he should perform pūjā to the
hunter who is afflicted with cold and hunger (142.15), that a dove’s livelihood
(vṛttiḥ kāpotī) has been prescribed for them according to jātidharma (the law
of species) (142.17), that a householder who follows dharma according to his
ability (yathāśakti) obtains imperishable worlds once he dies (142.18), and that
(142.19):
71 Jamison, Sacrificed Wife/Sacrificer’s Wife; cf. Leslie, The Perfect Wife, 198–205.
72 Jamison, Sacrificed Wife/Sacrificer’s Wife, 153–69, citing this tale among others, has espe-
cially drawn attention to the dangerous obligations of hospitality.
344 Bowles
Though “the past” here is not specified and therefore cannot be determined,
one is tempted to surmise that it is the very recent past, in which his wife has
been instructor and he the instructee. Whatever the case, the dove’s wife has
ably demonstrated a firm knowledge of her husband’s dharma, and success-
fully urged him to follow it. One suspects that without such urging her husband
would have continued to dither, crippled by grief for his absent wife.
Considered paradigmatically within the Mahābhārata, and as a mirror
to Draupadī, who we presume to be one of its auditors, what thematic reso-
nances might be detected between this story and the Mahābhārata’s heroines,
Draupadī in particular? Both the bird and Draupadī are praised as pativratās,
and each extols the virtues of husbands and devoted wifeliness (for Draupadī,
see e.g., her dialogue with Satyabhāmā in Mahābhārata 3.222–23). And, like
the female dove, Draupadī too shows herself to be a capable host to a surprise
guest who then places her in a situation of danger, when, with her husbands
away hunting, she receives Jayadratha as a guest in 3.250–51, showing towards
him the proper observances (3.251.10–11),73 only to be subsequently kidnapped.
However, there are rather more interesting parallels to be drawn with the
two contrasting aspects of the Kapotopākhyānam discussed above. On the one
hand, attention was drawn to the lot of the widow, who, despairing at what lies
ahead (Mahābhārata 12.144.2–8), decides to enter the fire after her husband
(12.144.9); Bhīṣma, in recapping the central points of the story, refers to the
rewards due a wife who “follows” (anu+vṛt) her husband. On the other hand,
focus was placed on the female dove’s taking control of a situation while her
husband dithers,74 and her guiding him towards the proper fulfillment of his
dharma in performing atithipūjana, but also fulfilling the expectations of his
jātidharma. We ought to note the dubious choices the female dove faces; nev-
ertheless, within the constraints of the cultural context drawn by the fable, the
female dove is a strong and decisive actor, who, by force of personality and
argument, manages to convince her husband to follow the dharma allotted
73 Cf. Leslie, The Perfect Wife, 202–3; see also above on page 339.
74 Cf. Jamison, Sacrificed Wife/Sacrificer’s Wife, 15: “In story after story women see what
needs to be done, take command, and order the bewildered, hand-wringing male partici-
pants into their supporting roles—and the enterprise fails only when one of these nin-
nies messes up his part of the woman’s plan.”
Reflections on the Upākhyānas in the āpaddharmapar 345
him. In both instances, the female dove proves herself a pativratā, a wife
upholding her vows to her husband.
It might seem surprising to mention Draupadī in the context of sahagamana,
since she does not mount a funeral pyre at the end of her life, but rather dies in
the course of the “great journey” (mahāprasthāna) undertaken by the Pāṇḍavas
in the Mahāprasthānikaparvan. Nevertheless, this perspective presents itself
on the basis of a highly stimulating and suggestive article by Sally Sutherland.75
Sutherland argues two things of relevance here. First, that the practice of
sahagamana is rather more common in the epics than some scholars have sug-
gested. She notes, for example, that in the Mahābhārata Mādrī and Kṛṣṇa’s
chief wives climb funeral pyres after the deaths of their husbands. Further,
Kuntī, Dhṛtarāṣṭra, and Gāndhārī, once the war has passed and their familial
duties are over, retire to the forest and are consumed in a forest fire in a man-
ner comparable to sahagamana.76 Even Draupadī, in fact, has a close encounter
with sahagamana, when, in the guise of the servant girl Sairandhrī in Virāṭa’s
court, she attracts the eye of Virāṭa’s general Kīcaka.77 After rebuffing his
advances, Bhīma lies in wait disguised as Draupadī (disguised as Sairandhrī!)
and brutally kills Kīcaka. Kīcaka’s family, outraged at his death, approach
Virāṭa and request that the asatī (woman “without virtue”) be burnt with her
lover (kāmin) (Mahābhārata 4.22.5–10).78 Kīcaka’s family takes her to the burn-
ing-ground. But, unbeknownst to them, Draupadī is in fact a virtuous woman
having husbands (nāthavatī satī; 4.22.11) and, calling out to her nātha, she alerts
Bhīma, who saves her in typically brutish fashion.
The second point to draw from Sutherland is more intriguing and, I believe,
an innovative approach to understanding the cultural background of saha
gamana. Sutherland argues that a widow’s self-immolation (sahagamana,
anugamana, or anumaraṇa) is merely one particular way of managing the
(perhaps overwhelming) challenges of widowhood, or a situation of being
bereft of a husband, within a general cultural context in which “following”
a loved one, a friend or a husband is a common expression of grief, love or
be more strident in asserting that her plight is unjust and the result of hus-
bandly neglect. As Draupadī’s abuses compound—by Karṇa and Duḥśāsana in
the sabhā, by Jayadratha in the forest, by Kīcaka in King Virāṭa’s court—her
irascibility and frustration with her husbands, especially Yudhiṣṭhira, multi-
plies. But if Draupadī and the female dove have quite different emotional
responses to their abuse, their verbal rejoinders are driven by a similar con-
cern: to compel their husbands to pursue the dharma appropriate to them.
This may explain the female dove’s calm response, since, as discussed already,
she seemingly suggests that it is the lot of birds according to their jātidharma
to be the food of fowlers. Draupadī, on the other hand, is famous for accusing
her eldest husband of being too much like a brahman, and therefore failing to
acquit himself as husband, warrior, and king. In both cases, the women dem-
onstrate a proficiency in dharma. But more is meant in this respect than their
respective knowledge of submissive wifely duties that a pativratā ought to
demonstrate; indeed, as others have noted,82 Draupadī’s explanation of the
qualities of a pativratā to Satyabhāmā in Mahābhārata 3.222–23 stand in ironic
juxtaposition to her behavior elsewhere, where she does, nevertheless, fre-
quently demonstrate a sound knowledge of dharma. Notoriously, this occurs in
the very scene in the sabhā that sets in train the events of the war, when, after
Yudhiṣṭhira has lost Draupadī in the first dicing match, she queries whether he
had the right to put her up as a stake in the first place, since having already lost
himself he was no longer her lord (2.59.4). The question, posed before a court
of experts in dharma, goes unanswered.83 Draupadī’s victimization in the
sabhā and her husbands’ inaction, and most especially that of the excessively
benign Yudhiṣṭhira, becomes an endless source of rancor, and she is frequently
wont to remind her eldest husband of his duties as warrior and king. In such
instances, Draupadī (like the female dove) shows herself to be (in contradis-
tinction to texts pursuing a stricter śāstric register, as Draupadī herself does in
her dialogue with Satyabhāmā) capable of acting independently, confidently
arguing over courses of action, and, yet, in doing so, demonstrating a key char-
acteristic of the pativratā, that in the epic’s construction of gender, it is women
who are especially responsible for urging men to maintain their dharmas.84
82 E.g., Sally Sutherland “Sītā and Draupadī: Aggressive Behaviour and Female Role-Models
in the Sanskrit Epics,” Journal of the American Oriental Society 109, no. 1: 68.
83 Falk, “Draupadī and the Dharma,” in Beyond Androcentrism: New Essays on Women and
Religion, ed. R.M. Gross (Missoula, MT: Scholars Press for the American Academy of Reli-
gion, 1977) 95; Hiltebeitel, Rethinking the Mahābhārata, 240–77.
84 Falk, “Draupadī and the Dharma,” 98; Angelika Malinar, “Arguments of a Queen:
Draupadī’s Views on Kingship,” in Gender and Narrative in the Mahābhārata, ed. Simon
Brodbeck and Brian Black (London and New York: Routledge, 2007), 85, 90.
348 Bowles
Kṛtaghnopākhyānam
The last of the upākhyānas considered here is also the closing narrative of the
Ādhp, constituting chapters 12.162–67 of the Critical Edition. There is some
internal evidence that suggests this textual unit was a relatively late addition to
the Ādhp.85 However, since in this instance we are concerned with how this
text functions within its varied contexts, we shall be pursuing an analysis
founded upon its syntagmatic and paradigmatic relations, together with con-
siderations of it as a mirror text for Yudhiṣṭhira.
The Kṛtaghnopākhyānam invites a number of syntagmatic consider-
ations. The first and perhaps most startling involves the slaying of the “king
85 See, e.g., Fitzgerald, trans., The Mahābhārata, 778, note to 161.48, and Bowles, Dharma,
Disorder and the Political in Ancient India, 392. This view is critiqued by Hiltebeitel in Alf
Hiltebeitel, “On Reading Fitzgerald’s Vyāsa,” Journal of the American Oriental Society 125,
no. 2 (2005), 246 n. 20.
350 Bowles
of cranes” by the brahman Gautama. The crane is given the name Nāḍījaṅgha
(“Hollow-Shanks”) in Mahābhārata 12.163.18,86 but this is only used this one
time. The subsequent verse (12.163.19) informs us that he became renowned as
Rājadharman (“King Dharma”), and it is this title he bears for the remainder
of the story. Clearly, therefore, it is a designation of some import. Some obser-
vations in this respect shall be made shortly in relation to Yudhiṣṭhira (the
“Dharmarāja”); for the moment we shall note that this tale draws to a close the
“royal” instructions (rājadharma) represented by the Rājadharmaparvan and
Ādhp in combination (the Ādhp continuing the royal orientation of the ear-
lier Rājadharmaparvan—see above pp. 320–21). The slaying of Rājadharman,
therefore, symbolically represents the closing of Bhīṣma’s discourse in its ori-
entation towards advice for royal conduct (rājadharma). In this respect, the
Kṛtaghnopākhyānam is one of a number of “transitional” texts that draw to
a close the Rājadharmaparvan and Ādhp sequence and, in some cases, fore-
shadow the following collection of the Mokṣadharmaparvan.87
In other aspects, this fable revisits some key themes developed elsewhere in
the Ādhp, and to some degree offers a reorientation towards these themes for
its primary auditor, King Yudhiṣṭhira. We should perhaps begin by observing in
general terms that this tale involves: (1) a brahman engaging in behavior not
normally approved; (2) culturally marginal actors (dasyus,88 hunters, śūdras,
mlecchas, śabaras, and rākṣasas); (3) the transubstantiation of the brahman
into a dasyu; (4) the juxtaposition of an “abrahmaṇya” brahman with a
“brahmaṇya” dasyu and rākṣasa; and (5) the juxtaposition of the “abrahmaṇya”
brahman who “destroys” what’s been done (kṛtaghna) with the honorable
“guest worshipping” king of cranes who is emblematic (together with his
“friend” the rākṣasa king Virūpākṣa) of the virtue of kṛtajña (“knowing what
has been done”).
In addition to these points, Bhīṣma frames the fable in terms of ascertaining
who should be allied with (saṃdheya), and who not (asaṃdheya) (see, e.g.,
Mahābhārata 12.162.5). This is typical of the way in which fables are deployed
in the Ādhp, as also in nītiśāstra “mirror of princes” literature, in as much as
the dramatization of relationships between archetypical “beings” illustrates
potential difficulties in relationships between archetypal stakeholders in
political contexts. Thus the vulnerable yet eager-to-please bird is opposed to
his endemic enemy the bird-hunter (a situation not unlike the fable already
discussed), who in this case retains a veneer of his former brahman self, which
serves to lure the bird into over-extending hospitality despite the inevitable
dangers that he is evidently blind to. Kings, in other words, ought to be always
reserving their judgment, if simultaneously acting with propriety. Problems of
this nature become especially crucial when kings encounter crises, for in such
circumstances alliances are of particular significance—it is for this reason
that the Ādhp (like the Arthaśāstra, which in this respect it often resembles)
frequently explores the problems of the “weaker” (king) in the face of the
“stronger” (king). Alliances can often come from surprising sources, since the
ruthless “robber-baron” and the noble “foreigner” may each have their use for
the king in crisis; recognizing “types” and judging “characters,” therefore, are
crucial tools of the royal trade.
A brahman engaging in behavior not normally approved is a typical exem-
plar of situations of āpaddharma in both the Ādhp and the dharmaśāstras.
Yet, while much of the discourse surrounding āpaddharma concerns the brah-
man’s legitimate pursuance of a livelihood not normally permitted him, there
are nevertheless two problems that accompany this commonsense provision:
first, it invites the question of when such provisions are legitimately invoked;
and, second, it poses the problem of how a king, who is responsible for ensur-
ing that varṇasaṃkara (“the intermixing of varṇas”) and dharmasaṃkara
(“the intermixing of dharmas”) are avoided, may recognize when a brahman is
legitimately engaged in an occupation not normally his to pursue. Therefore,
while this fable serves to illustrate the considerations a king ought to make in
contemplating alliances, it also serves to highlight a general problem that the
idea of āpaddharma poses. This is evident in the manner in which the story
presents the brahman Gautama’s justifications for his behavior—he arrives
at the pious dasyu king’s village to beg and seek shelter for the rainy season
(Mahābhārata 12.163.31), he claims to have done so merely for a “livelihood”
(vṛttyartha; 12.162.48), he travels north seeking a “living” (jīvitārthin; 12.163.4),
he explains to Rājadharman that he is a “beggar” (daridra; 12.164.10) and won-
ders what he can do sustain his life (prāṇasaṃdhāraṇa; 12.165.29) on the road
(which leads him to kill Rājadharman for a supply of meat). Such language,
which typically arises in the context of āpaddharma, is encountered elsewhere
in the Ādhp89 and reflects the way in which āpaddharma conceptually devel-
oped in the dharma literature.90 But it is apparent quite early that the brahman
does not intend to return to the conventions demanded by the orthopraxy
expected of him. He stays on in the village for years, living in a house with a
śūdra woman given to him by the dasyu king, learning how to hunt birds, the
way of life of the dasyus (12.162.32–35), until, eventually, he became the same
as them (dasyubhiḥ samatām iyāt; 12.162.36). It takes another brahman, an old
friend evidently more pious than Gautama, to point out Gautama’s slide. “How
did you become a dasyu?” (dasyubhāvaṃ gataḥ katham, 12.162.44) he asks,
while reminding him of his brahman roots in a respectable family and encour-
aging him to leave the village. As we know, the brahman did leave, but too
late for him to deny the impulses that had become second nature—when the
opportunity presents itself, he kills the crane. This story, therefore, expresses
a common anxiety associated with āpaddharma that is part and parcel of the
dogmas expressed through the brahmanic social pathologies of varṇasaṃkara
and dharmasaṃkara: if people do not follow their designated orthopraxies,
but pursue the livelihoods associated with classes other than their own, then
they risk transubstantiation into a member of the class for whom the occu-
pation he follows is figured as the orthopraxy.91 Considered as the closing
narrative of the syntagm of the Ādhp, therefore, the tale serves as a cautionary
reminder of the dangers—societal as well as individual (the brahman, it must
be recalled, is cursed to go to hell)—that lurk behind the relaxing or ignoring
of the orthoproxies underpinning brahmanic social theories. In other words, it
serves to emphasise that āpaddharmas are only legitimate when certain condi-
tions are met.
If these lessons—the seeking of appropriate alliances, the distinguishing of
betrayers from the trustworthy, and the ensuring that orthopraxies are
observed—are rather useful for Yudhiṣṭhira to be apprised of as he prepares to
assume royal office, other factors also evoke paradigmatic associations with
Yudhiṣṭhira and his characterization in the Mahābhārata. The very name of
the bakarāja (“king of cranes”), Rājadharman, suggests the association, since it
91 Such concerns are expressed in respect to dasyus in the Ādhp at Mahābhārata 12.139.1 and
6, elsewhere in the Mahābhārata at 12.68.20, 74.10 and 329.12, and in respect to mlecchas at
3.188.29 and 45. See also Bowles, Dharma, Disorder and the Political in Ancient India, 46 on
Manu 10.92–93, which express a similar concern in respect to a brahman transforming
into a śūdra or vaiśya due to pursuing their occupations (cf. the very graphic Manu 12.70–
72). Further Bowles, Dharma, Disorder and the Political in Ancient India, 217–18 and Wil-
helm Halbfass, India and Europe: An Essay in Philosophical Understanding (New Delhi:
Motilal Banarsidass, 1990), 178 (the latter discusses similar concerns in respect to speaking
mleccha languages).
Reflections on the Upākhyānas in the āpaddharmapar 353
92 “Near,” since there is a slight morphological variation between dharma (in Dharmarāja)
and dharman (in Rājadharman). See further Bowles, Dharma, Disorder and the Political in
Ancient India, 401.
93 Madeleine Biardeau, Études de mythologie hindoue II—bhakti et avatāra (Pondichéry:
Publications de l’École Française d’Extrême-Orient, 1994), 157–61; Alf Hiltebeitel, “Śiva,
the Goddess and the Disguises of the Pāṇḍavas and Draupadī,” History of Religions 20
(1980): 169–70.
94 On the Yakṣapraśna, see also David Shulman, “The Yakṣa’s Questions,” in Untying the Knot:
On Riddles and Other Enigmatic Modes, ed. G. Hasan-Rokem and D. Shulman (New York:
Oxford University Press, 1996), 151–67.
95 James L. Fitzgerald, “Some Storks and Eagles Eat Carrion; Herons and Ospreys Do Not:
Kaṅkas and Kuraras (And Baḍas) in the Mahābhārata,” Journal of the American Oriental
Society 118, no. 2 (1998): 257–61.
96 Biardeau, Études de mythologie hindoue II, 158.
354 Bowles
potentates, but from a social lassitude in which social actors become complicit
in the projection of disorder through their abnegation of orthopraxis. Needless
to say, it is for the king to ensure that such things do not occur.
The bakarāja and the Gautama brahman, therefore, represent a slightly
different thematic thread to the trope of carrion-eating bird as omen. If the
Gautama brahman is a device for Bhīṣma to delineate aspects of the nṛśaṃsa,
the “cruel man,” we would also do well to recall Yudhiṣṭhira’s identification
with the values evoked through a derivative of its antonym, ānṛśaṃsya (see
above p. 340). Indeed, Yudhiṣṭhira has already foregrounded his interest in
these terms in the Ādhp when, in Mahābhārata 12.158.1, he tells Bhīṣma that
he understands ānṛśaṃsya, but not the nṛśaṃsa.97 The subsequent brief text,
an exercise in Listenwissenschaft,98 stands in an apposite relationship to the
Kṛtaghnopākhyānam, which also has its own two-part Listenwissenschaft sec-
tion (12.162.6–2599). In the Yakṣapraśna, Dharma in the guise of the yakṣa is
so thoroughly impressed with Yudhiṣṭhira’s appeal to ānṛśaṃsya (the “highest
dharma”; 3.297.71; cf. 3.297.54–55) as the reason for him choosing to have Nakula
revived so that both Mādrī and Kuntī would then have a surviving son, that he
revives all his brothers, and then, “pleased with his benevolence” (12.298.10;
ānṛśaṃsyena tuṣṭo ’smi), offers Yudhiṣṭhira three further boons. Much later,
as Yudhiṣṭhira refuses to abandon the dog on his final mahāprasthāna into
the Himālaya despite Indra’s protestations, it is precisely to ānṛśaṃsya that
he appeals (17.3.17; cf. 17.3.30).100 The dog, of course, turns out to yet again be
Dharma in disguise. Rājadharman also becomes an exemplar of benevolence
and friendship, though the word ānṛśaṃsya is not used in connection with
him. After Rājadharman as been returned to life, he appeals to Indra to return
the nṛśaṃsa Gautama to life, too. Indra complies, and the kṛtajña Rājadharman
embraces his sakhi, the kṛtaghna nṛśaṃsa Gautama (12.167.11–13). These asso-
ciations between dharma (Dharma), ānṛśaṃsya, Yudhiṣṭhira the Dharmarāja,
and Rājadharman the bakarāja, are perhaps more suggestive than concrete.
Nevertheless, they serve again to remind us of the moral and existential plight
of the tale’s principal auditor, King Yudhiṣṭhira, at the junctures of the dis-
course on royal conduct and the following discourses on soteriologies (in the
References
Primary Sources
Arthaśāstra. The Kauṭilīya Arthaśāstra. Edited and translated by R.P. Kangle. 3 vols. New
Delhi: Motilal Banarsidass, 1997.
Mahābhārata. The Mahābhārata for the First Time Critically Edited. Edited by Vishnu S.
Sukthankar, Shripad K. Belvalkar, Parashuram L. Vaidya, et al. 19 vols. Poona:
Bhandarkar Oriental Research Institute, 1933–66.
Manusmṛti. Manu’s Code of Law: A Critical Edition and Translation of the Mānava-
Dharmaśāstra. Edited and translated by Patrick Olivelle. Oxford and New York:
Oxford University Press, 2006.
Viṣṇusmṛti. The Law Code of Viṣṇu: A Critical Edition and Annotated Translation of the
Vaiṣṇava-Dharmaśāstra. Edited and translated by Patrick Olivelle. Harvard Oriental
Series vol. 73. Cambridge, MA: Harvard University Press, 2009.
Yājñavalkyasmṛti. Yajnavalkyasmrti of Yogisvara Yajnavalkya. Edited by Narayan Ram
Acharya. Bombay: Nirnayasagara Press, 1949.
Secondary Sources
Bailey, Greg. Gaṇeśapurāṇa. Part I: Upāsanākhaṇḍa. Wiesbaden: Harrassowitz Verlag,
1995.
Biardeau, Madeleine. É tudes de mythologie hindoue II: Bhakti et avatāra. Pondichéry:
Publications de l’É cole Française d’Extrême-Orient, 1994.
———. Le Mahābhārata: Un récit fondateur du brahmanisme et son interprétation,
2 vols. Paris: Éditions du Seuil, 2002.
356 Bowles
———. “Nala et Damayantī. Héros épiques. Part 1.” Indo-Iranian Journal 27 (1984):
247–74.
Black, Brian. “Eavesdropping on the Epic: Female Listeners in the Mahābhārata.” In
Gender and Narrative in the Mahābhārata, edited by Simon Brodbeck and Brian
Black, 1–34. London and New York: Routledge, 2007.
Bowles, Adam. Dharma, Disorder and the Political in Ancient India: the Āpaddharmaparvan
of the Mahābhārata. Leiden: E.J. Brill, 2007.
———. “The Dasyu in the Mahābhārata.” In The Churning of the Epics and Purāṇas at
the 15th World Sanskrit Conference, edited by Simon Brodbeck, Alf Hiltebeitel, Adam
Bowles, 155–72. New Delhi: Rashtriya Sanskrit Sansthan and D.K. Printworld, in press.
———. “Framing Bhīṣma’s Royal Instructions: The Mahābhārata and the Problem of
its ‘Design’.” In Parallels and Comparisons: Proceedings of the Fourth International
Conference on the Sanskrit Epics and Purāṇas, September 2005, edited by Petteri
Koskikallio, 121–35. Zagreb: Croatian Academy of Art and Sciences, 2009.
Brick, D. “The Dharmaśāstric Debate on Widow-Burning.” Journal of the American
Oriental Society 130, no. 2 (2010): 203–23.
van Buitenen, J.A.B., trans. The Mahābhārata, vol. 1. 1. The Book of the Beginning. Chicago:
University of Chicago Press, 1973.
Cowell, E.B., ed. The Jātaka or Stories from the Buddha’s former births, translated from
the Pāli by various hands. London: Luzac for the Pali Text Society, 1957.
Edgerton, Franklin, ed. The Panchatantra Reconstructed: An Attempt to Establish the Lost
Original Sanskrit Text of the Most Famous of Indian Story-Collections on the Basis of
the Principal Extant Versions. 2 vols. American Oriental Series vols 2–3. New Haven,
CT: American Oriental Society, 1924.
Falk, Harry. Quellen des Pañcatantra. Wiesbaden: Otto Harrassowitz, 1978.
Falk, N. “Draupadī and the Dharma.” In Beyond Androcentrism: New Essays on Women
and Religion, edited by R.M. Gross, 89–114. Missoula, MT: Scholars Press for the
American Academy of Religion, 1977.
Fausbøll, Viggo, ed. The Jātaka, together with its commentary, being tales of the anterior
births of Gotama Buddha, for the first time critically edited in the original Pāli. London:
Luzac for the Pali Text Society, 1962–64.
Fitzgerald, James L., trans. The Mahābhārata, vol. 7. 11. The Book of Women. 12. The Book
of Peace, Part 1. Chicago: University of Chicago Press, 2004.
———. “Some Storks and Eagles Eat Carrion; Herons and Ospreys Do Not: Kaṅkas and
Kuraras (And Baḍas) in the Mahābhārata.” Journal of the American Oriental Society
118, no. 2 (1998): 257–61.
Gharpure, J.R., trans. Yājñavalkya Smṛti with the Commentaries of The Mitākṣarā by
Vijñāneśvara Bhikṣu and the Vīramitrodaya by Mitramiśra. Ā cārādhyāya Chapters
I-VII. Bombay: J.R. Gharpure, 1936.
Reflections on the Upākhyānas in the āpaddharmapar 357
Chapter 11
Introduction
Let me first state that I am very glad and honoured to contribute to the present
volume devoted to the Mahābhārata and its upākhyānas, designed as a tribute
to Alf Hiltebeitel’s essential analyses of the Mahābhārata in the last forty years.
My aim here is neither to provide a new reading of a specific upākhyāna nor to
propose general considerations about upākhyānas but to present Madeleine
Biardeau’s (1922–2010) understanding and study of the upākhyānas. This chap-
ter can thus be seen as a reflexive paper on how this important epic textual
material was dealt with by one of the most insightful scholars of the Mahāb
hārata and a crucial source of inspiration for Hiltebeitel, who translated into
English what is arguably Biardeau’s most original book, Histoires de poteaux
(Stories about Posts). Considering the massiveness of Biardeau’s work on the
Mahābhārata which spanned some four decades, I have tried to distinguish
phases in her writing on the upākhyānas and I have tentatively identified four
of them, which account for the four-part frame of this contribution:
Before presenting those four phases in their succession, one point must be
stressed: once the essential overall coherence of the Mahābhārata as a unitary
Biardeau devoted the first fifteen years of her work in the field of classical
Indology to an exacting and extensive study in Indian darśanas and philoso-
phy of language, which culminated with the publication in 1964 of her book
Théorie de la connaissance et philosophie de la parole dans le brāhmanisme
classique.2 In the mid-1960s, her interest gradually shifted from Brahmanical
3 Biardeau’s intellectual and academic biography was richly evoked in a one-day seminar—Du
texte au terrain, du terrain au texte: Dialogues disciplinaires autour de l’œuvre de Madeleine
Biardeau—organized in Paris in April 2011 by the Centre d’Etudes de l’Inde et de l’Asie du Sud.
For a remarkable reflection about these “critical years,” we refer the reader to Gérard Colas’s
communication delivered at this occasion “Histoire, oralité, structure: à propos d’un tournant
dans l’œuvre de Madeleine Biardeau,” Journal Asiatique 300, no. 1 (2012): 17–32. In one of her
last publications, a book review, Biardeau herself summarized the main stages of her “journey”
in studying Indian civilization: “Peut-être aussi dois-je faire une part à la formation phi-
losophique que j’ai reçue pendant mes études, avant de me tourner vers l’Inde et de passer
après quelques années de la philosophie indienne aux mythes brahmaniques qui ne se sont
jamais, du moins c’est ainsi que je les vois, complètement séparés de leurs origines védiques.
Peut-être aussi devrais-je avouer que j’ai cru bon d’acquérir quelques connaissances en
ethnologie et que je me suis donné par de longs séjours en Inde la possibilité d’explorer le
«terrain» correspondant de près ou de loin à ce que je trouvais dans les textes mythologiques.”
Madeleine Biardeau, Review of The Mahābhārata and the Yugas, India’s Great Epic Poem and
the Hindu System of World Ages, by L. González-Reimann, Bulletin de l’École Française
d’Extrême-Orient 90–91 (2003–4): 511.
4 The EPHE publishes every year a summary of the conferences given by the professors in the
Annuaire de l’École Pratique des Hautes Études, Section des Sciences religieuses. M. Biardeau
was careful to provide rather detailed summaries of her teaching, complementing her articles
and books. She regularly refers to these accounts of her teaching in her publications. This
material is now freely available online on the French bibliographical website for humanities
and social sciences called PERSEE: <http://www.persee.fr/web/ouvrages/home/prescript/
fond/ephe>.
5 Madeleine Biardeau, “Some More Considerations About Textual Criticism,” Purāṇa 10, no. 2
(1968): 115–23; “Letter to the Editors [January 21st 1970],” Purāṇa 12, no. 1 (1970): 180–81; “The
Story of Arjuna Kārtavīrya without Reconstruction,” Purāṇa 12, no. 2 (1970): 286–303.
362 Dejenne
as early as 1929 and 19346 his doubts about the project of a critical edition of
the Mahābhārata launched by Pune’s Bhandarkar Oriental Research Institute.
The 1968–70 papers deal with what we can call upākhyānas, mostly Jāma
dagnya’s story (the very first epic and Purāṇic subject she chose for study at the
EPHE in 1963), but at that time these sub-stories were studied neither because
they were upākhyānas nor because of their relationship with the epic’s main
narrative. During this phase Biardeau focused on variants of different myths
from the epics as well as from very late Purāṇic sources (for instance, in her
1969 article, “La décapitation de Reṇukā dans le mythe de Paraśurāma,” where
she referred to the Reṇukā-māhātmya, a text most likely composed in the mod-
ern period, to shed light on the epic episode of Reṇukā’s beheading); but she
later abandoned the comparison of texts separated by such a huge temporal
gap, once she began in 1968 her continuous and systematic study of the
Mahābhārata in her EPHE teaching.
6 Lévi is cited by Biardeau for the last time in Madeleine Biardeau, Le Mahābhārata: Un récit
fondateur du brahmanisme et son interprétation, vol. 1 (Paris: Éditions du Seuil, 2002), 19.
7 The temporal distance between the dates of the first publication of these five studies as ar-
ticles and their dates of publication in book-form justify their being mentioned in the bibli-
ography of the present article concerned with the chronology and evolution of Biardeau’s
work. The original BEFEO articles are now freely available online on the same PERSEE elec-
tronic database.
The Status of Upākhyānas in Madeleine Biardeau’s Reflections 363
At the turn of the 1970s and 1980s, Biardeau published two articles in French
(one of them in two parts) that meticulously investigated two of the most
famous upākhyānas whose epic grounding, or maybe even epic origin,10 is
most often downplayed or ignored: the stories of Śakuntalā and of Nala and
Damayantī.11 Biardeau insisted on the epic context of those stories in the title
of both articles, “Śakuntalā dans l’épopée” (“Śakuntalā in the Epic”) and Nala et
Damayantī. Héros épiques” (“Nala and Damayantī as Epic Heroes”). The long
two-part article on the Nalopākhyāna published in the Indo-Iranian Journal12
undoubtedly represents Biardeau’s strongest and most detailed analysis of an
upākhyāna and of its intimate connection with the main narrative of the
Mahābhārata: this article presents the notion of upākhyāna as mirror-story
(“récit-miroir”) of the epic’s central plot. This notion, which is defined for the
first time with such clarity by Biardeau, remained till the end of her work on
the Mahābhārata and proved her main instrument to articulate upākhyānas
and the rest of the epics. The English abstract provided by Biardeau at the
beginning of her article deserves to be quoted in its totality:
When regarded separately, the story of Nala and Damayantī does not
amount to much more than a tale. Therefore it has been inserted in book
III of the Mahābhārata as a “sub-story,” according to the usual practice
of the Epic. The aim of this article is to show that this insertion is not
without importance for an understanding of the story itself. The Epic
provides a context for it and thereby helps in distinguishing the levels
of significance which would otherwise pass unnoticed. Reciprocally, the
“sub-story” brings to the Epic an outlook which is in part new and more
profound by taking up the Epic plot in its own way and at its proper level.
In this way, the role of the Avatāra in the Nalopākhyāna is taken up by the
wife of the exiled king. This leads one to ask oneself about the link con-
necting the Avatāra Kṛṣṇa and the princess Kṛṣṇā Draupadī, wife of the
five Pāṇḍava kings, in the Epic ideology. At the background of all of this,
there are an active concept of the entire Epic, and also the aspects of the
method which results from this.13
13 Biardeau, “Nala et Damayantī: Héros épiques. Part 1,” 247. The French summary, also pro-
vided by Biardeau herself, is more apposite: “Prise en elle-même, l’histoire de Nala et
Damayantī n’est guère plus qu’un conte. Cependant elle est donnée comme un «sous-
récit» à l’intérieur du livre III du Mahābhārata, selon un procédé courant de l’épopée. Le
propos du présent article est de montrer que cette insertion n’est pas sans importance
pour une compréhension de l’histoire. L’épopée lui sert de contexte et aide à en dégager
des niveaux de signification qui passeraient inaperçus sans cela. Réciproquement le
«sous-récit», en reprenant l’intrigue épique à sa manière et dans son registre, renvoie de
l’épopée une image en partie renouvelée et approfondie. C’est ainsi que le Nalopākhyāna
fait presque totalement disparaître le rôle de l’avatāra et donne le rôle principal à l’épouse
du roi exilé. On est amené à s’interroger sur le lien qu’entretiennent dans l’idéologie
épique l’avatāra Kṛṣṇa et la princesse Kṛṣṇā Draupadī, épouse des rois Pāṇḍava. A
l’arrière-plan c’est toute une conception de l’ensemble de l’épopée qui est en jeu et les
aspects de la méthode qui en découlent.” Ibid.
The Status of Upākhyānas in Madeleine Biardeau’s Reflections 365
Barring a few articles—notably the 1991 one on Nara and Nārāyaṇa, the 1997
article on the Kīcaka episode, and the 1997 contribution to Gerhard Ober
hammer’s edited volume Studies in Hinduism14—the last phase of Biardeau’s
publication on the epic can be broadly characterized as a period of populariza-
tion15 of a decade-long work on the Mahābhārata towards a cultured
French-speaking audience.16 She first recapitulated in 1985–86 her previous
findings in a two-volume pocket book, in a collection of classics of world litera-
ture, where selected passages (translated into French by Jean-Michel Peterfalvi),
were accompanied by her copious comments. In 2002, she presented her “final
word” on the Mahābhārata in a huge two-volume book of nearly 2000 pages;
this volume does not bring anything really new for her understanding of the
articulation and of the close and incredibly refined imbrication between the
overall meaning of the Mahābhārata and its “mirror-stories.” The innovation in
the book consists in her strong, but not very convincing, assertion of the
Mahābhārata as an ideological and narrative weapon to retaliate against
Buddhist progress and prevalence in Indian political, religious, and social
mores at the time of the Mahābhārata composition. It must be noted here that
the presentation and the analysis of the upākhyānas are not altered or modi-
fied in any way by this new more historical outlook on the epic and that their
interpretation remains grounded in her previous essentially structuralist read-
ing: in keeping with her earlier study, Biardeau still reads the epic’s main
narrative and the upākhyānas as constituting parts of a single textual corpus.
14 Madeleine Biardeau, “Nara et Nārāyaṇa,” Wiener Zeitschrift für the Kunde Südasiens 35
(1991): 75–108; “Un certain Kīcaka,” in Lex et Litterae: Studies in Honour of Professor Oscar
Botto, ed. Siegfried Lienhard and Irma Piovano (Alessandria: Edizioni dell’Orso, 1997),
35–52; “Some Remarks on the Links between the Epics, the Purāṇas and their Vedic
Sources,” in Studies in Hinduism: Vedism and Hinduism, ed. Gerhard Oberhammer (Vienna:
Verlag der Österreichischen Akademie der Wissenschaften, 1997), 77–119.
15 In order to attract a wide range of readers, Biardeau had a little earlier contributed more
than ten popular articles in an encyclopaedia on world religions and mythologies (Yves
Bonnefoy, ed. Dictionnaire des mythologies et des religions des sociétés traditionnelles et du
monde antique, 2 vols. [Paris: Flammarion, 1981]). We have mentioned in the bibliography
only the entries directly linked with upākhyāna topics.
16 We must also classify under this rubric a new complete annotated French translation of
the Rāmāyaṇa that she supervised, in collaboration with Marie-Claude Porcher, in the
most prestigious French collection of classics, la Bibliothèque de la Pléiade (Madeleine
Biardeau, Le Rāmāyaṇa de Vālmīki [Paris: Gallimard, 1999]).
366 Dejenne
17 This “anti-Buddhist” dimension is not totally new in Biardeau’s work but there had only
been hints of it before the 2002 book.
18 As a difference with the 1985–86 publication, Biardeau is not accompanied in 2002 by a
collaborator in charge of the translation and the volume, so that the text as well as the
interpretation is thus totally her own.
19 Biardeau sometimes preferred to give only the gist of previous analyses published in sci-
entific journals and collections to which she referred the reader (in spite of this book
being intended for a non-specialist readership). We can mention for instance the treat-
ment of Kārtavīrya-Jāmadagnya-Upākhyāna in its Vanaparvan version: “The myth of
Rāma-with-the-axe […] was specifically studied in [Biardeau 1994: p. 73 sq.], particularly
in this Epic version. We thus try here to shorten the narrative as much as possible.”
Biardeau, Le Mahābhārata, vol. 1, 517, n. 20.
The Status of Upākhyānas in Madeleine Biardeau’s Reflections 367
II for the list of words (mainly names of characters) directly connected with
upākhyānas in Biardeau’s rather extensive glossary20 to have an indication of
the sub-stories she thus seemed to find indispensable for a general knowledge
or understanding of the Mahābhārata—Biardeau seems in a few cases to have
used this glossary to provide extremely short summaries of upākhyānas she
did not want to include in the body of her narrative and interpretation but that
deserved to be mentioned at some point as significant cases of stories of kings
and brahmans.21
Biardeau’s concern with upākhyānas as “mirror-stories” is almost exclusive:
it accounts for her detailed analysis of the upākhyānas appearing only in the
first parvans of the Mahābhārata, crucially the Vanaparvan which concen-
trates a number of “mirror-stories” providing the reader—and the Pāṇḍavas
themselves—with clues to understand and decode the meaning of the cos-
mic and dynastic crisis of the main plot. This concern also explains why the
upākhyānas of Books 12 and 13, the two other main storehouses of upākhyānas
in the Mahābhārata, which have a broader ideological and didactic intent,
are not studied or even alluded to in 2002 as well as in 1985–86—in spite of
the far greater size of the 2002 book as was already mentioned. If Biardeau
focuses on upākhyānas that “mirror” the whole epic, there is no complete
equivalence and synonymity between the terms upākhyāna and mirror-story:
many upākhyānas cannot be deemed mirror stories and, reciprocally, some
stories labeled “mirror-stories” by Biardeau involve the main characters of the
Mahābhārata and, as such, are not upākhyānas stricto sensu—this is especially
the case in the Virāṭaparvan with the story of Kīcaka: “We can see that the
Virāṭaparavan reproduces like a mirror part of the epic plot and that the distri-
bution of characters alludes to a well-known [epic] opposition: in the Matsya
kingdom, the Sūtas group led by Kīcaka is opposed to the legitimate king and
to the kingdom he rules.”22
On the whole, in her last and huge statement on the Mahābhārata, Biardeau
has managed to make sense of all the portions of the epic, even the upākhyānas
Conclusion
1. “The narrative” (“Le récit”) offers a summary and preliminary analysis of the
Nalopākhyāna proper and covers the whole of part 1 (pp. 247–74);
2. “The epic context” (“Le contexte épique”; part 2, pp. 1–17) highlights the major
narrative and thematic correspondences, the structural ones so to say, between
the Nalopākhyāna and the main plot of the Mahābhārata;
3. “Mirror-effects” (“Effets de miroirs”; part 2, pp. 17–32) comes back to the analysis
of the Nalopākhyāna to qualify it and to deepen it and try to draw conclusions
for the epic—“Conclusions” that are summarized in part 2 (pp. 32–33).
Elements that had been left aside or remained cryptic in the first two sections
now receive new light.
1. At the outset of her analysis of the Nalopākhyāna, Biardeau clearly stresses the
necessity of the first phase of her undertaking:
Biardeau identifies five segments: “the wedding” (part 1; pp. 248–51; the part of the
story from Damayantī’s miraculous birth to her svayaṃvara and subsequent conjugal
life with reigning Nala), “the game of dice” (pp. 251–54), “the exile” (pp. 254–63), “the
hidden life and the recognition” (pp. 263–71), and “the final game of dice” (pp. 271–72).
At the end of this first section of the article, the complex narrative of the Nalopākhyāna
is very aptly and conveniently summarized by Biardeau in two pages (pp. 272–74) dis-
playing the five segments and their various sub-segments (up to five sub-segments for
the fourth segment, “the hidden life and the recognition”).
25 Ibid., 248.
26 Biardeau, “Nala et Damayantī: Héros épiques. Part 2,” 1.
27 Ibid., 2.
The Status of Upākhyānas in Madeleine Biardeau’s Reflections 371
go beyond them.”28 Much of the analysis in this “epic context” section aims at identify-
ing and delineating the analogies between the main characters of the Nalopākhyāna
and those of the Mahābhārata: Nala reminds the reader of Yudhiṣṭhira, because of his
being the legitimate king and of some aspects of his personality, notably his patience
and his magnanimity, and at the same time of Arjuna, especially for onomastic reasons
(Nala can be read as a slight variation of Nara, the name of the ṛṣi whose Arjuna is the
incarnation). As far as she is concerned Damayantī obviously appears as a reflection of
Draupadī as the image of the Earth-wife of the king and as the suffering and active
queen of kingly husband(s) under duress: the supernatural conditions of their birth,
their śyāma complexion, their svayaṃvaras, their forceful but only partially successful
interventions during their husband’s game of dice, the power of their complaints and
curses, the sameness of the name—sairandhrī—adopted during their phases of hid-
ing in a foreign court … are as many elements strengthening the analogy between both
queens.
Next to those analogies between characters, the Nalopākhyāna as a whole may be
read as “an optimistic version of the crisis of dharma at the attention of Yudhiṣṭhira
[…], which contributes to its aspect of tale”29 and accounts for the “major omissions”
in the summary of the Mahābhārata at the beginning of this section: the “obsessional”
theme of the “disturbed relationship between the two highest varṇas” (in the
Nalopākhyāna, “kṣatriyas and Brahmins are complementary as they must be in a har-
monious society”30); the second major omission, “intimately connected with the first
one, concerns the key-character of the cosmic crisis, the avatāra” as the Nalopākhyāna
does not present on Nala’s side a character who would be an alter ego of Krṣṇa in his
role next to Arjuna (in the Nalopākhyāna a faint but very significant reminder of Kṛṣṇa,
born among the Vṛṣṇis, consists in the very name of Nala’s own charioteer, Vārṣṇeya, to
whom Damayantī entrusts on the eve of the exile in the forest Nala’s chariot, weapons
as well as their children). As a consequence of the much downplayed aspect of the
crisis of dharma in the Nalopākhyāna and of the absence of the avatāra, part of the
latter’s functional role seems to be taken by Damayantī herself who “dominates the
action”31—even if she remains almost totally unaware during the whole Nalopākhyāna
of her being “at the centre of the action” and the main stake in Nala’s rivalry with the
gods, Kali and Dvāpara, his brother Puṣkara or his host-king Ṛtuparṇa. Nala himself is
credited with other aspects of the avatāra as the salvation-giver divinity of the bhakti
ideology (as when he grants abhaya to some characters, including his enemies Kali and
Puṣkara).
28 Ibid., 3.
29 Ibid., 4.
30 Ibid., 5.
31 Ibid., 6.
372 Dejenne
She then summarizes a set of four oppositions “functioning inside a same whole and
being as much reference points as the constant features are [les oppositions … servent
tout autant de points de repère que les traits constants]”33:
The term of an opposition never fully obliterates its opposite so that the mean-
ing is never lost sight of, even if the register war-avatāra-warior king-blood is
more “readable” as a socio-cosmic crisis necessary for the salvation of the world
than the opposite register game of dice-princess-magnanimous king-dust. The
terms of the second register are indeed implied by the terms of the first one: the
game of dice heralds the war, the avatāra appears in the Mahābhārata when the
Pāṇḍavas wed Draupadī, Nala is the Yudhiṣṭhira aspect of the Pāṇḍava kingship
while Arjuna is the central actor in the epic […]. Maybe the conversion of the
first set of terms into the second one was needed to highlight so clearly the role
of the princess.35
3. The third and final section of the article, devoted to the study of “mirror effects”
or “mirroring effects,” is more eclectic than the first two as it investigates some puz-
zling episodes happening during the eventful segment of the exile in the forest of Nala
and Damayantī, first together then separated from each other. Biardeau’s interpreta-
tions may seem more tentative here but the extreme attention devoted to names of
32 Ibid., 16.
33 Ibid., 17.
34 Ibid.
35 Ibid.
The Status of Upākhyānas in Madeleine Biardeau’s Reflections 373
characters or of places helps once more to shed light on an episode as bizarre as the
destruction of the merchants’ caravan which had rescued Damayantī, through a paral-
lel drawn between the roaming of Bhīma on the Gandhamādana mountain and his
killing of the rākṣasa Maṇimat (guardian of this mountain and of Kubera’s treasures),
on the one hand, and the protective role of the merchants supposedly played by the
yakṣa Maṇibhadra, on the other.36 The main result of this section is indeed most likely
the underlining of the rather unexpected analogy between some functions of Bhīma
in the Mahābhārata and the ones of Damayantī in the Nalopākhyāna. This analogy,
expressed in a concise phrase at the end of the second section of the article (“the Bhīma
component of kingship is transferred onto the princess in the Nalopākhyāna”37), is
elaborated mainly thanks to the connection, explored beforehand in the analysis,
between Damayantī and Draupadī, to whom Bhīma is blindly obedient; a more direct
comparison can be made between the grasping of Damayantī by a python just after
being forsaken by Nala and the episode where Bhīma is also imprisoned by a python
which appears to be the cursed wicked king Nahuṣa.38 The violence Damayantī effi-
ciently but blindly displays in her cursing of the lustful hunter and of the unknown
creature responsible for Nala’s sufferings reminds Biardeau of major features of
Bhīma’s personality and destructive action. In the final part of this third section,
Biardeau re-examines once more the respective periods of life incognito of the heroes
in the Mahābhārata and of the Nalopākhyāna because both shed light on one another;
she here draws the attention on the strange relationship between “eater” and “eaten”
that can be observed in both contexts. The Brahman Parṇāda (“eater of leaves”), sent
by Damayantī to find Nala and who succeeds in his mission, is in the same relationship
with king Bhāṅgāsuri Ṛtuparṇa (“who gives leaves at the right period”?) as Yudhiṣṭhira
disguised as the Brahman Kaṅka (seemingly the name of a fish-eating bird in Sanskrit)
is with the Mastya (“fish”) king Vīrāṭa. This homology is not a gratuitous one as “the
kingdom of fishes is mythically ruled by the law of the strongest [la loi du plus fort]
which the war must replace by the dharmic kingship […] but the passage from the car-
nivorous register to the vegetarian metaphor is justified by the fact that there will be
nothing more than a game of dice, there will be neither war nor any kind of execution
even after Nala’s final victory over his brother.”39 In the Nalopākhyāna story, sacrifi-
cial and most blatantly warrior elements are downplayed not to say overlooked by the
center-stage place granted to the game of dice and to the heroine.
In her conclusions, for Biardeau, “the ramification of the main plot in well-built
‘sub-narratives’ [sous-récits], or else in secondary narratives aimed at ‘explaining’ an
36 Ibid., 27–30.
37 Ibid., 17.
38 Ibid., 19–22.
39 Ibid., 31.
374 Dejenne
abnormal situation, satisfies other demands”40 than the mere enjoyment of the reader.
She continues:
The narration intends to be cryptic: one of the charms of the narrative consists
in this display of various levels of meaning […] but this charm requires at some
point for one of these levels to become suddenly dominant, thanks to the insis-
tent repetititon of a word or to a totally unlikely situation or to the highlighting
of a detail, which results in an opening toward a new meaning. The “sub-narra-
tive” [sous-récit] is a most favored device used by the [epic] authors to give a
wink at their audience. Not only does the Nalopākhyāna require the epic context
to be fully understood but we must not omit to return the mirror and to look at
the epic plot under the light of the Nalopākhyāna. We are thus surprised to see
Damayantī take so much importance in the action without any alteration of her
femininity [sans que sa personnalité féminine en soit altérée]. Above all [we are]
surprised to see her almost take in the narrative the place of the avatāra […]. It
obliges the auditor of the epic to connect the princess with the avatāra, Kṛṣṇā
[i.e., Draupadī] with Kṛṣṇa: a whole new understanding makes its way.41
More precisely the Nalopākhyāna helps the reader or auditor of the epic to understand
“the close link of the avatāra not only with the king but with the [princess and] queen”
who, as the king’s wife and the incarnation of Earth, is “most interested in the final suc-
cess of the avataric and kingly action.” As such the queen evokes the Goddess and, in
the power-granting asceticism which forms a common thread in the Nalopākhyāna
and during the twelve-year exile of the Pāṇḍavas, and which is illustrated in the queen’s
estrangement from her king-husband, “we are at the source of the difficulties that we
have to conceive of the Goddess both as a virgin and as Śiva’s wife, both being a spouse
and staying alone in her shrine. […] The Nalopākhyāna, by putting forward a heroine,
helps at the same time to deepen the main feminine character of the epic.”42
40 Ibid., 32.
41 Ibid.
42 Ibid., 33.
43 See Biardeau, Le Mahābhārata, vol. 2, 789–895 (“Glossaire”).
The Status of Upākhyānas in Madeleine Biardeau’s Reflections 375
References
Biardeau, Madeleine. “Avātara.” In Dictionnaire des mythologies et des religions des so-
ciétés traditionnelles et du monde antique, edited by Y. Bonnefoy. 2 vols. Paris:
Flammarion, 1981.
———. “Kāmadhenu: La vache mythique, symbole de prospérité.” In Dictionnaire des
mythologies et des religions des sociétés traditionnelles et du monde antique, edited by
Y. Bonnefoy. 2 vols. Paris: Flammarion, 1981.
———. “Mahābhārata: Le mythe principal.” In Dictionnaire des mythologies et des reli-
gions des sociétés traditionnelles et du monde antique, edited by Y. Bonnefoy. 2 vols.
Paris: Flammarion, 1981.
———. “Nara/Narāyaṇa.” In Dictionnaire des mythologies et des religions des sociétés
traditionnelles et du monde antique, edited by Y. Bonnefoy. 2 vols. Paris: Flammarion,
1981.
———. “Paraśurāma.” In Dictionnaire des mythologies et des religions des sociétés tradi-
tionnelles et du monde antique, edited by Y. Bonnefoy. 2 vols. Paris: Flammarion, 1981.
———. “Rāmāyaṇa.” In Dictionnaire des mythologies et des religions des sociétés tradi-
tionnelles et du monde antique, edited by Y. Bonnefoy. 2 vols. Paris: Flammarion, 1981.
———. “Terre: Les symboles de la terre dans les religions de l’Inde.” In Dictionnaire des
mythologies et des religions des sociétés traditionnelles et du monde antique, edited by
Y. Bonnefoy. 2 vols. Paris: Flammarion, 1981.
376 Dejenne
———. “Nara et Nārāyaṇa.” Wiener Zeitschrift für the Kunde Südasien 35 (1991):
75–108.
———. “Śakuntalā dans l’épopée.” Indologica Taurinensia 7, Dr. Ludwig Sternbach
Felicitation Volume, Part I (1979): 115–25.
———. “Some More Considerations About Textual Criticism.” Purāṇa 10, no. 2 (1968):
115–23.
———. “Some Remarks on the Links between the Epics, the Purāṇas and their Vedic
Sources.” In Studies in Hinduism: Vedism and Hinduism, edited by Gerhard Ober
hammer, 77–119. Vienna: Verlag der Österreichischen Akademie der Wissenschaften,
1997.
———. “The Story of Arjuna Kārtavīrya without Reconstruction.” Purāṇa 12, no. 2 (1970):
286–303.
———. Théorie de la connaissance et philosophie de la parole dans le brāhmanisme clas-
sique. Paris: Mouton et Cie, 1964.
———. “Un certain Kīcaka.” In Lex et Litterae: Studies in Honour of Professor Oscar Botto,
edited by Siegfried Lienhard and Irma Piovano, 35–52. Alessandria: Edizioni dell’Orso,
1997.
Bonnefoy, Yves, ed. Dictionnaire des mythologies et des religions des sociétés tradition-
nelles et du monde antique. 2 vols. Paris: Flammarion, 1981.
Colas, Gerard. “Histoire, oralité, structure: à propos d’un tournant dans l’œuvre de
Madeleine Biardeau.” Journal Asiatique 300, no. 1 (2012): 17–32.
Defourny, Michel. Le Mythe de Yayāti dans la Littérature Épique et Purānique. Bibliothèque
de la Faculté de Philosophie et Lettres de l’Université de Liège, 221. Paris: Les Belles
Lettres, 1978.
Hänggi, Hubert. “Viśvāmitra et Vasiṣṭha dans la littérature épique et purāṇique.” PhD
diss., Paris Sorbonne University, 1971.
Hiltebeitel, Alf. “Not Without Subtales: Telling Laws and Truths in the Sanskrit Epics.”
Journal of Indian Philosophy 33, no. 4 (2005): 455–511.
———. “‘You Have to Read the Whole Thing’: Some Reflections on Madeleine Biardeau’s
Mahābhārata.” In Du texte au terrain, du terrain au texte. Dialogues interdisciplinaires
au tour du l’oeuvre de Madeleine Biardeau. Journée 2011 du Centre d’Études de l’Inde
et de l’Asie du Sud. Paris, 2011. <http://ceias.ehess.fr/docannexe.php?id=1862>.
Scheuer, Jacques. Śiva dans le Mahābhārata. Paris: Presses universitaires de France, 1982.
378 Mahadevan
Chapter 12
1 Max Müller, A History of Sanskrit Literature So Far as it Illustrates the Primitive Religion of the
Brahmans, rev. 2nd ed. (London: Williams and Norgate, 1860), 368, note 6. Mudgala is named
as one of the five adherents of the Śākala Saṃhitā in the commentary to its prātiśākhya text.
2 Alf Hiltebeitel, Rethinking the Mahābhārata: A Reader’s Guide to the Education of the Dharma
King (Chicago: University of Chicago Press, 2001); “Not Without Subtales: Telling Laws and
Truths in the Sanskrit Epics,” Journal of Indian Philosophy 33, no. 4 (2005): 455–511.
3 Hiltebeitel, Rethinking the Mahābhārata, 19.
4 Ibid., 20.
5 During and after the development of the classical scheme of the soma rituals, starting with
the Agniṣṭoma and ending with the Āptoryāma, Vedic ritualism seems to go into a high gear,
with the year-long gavamāyana and the years-long sattra rituals. Indeed the Mahābhārata
presents itself to the world twice with the sattra-type rituals as the setting, first from
Vaiśaṃpāyana in Kurukṣetra and then by Śaunaka in Naimiṣa Forest.
6 See Thennilapuram Mahadevan, “The Institution of Gotra, Ṛgveda, and the Brahmans,” paper
presented at the Fourth Vedic Workshop, Austin, TX, 2007.
7 Thennilapuram Mahadevan, “The Three Rails of the Mahābhārata Textual Tradition,” Journal
of Vaishnava Studies 19, no. 2 (2011): 23–69.
8 Joel Brereton, “The Race of Mudgala and Mudgalānī,” Journal of the American Oriental Society
122, no. 2 (2002): 224–34.
380 Mahadevan
(at ṚV 10.102.8c) “many people”: “[T]he ‘many people’ would be the many
descendants that the impotent husband will win through the niyoga,”9 thus
establishing a patrilineal lineage of the Maudgalyas, with the pravara formula,
later, in BŚS, of Āṅgirasa-Bhārmyśva-Maudgalya, the prefix bhār- (“many”10) in
the immediate progenitor of Mudgala, perhaps alluding to “bahave” in the ṛk:
“he of the many horses.”
We can flesh out this figure further. Mudgala is one of Kuiper’s three hun-
dred-odd non-Vedic items,11 the –gala suffix signifying an unknown Panjab
substrate.12 There are others like him in the ṚV, most notably the Kaṇvas, with
the second largest collection of hymns in the corpus. Witzel suggests a
Dravidian origin for the Kaṇvas,13 but we don’t know of a similar origin for the
term mudgala. Placed at ṚV 10.102, part of the appendix of the corpus, it is a
late composition, probably composed in eastern Panjab toward the Kuru realm
and Kurukṣetra.
The lineage he establishes possesses, as noted above, in the pravara list of
the BŚS (sixth century bce), the genetic formula of Āṅgirasa-Bhārmyaśva-
Maudgalya: the mythical Aṅgirasa, often a metonym for Agni Himself; an
intermediate Bhārmyaśva, one of many horse-linked names in the Vedic
period; and Mudgala the last figure after whom we have a lineage, a group
of Vedic oral agents, many descendants from a male Mudgala marrying only
from other such similar gotra lineages and not in his own. The whole scheme
is pregnant with the import of the Mudgala hymn of the ṚV 10.102, Mudgala-
Mudgalānī engendering a race of Maudgalyas.
The BŚS pravara list designates the line as a “Kevala” Āṅgirasa, (a rubric occur-
ring outside the Aṅgirasas only for the Bhṛgus in the Pravara Lists14). Mudgala
shares this prefix with six other such groupings: Kutsa, Kaṇva, Virūpa,
Pourugutsa, Gaurivita, Āmahīyava. Several of these groupings, as noted with
9 Ibid., 231.
10 Macdonell, s.v. “bhār.”
11 F.B.J. Kuiper, “Ṛgvedic Loan Words,” Studia Indologica 51, no. 3 (1955): 137–85 and
F.B.J. Kuiper, Aryans in the Ṛgveda (Amsterdam and Atlanta, GA: Rodopi, 1991).
12 Michael Witzel, “Substrate Languages in Old Indo-Aryan (Rgvedic, Middle and Late
Vedic),” Electronic Journal of Vedic Studies 5, no. 1 (1999): 13.
13 Ibid., 24.
Witzel, citing Kuiper (Aryans in the Ṛgveda, 7) suggests that the Kaṇvas may be Dravidian
immigrants into the Panjab in the later phases of the ṚV formations. Witzel, “Substrate
Languages in Old Indo-Aryan,” 24.
14 See Mahadevan, “The Three Rails of the Mahābhārata Textual Tradition,” 66–73.
Toward a Geography of the Mahābhārata’s Uñchavṛtti Brahman 381
15 By the Brāhmaṇa period, the “main” ṛṣis, forming the Great Bear in the Milky Way, are
cited as “madhyama ṛṣis,” Middle Ṛṣis, the composers of the middle books, the Family
Books, of the Ṛgveda. The Anukramaṇī index lists seven such figures as the composers of
two different hymns (ṚV 9. 67.1–21 and 10. 137.1–7), in each the list being Bharadvāja,
Kaśyapa, Gotama, Atri, Viśvāmitra, Jamadagni, and Vasiṣṭha and the Bhāradvāja-Āṅgirasas
leading the list in both with the Gotama-Āṅgirasas placed third.
16 Mahadevan, “The Three Rails of the Mahābhārata Textual Tradition.”
17 Macdonnell, s.v. “kevala.”
18 The other nine names are Hiraṇākṣa, Ṛṣabhā, Mitākṣa, Ṛśya, Ṛśyāyana, Dīrghjanghaḥ,
Pralambajanghās, Taruṇā, Bhindavā. See John Brough, The Early Brahmanical System of
Gotra and Pravara: A Translation of the Gotra-Pravara-Mañjarī of Puruṣottama-Paṇḍita.
With an Introduction by John Brough (Cambridge: Cambridge University Press, 1953), 123.
19 See Mahadevan, “The Institution of Gotra, Ṛgveda, and the Brahmans,” Appendix I; Then-
nilapuram Mahadevan, “On the Southern Recension of the Mahābhārata, Brahman
Migrations, and Brāhmī Paleography,” Electronic Journal of Vedic Studies 15, no. 2 (2008):
183–85.
382 Mahadevan
20 Brough, The Early Brahmanical System of Gotra and Pravara, 150; Item 64.
21 Michael Witzel, “The Development of the Vedic Canon and Its Schools: The Social and
Political Milieu,” in Inside the Texts—Beyond texts: New Approaches to the Study of the
Vedas, ed. Michael Witzel, Harvard Oriental Series, Opera Minora, vol. 2 (Cambridge, MA:
Department of Sanskrit and Indian Studies, Harvard University, 1997), 305.
22 Paila is a Jāmadagnya-Bhṛgu (Brough, The Early Brahmanical System of Gotra and Pravara,
79; Item 15) and Jaimini, not listed in BŚS, appears in the Āśvalāyana Śrauta Sūtra Pravara
List as a Yāska-Bhārgava, a Kevala Bhṛgu genus, with the Pravara formula Bhārgava-
Vaitahavya-Sāvetasa (ibid., 94; Item 10).
23 Müller 1860: 386.
Toward a Geography of the Mahābhārata’s Uñchavṛtti Brahman 383
24 See Witzel, “The Development of the Vedic Canon and Its Schools,” 315 and passim.
25 Thennilapuram Mahadevan, “The Ṛṣi Index of the Vedic Anukramaṇī System and the Pra-
vara Lists: Toward a Pre-History of the Brahmans,” Electronic Journal of Vedic Studies 18,
no. 2 (2011): 91–93.
26 John D. Smith, trans. The Mahābhārata: An Abridged Translation (Delhi: Penguin, 2009),
209.
27 J.A.B. Buitenen, trans. The Mahābhārata, vol. 2. 2. The Book of the Assembly Hall; 3. The
Book of the Forest, (Chicago: University of Chicago Press, 1975), 705.
Hiltebeitel’s use of “nirvāṇa” for the final release (Hiltebeitel, Rethinking the Mahābhārata,
54) is interesting; in the original it reads: jagāma śāśvatīṁ siddhiṁ parāṁ nirvāṇalakṣaṇām
(3.247.43 cd).
28 Hiltebeitel, Rethinking the Mahābhārata, 20 n. 78.
384 Mahadevan
Brahman movements, for instance, to the Magadha area and the rise of the
Śukla Yajurveda traditions, centering around the Vājanaseyi Saṃhitā. Likewise,
some of them were part of the Brahman migration to the peninsula comprised
of the Pūrvaśikhā and Aparaśikhā Brahmans.29 The first of these groups, the
Pūrvaśikhās, arrived in the peninsula with the epic ca. 150 bce and are attested
in the Sangam poetry (50 bce–300 ce). The second group, the Aparaśikhā,
arrived half a millennium afterward from sixth through the seventeenth centu-
ries ce and also brought with them the Mahābhārata. The latter group has left
behind enough historical evidence in the form of epigraphy and corroborative
field work for us to reconstruct their migration.
The Mudgala gotra is attested among both the Pūrvaśikhā and Aparaśikhā
Brahmans and thus provides us a compelling picture of the later history of the
Maudgalya group. Their numbers range around 2–3% in the Aparaśikhā data
available from epigraphy and fieldwork, about the same as of the other Kevala
Āṅgirsas.30
An early Mudgala in the Tamil country was Viṣṇucitta or Periyāḻvār, a Cōḻiya
Pūrvaśikhā. He is one of the three Brahman Āḻvārs, besides being the father
of the foundling daughter, Āṇḍāḷ the only woman Āḻvār. Periyāḻvār is also the
composer of the Periyāḻvār Tirumoḻī that comprises the first 473 pāsurams of
the Nālāyiradivyaprabhandam (NVP). The Uttamanaṃbi family of Śrī Raṅgam,
the administrators of the great temple in the middle ages, is descended
directly from him.31 He plays a decisive role in the rise of the Āḻvār phase of
Śrīvaiṣṇavism in the Tamil country, generating a bhakti discourse from the
Mahābhārata and the Harivaṃśa.32
29 Mahadevan, “On the Southern Recension of the Mahābhārata,” 183–85 (Appendices I and
II).
30 For instance, the Pourugutsas and Rahtitaras. But the Kaṇvas are rare in the peninsular
samples; see Thennilapuram Mahadevan, “Toward a Gotra Census from Peninsular Data,”
paper presented at the 223rd Annual Meeting of the American Oriental Society, Portland,
OR, 2013
31 Interview with Singapperumal Uttamanambi, the ninty-fifth descendant of Periyāḻvār
(interview conducted August 6, 2012 at Śrī Raṅgam).
32 Thennilapuram Mahadevan, “The Southern Recension of the Mahābhārata, the Hari
vaṁśa, and Āḻvār Vaiṣṇavism,” in Ways and Reasons for Thinking about the Mahābhā
rata as a Whole, ed. Vishwa Adluri (Pune: Bhandarkar Oriental Research Institute, 2013),
63–117.
Toward a Geography of the Mahābhārata’s Uñchavṛtti Brahman 385
The history of the Mudgalas among the Aparaśikhā Brahmans is even more
dramatic. The first attested Aparaśikhā Mudgala in the peninsula is possibly
Deva Śarman (son of Droṇa Śarman and grandson of Svāmy Śarman). He is
described as an “expert in the teachings of Vedas, Vedāṅgas, Itihāsa, [and]
Purāṇa” in the Vunnuguruvapālem Plates of the Pallava king Parameśvaravarman
(Regnal Year 19 [= 687 ce]),33 the archaeological sites where where the plates
were discovered still in Nellore district in Andhra Pradesh but abutting into the
Toṇḍaimaṇḍalam region of the Tamil country. There are seven Mudgalas (out
of 215 donees) mentioned in the Taṇḍantottam Plates of Nandi Varman exactly
a hundred years later (Regnal Year 58 [= 789 CE]), these archaeological sites
now deep in the Kaveri delta with the Pallava expansion southward and in
present-day Kumbakonam district. As we know, these plates also set aside land
for a “reader” of the Mahābhārata. The Mudgala number reaches seventeen
(out of 1083) in the Karandai Plates of Rajendra Cōḻa (Regnal Year 8 [= 1929–31
CE]); many of them are referred to in these deeds with the last name
Daśapuriyan or originating from Daśapuri, the fabled city of Malva along the
Chambal river. These epigraphic data have been amply confirmed by field data
available about the Aparaśikhā Brahmans: a fairly reliable Gotra census of this
group places the Maudgalya numbers regularly around 3% of the total.34
Is there continued evidence of the twin confluence of ritualism and the epic
among the historical Maudgalyas in Tamil country? I have already noted the
significant role Periyāḻvār plays in generating the bhakti discourse of the Āḻvār
phase of the Śrīvaiṣṇavism of the Tamil country from the epic and the
Harivaṃśa, both of which would have arrived in the Tamil country around the
fourth century ce. His descendants, the extant Uttamanambi family of Śrī
Raṅgam, were masters of the Śrī Raṅgam temple in the middle ages. Perhaps
the historical family where both śrauta Vedism and epic erudition com-
bine most graphically is an Aparaśikhā Brahman family on the Kaveri river in
the village of Śēṅkhālipuram just west of Kumbakonam: that of the late
Anantarāma Dīkṣitar.35 The family has been performing the Agniṣṭoma-
Agnicayana-Vājapeya complex of the soma rituals for the past sixty genera-
tions36 and they have also been the leading exponents of the kathā-kālakṣepa
tradition in the Tamil country, narrating, among other stories, the Mahābhārata
to great acclaim and regional distinction. The tradition is continued by the
thirty-three-year old Kesava Dīkṣitar of Śrī Raṅgam, a third-generation descen-
dant of Anantarama Dīkṣitar. Kesava Dīkṣitar has mastered the Taittirīya
Saṃhitā upto the krama recitation; he performed the Agniṣṭoma at age twenty-
four in 2003 and the Agnicayana at age thirty-three in 2012. He is set on
performing the Vājapeya. His kathā-kālakṣepa repertory includes, since the age
of eighteen, the Śankaravijayam, the Rāmāyaṇa, the Bhāgavatam, the Narāyaṇī
yam, and the Mahābhārata.
It is seen thus that, using gotra markers, the Mudgala of the Mudgala
Upākhyāna can be fully delineated backward from the epic period to his Vedic
origins and milieu and forward through the historical period to the present. It
constitutes a remarkable case history spanning almost 3000 years. It also rein-
forces my call for a radical revision of our understanding of the Brahman
migrations. The migrations were not haphazard and they are always organized
along and within a multi-gotra group—at least four gotra affiliates from extant
epigraphy and fieldwork.37 There is nothing like it in human history.
References
Brereton, Joel. “The Race of Mudgala and Mudgalānī.” Journal of the American Oriental
Society 122, no. 2 (2002): 224–34.
Brough, John. The Early Brahmanical System of Gotra and Pravara: A Translation of the
Gotra-Pravara-Mañjarī of Puruṣottama-Paṇḍita. With an Introduction by John Brough.
Cambridge: Cambridge University Press, 1953.
van Buitenen, J.A.B., trans. The Mahābhārata, vol. 2. 2. The Book of the Assembly Hall; 3.
The Book of the Forest. Chicago: University of Chicago Press, 1975.
Hiltebeitel, Alf. “Not Without Subtales: Telling Laws and Truths in the Sanskrit Epics.”
Journal of Indian Philosophy 33, no. 4 (2005): 455–511.
———. Rethinking the Mahābhārata: A Reader’s Guide to the Education of the Dharma
King. Chicago: University of Chicago Press, 2001.
36 Interview with Śrī Kesava Dikshitar (interview conducted August 6, 2012 at Śrī Raṅgam).
37 Such is the case where historical reconstruction is possible. For instance a Brahman
migration from around the fourteenth century ce brought a group of Jaiminīya Sāmavedis
from the Śrī Raṅgam area in Tamil Nadu to the Koḻunthirapuḷḷi agrahāram in Palghat,
made up of four gotra affiliations: Kautsa-Āṅgirasa, Jāmadagni-Bhṛgus, Vaiśvāmitra-
Aghamarṣaṇa, and Vāsiṣṭha-Kauṇḍinyas.
Toward a Geography of the Mahābhārata’s Uñchavṛtti Brahman 387
Kuiper, F.B.J. Aryans in the Ṛgveda. Amsterdam and Atlanta, GA: Rodopi, 1991.
———. “Ṛgvedic Loan Words.” Studia Indologica 51, no. 3 (1955): 137–85.
Macdonell, Arthur Anthony. A Practical Sanskrit Dictionary with Transliteration,
Accentuation, and Etymological Analysis Throughout. London: Oxford University
Press, 1929.
Mahadevan, Thennilapuram. “The Institution of Gotra, Ṛgveda, and the Brahmans.”
Paper presented at the Fourth Vedic Workshop, Austin, TX, 2007.
———. “On the Southern Recension of the Mahābhārata, Brahman Migrations, and
Brāhmī Paleography.” Electronic Journal of Vedic Studies 15, no. 2 (2008): 1–143.
———. “The Three Rails of the Mahābhārata Textual Tradition.” Journal of Vaishnava
Studies 19, no. 2 (2011): 23–69.
———. “The Ṛṣi Index of the Vedic Anukramaṇī System and the Pravara Lists: Toward
a Pre-History of the Brahmans.” Electronic Journal of Vedic Studies 18, no. 2 (2011):
1–169.
———. “Toward a Gotra Census from Peninsular Data.” Paper presented at the 223rd
Annual Meeting of the American Oriental Society, Portland, OR, 2013.
———. “The Southern Recension of the Mahābhārata, the Harivaṁśa, and Āḻvār
Vaiṣṇavism.” In Ways and Reasons for Thinking about the Mahābhārata as a Whole,
edited by Vishwa Adluri, 63–117. Pune: Bhandarkar Oriental Research Institute, 2013.
Mahalingam, T.V. Inscriptions of the Pallavas. New Delhi: Indian Council of Historical
Research, 1988.
Müller, Max. A History of Sanskrit Literature So Far as it Illustrates the Primitive Religion
of the Brahmans, rev. 2nd ed. London: Williams and Norgate, 1860.
Smith, John D., trans. The Mahābhārata: An Abridged Translation. Delhi: Penguin, 2009.
Witzel, Michael. “The Development of the Vedic Canon and its Schools: The Social and
Political Milieu.” In Inside the Texts—Beyond Texts: New Approaches to the Study of
the Vedas, edited by Michael Witzel, 247–345. Harvard Oriental Series, Opera Minora,
vol. 2. Cambridge, MA: Department of Sanskrit and Indian Studies, Harvard University,
1997.
———. “Substrate Languages in Old Indo-Aryan (Rgvedic, Middle and Late Vedic).”
Electronic Journal of Vedic Studies 5, no. 1 (1999): 1–67.
388 Brodbeck
Chapter 13
5 Alf Hiltebeitel, Rethinking the Mahābhārata: A Reader’s Guide to the Education of the
Dharma King (Chicago: University of Chicago Press, 2001), 132–91.
6 Simon Brodbeck, The Mahābhārata Patriline: Gender, Culture, and the Royal Hereditary
(Farnham: Ashgate Publishing, 2009), 217–66; Simon Brodbeck, “Janamejaya’s Big Brother:
New Light on the Mahābhārata’s Frame Story,” Religions of South Asia 2, no. 2 (2009): 161–
76.
7 Hiltebeitel, Rethinking the Mahābhārata, 92–176; Laurie L. Patton, “Traces of Śaunaka: A
Literary Assessment,” Journal of the American Academy of Religion 79, no. 1 (2011): 113–35.
8 Vishwa Adluri, “Frame Narratives and Forked Beginnings: Or, How to Read the Ādiparvan,”
Journal of Vaishnava Studies 19, no. 2 (2011): 143–210; Vishwa Adluri, “Hermeneutics and
Narrative Architecture in the Mahābhārata,” in Ways and Reasons for Thinking about the
Mahābhārata as a Whole, ed. Vishwa Adluri (Pune: Bhandarkar Oriental Research Insti-
tute, 2013), 1–27.
9 Simon Brodbeck, “Analytic and Synthetic Approaches in Light of the Critical Edition of
the Mahābhārata and Harivaṃśa,” Journal of Vaishnava Studies 19, no. 2 (2011): 223–50.
10 Alf Hiltebeitel, “The Geography of the Mahābhārata’s Upākhyānas,” paper presented
at the Sixth Dubrovnik International Conference on the Sanskrit Epics and Purāṇas
(Dubrovnik, August 2011), 1.
11 I use the word “stories” rather than “tales,” as I think it has fewer incidental connotations;
see, e.g., Adheesh Sathaye, “Magic Cows and Cannibal Kings: The Textual Performance of
390 Brodbeck
the Viśvāmitra Legends in the Mahābhārata,” in Battle, Bards, and Brāhmins, ed. John
Brockington (Delhi: Motilal Banarsidass, 2012), 198 n. 9.
12 See Hiltebeitel’s list of the upākhyānas (Hiltebeitel, “Not Without Subtales,” 467–69). The
criteria for inclusion in this list are detailed immediately after the list, and will be dis-
cussed later.
13 For examples of this kind of explication in my own work, see Simon Brodbeck, “Gendered
Soteriology: Marriage and the Karmayoga,” in Gender and Narrative in the Mahābhārata,
ed. Simon Brodbeck and Brian Black (London: Routledge, 2007), 150–62; Simon Brodbeck,
“The Story of Sāvitrī in the Mahābhārata: A Lineal Interpretation,” Journal of the Royal
Asiatic Society of Great Britain and Ireland (third series) 23, no. 4 (2013): 538–43.
14 Hiltebeitel, “Not Without Subtales,” 472.
Upākhyānas and the Harivaṃśa 391
15 When detailing the audiences of his listed upākhyānas, Hiltebeitel states that “Vaiśam
pāyana addresses ten to Janamejaya” (ibid.); but these are only the ones that Vaiśaṃpāyana
addresses directly to Janamejaya—that is, the ones that are not addressed to a character
within a story told to Janamejaya. In fact Janamejaya hears all of Vyāsa’s upākhyānas, one
from the horse’s mouth (the Hayaśira-Upākhyāna at Mahābhārata 12.335; see Hiltebeitel,
“Not Without Subtales,” 472–73; Alf Hiltebeitel, “The Nārāyaṇīya and the Early Reading
Communities of the Mahābhārata,” in Between the Empires: Society in India 300 bce to 400
ce, ed. Patrick Olivelle [New York: Oxford University Press, 2006], 245–49), and the rest
from Vaiśaṃpāyana.
16 The word paryākhyāna is found in Patañjali’s Mahābhāṣya (ad Pāṇini’s Aṣṭādhyāyī 2.4.54),
but apparently not with this meaning; Monier Monier-Williams, A Sanskrit–English Dic-
tionary (Oxford: Clarendon Press, 1964), 607; The Vyākaraṇa-Mahābhāṣya of Patañjali, vol.
1, ed. Franz Kielhorn, rev. Kashinath V. Abhyankar (Poona: Bhandarkar Oriental Research
Institute, 1985), 487.
392 Brodbeck
frame; although the physical frame is sometimes painted over as a frame (i.e.,
a rectangle is painted approximately around the outside), it is clear that the
frame is thus an integral part of the artwork as the artist conceives it.
Although the Pāṇḍava story at the intermediate level usually passes for the
Mahābhārata’s “main story,” since it is all relative and there are not just three
levels of narration, one might just as well normalize the story of Janamejaya
and the sarpasatra, and see the story of the Pāṇḍavas as a substory, even though
it is never called an upākhyāna in the text (and even though it would still be a
frame story for its own substories). And if one does so, the tools to use for expli-
cating that substory as a substory are those already applied to Yudhiṣṭhira’s 49
upākhyānas on the lower level; that is, the substory can be understood in terms
of the context in which it is told.17 But as soon as one de-normalizes the
17 For example, in past work I have assumed that the Pāṇḍava story told to Janamejaya
relates to Janamejaya’s situation, and I have tried to think about how it does so, and to use
it as supporting evidence for inferences made about that situation (Brodbeck, The
Mahābhārata Patriline, 217–66; Brodbeck, “Janamejaya’s Big Brother”). This follows up the
work of Minkowski, who identified the thematic congruence between the story of the
sarpasatra and the story of the Pāṇḍavas. Christopher Z. Minkowski, “Janamejaya’s Sattra
and Ritual Structure,” Journal of the American Oriental Society 109, no. 3 (1989): 403–4;
Upākhyānas and the Harivaṃśa 393
Figure 13.2 Howard Hodgkin, Dark Evening, 2011; oil on wood, 20 ¾ × 26 × 2 inches.
© Howard Hodgkin. courtesy gagosian gallery.
Pāṇḍava story and steps up to the higher dialogue level, the Harivaṃśa comes
into play.
I have written elsewhere of the Harivaṃśa’s integral status within the
Mahābhārata as the Mahābhārata presents itself.18 I summarize my findings
briefly here. The Mahābhārata’s tables of contents include the khila parvans
among the hundred parvans (listed at Mahābhārata 1.2.34–69), and also
include them in the listed contents of the text even after the Mahābhārata’s
allegedly later division into eighteen parvans, which they stand outside of
(Mahābhārata 1.2.72–234). In both cases I have argued19 that all three parvans
of our Harivaṃśa are mentioned, although the second of those parvans—
called Viṣṇuparvan and Āścaryaparvan in the colophons20—is mentioned out
the Bhaviṣya Parvan (HV 114–18) ends the Pune Critical edition of the
Harivaṃśa, which completed the Pune Critical Edition of the
Mahābhārata.23
There is some hedging and/or fudging here, as the final quotation requires the
Harivaṃśa not to be part of the Mahābhārata in order that there be some
(reduced) temporal remove between the two texts, and the previous quotation
Time Critically Edited, ed. Parashuram L. Vaidya, vol. 1 [Poona: Bhandarkar Oriental
Research Institute, 1969]), or at chapter 30ish as per the colophons; see Horst Brinkhaus,
“The Division into Parvans and the Bhaviṣyaparvan of the Harivaṃśa,” in Stages and Tran-
sitions: Temporal and Historical Frameworks in Epic and Purāṇic Literature, ed. Mary
Brockington (Zagreb: Croatian Academy of Sciences and Arts, 2002), 164–69.
21 This view is to be differentiated from the idea that “the two [parvans of the Harivaṃśa]
that the Mahābhārata refers to as upaparvans may be meant to encompass all three,”
which Hiltebeitel traces to personal communication from myself (Alf Hiltebeitel,
Dharma: Its Early History in Law, Religion, and Narrative [New York: Oxford University
Press, 2011], 577), but is a view I have never knowingly expressed (though it may well have
some mileage).
22 Ibid., 571.
23 Ibid., 578.
24 Ibid., 581.
25 Ibid., 584.
Upākhyānas and the Harivaṃśa 395
On “Upākhyānas”
26 John D. Smith, trans. The Mahābhārata: An Abridged Translation (Delhi: Penguin, 2009).
27 Hiltebeitel, “Not Without Subtales,” 467–69.
396 Brodbeck
from upākhyānas (the former may be interrupted, the latter may not28), none-
theless he admits that his list of upākhyānas “is not a boundaried group without
overlap with other ‘ancillary story’ material.”29
That the list appears as a boundaried group is due to the manner of its
compilation:
This number [i.e., 67] is reached by including all units that are mentioned
to be upākhyānas either in passing in the text,30 cited as upākhyānas in
the PS [the Parvasaṃgraha of Mahābhārata 1.2], or called upākhyānas in
the colophons and/or the running heads for units in the Pune Critical
Edition.31
Nonetheless, since the different manuscripts used by the critical editors vary in
their labelling of specific chapters and units, Hiltebeitel has sometimes used
his judgement in order to decide whether an upākhyāna labelled as such by
only some manuscript colophons deserves a place in the list. And such variety
does not just tell against our ability to compile a trustworthy list; it also tells
against the idea that an upākhyāna is an identifiable thing. If our only way of
determining what is an upākhyāna is to see whether the label is applied, then
how did the labellers determine when to apply the label? On the basis of crite-
ria both unknown and various, it would seem. My sense is that such terms may
often be used impressionistically, and that it is risky to press them into service
as technical genre terms.32
28 Ibid., 471–72.
29 Ibid., 469. The term “ancillary story” is taken from Barbara Gombach.
30 Here Hiltebeitel is not telling the whole truth, since he has not included, in the list of 67
upākhyānas, the Mahābhārata itself as a whole, which is said to be an upākhyāna at
Mahābhārata 1.2.236 (with no variants). See Hiltebeitel, “Not Without Subtales,” 465, 471,
in the first instance noting his surprise at this appellation, and in the second instance
asking “the Mahābhārata would be a subtale to what?”—and thus perhaps drawing atten-
tion to the inadequacy of such a translation of the word upākhyāna.
31 Ibid., 469. Regarding the colophon titles, cf. Fitzgerald’s comment: “My general practice in
translating and writing about the texts collected in the Mokṣadharmaparvan is to use and
translate the name most commonly found in the colophons of the manuscripts used for
S.K. Belvalkar’s Critical Edition of the Śāntiparvan.” James L. Fitzgerald, “The Sāṃkhya-
Yoga ‘Manifesto’ at Mahābhārata 12.289–290,” in Battle, Bards and Brāhmins, ed. John
Brockington (Delhi: Motilal Banarsidass, 2012), 260 n. 3.
32 This remark would apply also to terms such as itihāsa and kāvya. On the face of it, itihāsa
could be any past-tense narration, and kāvya could be any of the output of a kavi.
Upākhyānas and the Harivaṃśa 397
There is also the difficulty of using, on the one hand, mentions of upākhyā
nas as such within the text (this covers the first two of Hiltebeitel’s criteria for
inclusion in the list), and, on the other, mentions of upākhyānas within the
paratextual apparatus (this covers his third criterion). For although the recon-
stituted Mahābhārata is intended to approximate an ancient text as precisely
as possible,33 the colophon details are merely part of its eventual filing and
storage system, and Brockington notes that “these manuscript colophons are
undoubtedly later than anything included in the text by a considerable period.”34
In other words, the colophons might label as an upākhyāna something that
the Mahābhārata’s authors would have thought not to be an upākhyāna. Thus
when Hiltebeitel states that “there does not seem to be anything to discourage
the view that traditional unit titles would have been part of the text’s earliest
self-conception,”35 we can agree and yet still admit that it is guesswork to sug-
gest that a colophon title preserves an aspect of that “earliest self-conception”
in a case where that title is not also used in the text.
Taken together, the foregoing caveats may seem to suggest, at limit, either
that there is no such specific thing as an upākhyāna, or that there is, but we
cannot tell which textual units are upākhyānas and which are not. It is worth
noting that even if such a sceptical position is adopted, Hiltebeitel’s 2005 paper
successfully achieves its brief as set out by the title—that is, to argue that view-
ing the substories of the Mahābhārata and Rāmāyaṇa as later additions is not
only historically indefensible (since there is no evidence of a streamlined ur-
text in either case), but also risks beggaring the texts as works of literature,
since so much of what they do is done through and with the aid of substories.
Beyond that general assessment of the paper, however, I think it remains to be
seen how far one can press the idea that some substories are upākhyānas and
some are not. Nonetheless, the question is an open one, and so, whilst making
no attempt to answer that question on the basis of the meagre evidence pre-
sented here, I will try, for what it is worth, to fill out Hiltebeitel’s data to include
the Harivaṃśa. What might he additionally have said in his paper if, at the time
of writing it, he had been including the Harivaṃśa within the Mahābhārata?
33 “Our objective can only be to reconstruct the oldest form of the text which it is possible to
reach, on the basis of the manuscript material available.” Vishnu S. Sukthankar, “Prolegom-
ena,” in The Mahābhārata for the First Time Critically Edited, vol. 1. The Ādiparvan (Poona:
Bhandarkar Oriental Research Institute, 1933), lxxxvi (italics original).
34 John L. Brockington, “The Spitzer Manuscript and the Mahābhārata,” in From Turfan to
Ajanta: Festschrift for Dieter Schlingloff on the Occasion of his Eightieth Birthday, ed. Eli
Franco and Monika Zin (Lumbini: Lumbini International Research Institute, 2010), 77;
cf. 75, 86.
35 Hiltebeitel, “Not Without Subtales,” 470.
398 Brodbeck
36 On which see Brockington, “The Spitzer Manuscript and the Mahābhārata,” 76.
37 Parashuram L. Vaidya, “Introduction,” in The Harivaṁśa, Being the Khila or Supplement to
the Mahābhārata, for the First Time Critically Edited, vol. 1 (Poona: Bhandarkar Oriental
Research Institute, 1969), xxxi.
38 These are mentioned in the colophons of the critically reconstituted Harivaṃśa. The
colophons within the critical apparatus mention additionally a haṃsaḍibhakopākhyānam
(or haṃsaḍibakopākhyānam, or haṃsaḍimbhakopākhyānam) in appx 31 (The Harivaṁśa,
Being the Khila or Supplement to the Mahābhārata, for the First Time Critically Edited, vol.
2, ed. Parashuram L. Vaidya [Poona: Bhandarkar Oriental Research Institute, 1971], 437–
77), which I do not discuss here. Three upākhyānas mentioned in Vaidya’s “Table of Con-
tents”—the urvaśyupākhyānam (appx 6), the ṣaṭpuravadhopākhyānam (appx 29B), and
the andhakavadhopākhyānam (appx 29C; ibid., 5–6)—are not known as upākhyānas in
the colophons.
Upākhyānas and the Harivaṃśa 399
Hv 7 caturdaśamanvataropākhyānam [sic]39 V3
Hv 15 pūjanīyopākhyānam Ñ2 B1.3 Ds; V3 B2
Hv 15 caṭakopākhyānam V1; V2
Hv 15 śakuntikopākhyānam D1.2
Hv 16 saptavyādhopākhyānam G3
Hv 28 syamantakopākhyānam B1 D1–3.6 T2 G2.3 M2
Hv 29 syamantakopākhyānam K4 D1.2.6
Hv 100 dhanyopākhyānam K Ñ V B Dn Ds D1.2.4–6
Hv 100 dhanyāścaryopākhyānam D3
Hv 100 āścaryopākhyānam T2 G1.4 M4
In what follows, I will summarize and briefly discuss each of these four.
39 Presumably an “n” is missing. The same manuscript V3 uses the label manvantaro
pākhyānam in the additional colophon that fifteen mansucripts have after Harivaṃśa
7.47.
40 Hiltebeitel, “Not Without Subtales,” 469.
400 Brodbeck
Pṛthu-Upākhyāna
By Aṅga, Sunīthā gave birth to a son: one Vena, whose bad conduct caused
severe fury. The seers, seeking offspring, drilled Vena’s right hand, and a
great seer sprang from his drilled hand. On seeing him, the sages said:
“This splendid man will certainly bring joy to the people, and will attain
great fame.” Born with bow and armor, and blazing with splendor, Pṛthu
Vainya, the first kṣatriya, looked after this lady; he was the ruler of the
jewel-bearing earth, and the first to be anointed at a rājasūya; and the
skilful storytellers and praise-singers sprang into being because of him.
Bhārata, your majesty! Seeking livelihoods for his subjects he milked this
cow, and she yielded grains. And the gods and the companies of seers
milked her too, as did the ancestors, the Dānavas, the gandharvas together
with the companies of apsarases, the snakes, the puṇyajanas, the plants,
and the mountains. As she was milked into this bucket and that bucket
the jewel-bearing earth provided whatever milk was desired, and they
kept themselves alive with it.41
The story of Pṛthu has also already been mentioned at Mahābhārata 12.29.129–
36 in the Ṣoḍaśarājika-Upākhyāna, and narrated at Mahābhārata 12.59.93–130
in response to Yudhiṣṭhira’s question about the origin of kingship. The sum-
mary at Harivaṃśa appx 44 calls the Pṛthu story an ākhyāna (appx 44.3), and
some of the manuscript colophons call it a carita. I summarize the upākhyāna
here chapter by chapter, giving the colophon details each time:42
Harivaṃśa 4 (pṛthūpākhyānam, K1.2.4 Ñ3.3 [sic] V B1.2 Dn Ds D1–5; somā
bhiṣekaḥ, T2; caturthasargaḥ, T3; somādyabhiṣekaḥ, G1 M4; somādyabhiṣeke
dikpālasthāpanam, G2.3.5). Vaiśaṃpāyana narrates. Brahmā appointed sover-
eigns for every type of being, and also appointed the guardians of the directions
(1–17). Vaiśaṃpāyana offers to tell about Manu Vaivasvata (18), but Janamejaya
asks instead to hear about the birth (janman) of Pṛthu, and details of the
milking of the earth, and the reason why the ṛṣis drilled Vena’s hand (19–22).
Vaiśaṃpāyana praises and prepares to narrate “The Birth (saṃbhava) of Pṛthu
Vainya” (23–26).
Harivaṃśa 5 (pṛthivīvākyam, V2; pṛthūpākhyānam, V3 B Ds D2.5; pṛthuca
ritam, D3 T2 G1; vainyotpattau vasudhāvākyam, D6 G2.3.5 M4; vainyotpatti, T4).
Vena’s family and rule are described; he outlaws offering to the gods, elevating
himself in their place (1–7). The seers try to reason with him, but are rejected
(8–13). They grab him; from his left thigh they drill the ancestor of the niṣādas
(14–19), and from his right hand they drill Pṛthu, who immediately takes up
weapons (20–22). Because he now has a good son, Vena is saved from hell (23–
24). Pṛthu is anointed king, to the joy of all, and is praised (25–31). The storyteller
and the praise-singer are created and told to praise Pṛthu; they say they don’t
know about him; they are told to praise him according to his future deeds; they
do so and are rewarded (32–39). The other subjects ask Pṛthu for livelihoods;
he turns upon the earth, and she flees in the form of a cow, but she cannot
escape, so she pleads with him, explaining how necessary she is (40–52). He
calms down (53).
Harivaṃśa 6 (vainyopākhyānam, Ś1; pṛthivīdohaḥ, K1–3 Dn; pṛthūpākhyānam,
K4 Ñ2 V2.3 B D1.2.4 T3 M3; pṛthor upākhyānam, Ñ3 V1; pṛthucaritam, D3; pārijāte
pṛthūpākhyānam, D5; vasuṃdharādohaḥ, D6 T2 G1; pṛthumāhātmye vasuṃdha
rādohaḥ, G2.3.5; pṛthuvaṃśacaritam, M4). Pṛthu says he will spare the earth if
she will be his daughter and nourish all the subjects (1–6). She tells him to
give her a calf so she can lactate, and to make her level so the milk will go
42 For translations, see Manmatha N. Dutt, trans. A Prose English Translation of Harivamsha
(Calcutta: Elysium Press, 1897), 20–30; Dhirendra N. Bose, ed. Harivamsha Translated into
English Prose from the Original Sanskrit Text (Dum Dum, Bengal: Datta Bose and Co., n.d.),
14–21 (with some unexpected chapter titles).
402 Brodbeck
everywhere (7–8). Pṛthu levels and de-boulders the earth with the tip of his
bow (9–11); then, for the good of his subjects, he makes Manu the calf and
milks grains from the earth (12–15). The earth is then milked successively by
seers, gods, ancestors, snakes, asuras, yakṣas, rākṣasas and piśācas, gandharvas
and apsarases, mountains, and trees and plants, using a different bucket and
calf each time; to each group she yields a different power (16–37). Praise of
the earth (38–41). Praise of King Pṛthu Vainya; the rewards of his veneration
(42–48). Vaiśaṃpāyana asks what else Janamejaya wants to hear (49).
The upākhyāna is thus very much as summarized in the preparatory trailer
at Harivaṃśa 2.19–26. It is also well summarized by its colophons, although the
inclusion of Harivaṃśa 4 as part of the upākhyāna may seem slightly awry, as
the story proper is restricted to Harivaṃśa 5 and 6.
Since the time of Huntington’s ambitious essay in terms of the individua-
tion process,43 the story of Pṛthu has been studied by Doniger, Bailey, Miller,
and Fitzgerald,44 and it was also discussed by Tamar Reich in a paper presented
recently.45 Here I would like to comment on the circumstances of its introduc-
tion at this textual juncture, and on its connection with an earlier upākhyāna.
After mentioning the establishment of the direction-guardians, Vaiśaṃ
pāyana says:
And these protectors of the people anointed Pṛthu at a rājasūya, for sov-
ereignty over the kings—as did the kings themselves, according to the
ordinance found in the Veda. Then, when the boundlessly brilliant period
of Manu Cākṣuṣa had passed, he [i.e., the Grandfather] appointed Manu
Vaivasvata to the sovereignty of the earth. If you would like to hear the
detailed account of Manu Vaivasvata, I will narrate it as a favor for you,
supreme and impeccable king. For it is the basic foundation that is estab-
lished in the old lore.46
But Janamejaya says he wants to hear details about Pṛthu instead. This posi-
tioning is interesting because both of those characters, Pṛthu and Manu, are
said to be the first king. Manu has been described in this way at Mahābhārata
12.67.20–31 and 12.160.66–71.47 The mechanism for allowing both Pṛthu and
Manu to play this role without contradiction is, at this point, the notion of the
various Manu periods. Manu Vaivasvata has his own period, and Pṛthu was
the first king in the period of Cākṣuṣa, the previous Manu.48 The full details
of the various Manu periods are given in Harivaṃśa 7, straight after the Pṛthu
narrative has finished. Janamejaya requests them, as well he might—perhaps
he is bemused by the overlap between Pṛthu and Manu. But if Manu’s thun-
der has been stolen slightly here, he makes up for it not just through his role
in the solar lineage shortly to be described in chapters 8–10, but also in his
role as śrāddhadeva, the god of ancestral rites. Curiously, and perhaps appro-
priately, this latter role is also seemingly shared with someone else, this time
Manu’s father: Manu is identified as śrāddhadeva at Mahābhārata 12.122.39
and at Harivaṃśa 8.7 and 13.65, and Vivasvat is identified as śrāddhadeva at
Mahābhārata 12.329.44 and at Harivaṃśa 10.80 and 11.1.49 This dual identity is
highlighted soon after the dual identity of the first king, but there is no attempt
to explain it with reference to different Manu periods.
It is notable that only one manuscript uses the label vainyopākhyānam in
connection with the Pṛthu story (Ś1, at Harivaṃśa 16), because the Vainya-
Upākhyāna has already been presented: it is mentioned in the Parvasaṃgraha
(at Mahābhārata 1.2.126) and occurs at Mahābhārata 3.183, where it is labelled
as such by the colophon in more than a dozen manuscripts.50 Here is Smith’s
summary:
[183] Next Mārkaṇḍeya tells the story of Atri. Atri planned to retire to the
forest, and went to see King Vainya in hopes of receiving wealth to pass
on to his sons. But when he offered praise to Vainya he was contradicted
by another ascetic, Gautama, and a dispute arose as to whether kings
hold supreme power. Sanatkumāra, asked to resolve it, agreed with Atri
and declared that kings have been granted supremacy by Brahmins in
order to maintain dharma. Pleased with this outcome, Vainya rewarded
Atri generously.51
49 Brinkhaus has argued that Harivaṃśa 10.80 names Manu as śrāddhadeva, not Vivasvat,
and that Harivaṃśa 11.1 is an interpolation (Horst Brinkhaus, “Manu Vaivasvata as
Śrāddhadeva: On the Insertion of the Pitṛkalpa into the Harivaṃśa,” in Epic Undertakings,
ed. Robert P. Goldman and Muneo Tokunaga [Delhi: Motilal Banarsidass, 2009], 1–11); but
Söhnen-Thieme has defended the more obvious sense of 10.80 (Renate Söhnen-Thieme,
“The Setting of Purāṇas: Frame Stories, Layers of Interlocution, and Ultimate Authority,”
in Epics, Khilas, and Purāṇas: Continuities and Ruptures, ed. Petteri Koskikallio [Zagreb:
Croatian Academy of Sciences and Arts, 2005], 448 n. 30).
50 These include “Ś1”; but the manuscript sigla only sometimes carry over from one volume
of the critical edition to another, and judging by the editorial notes this may or may not
be the Ś1 used for the Harivaṃśa edition. See Vishnu S. Sukthankar, “Introduction,” in The
Mahābhārata for the First Time Critically Edited, vol. 3. The Āraṇyakaparvan (Poona: Bhan-
darkar Oriental Research Institute, 1942), ii–iii; Sukthankar, “Prolegomena,” x–xi; Vaidya,
“Introduction,” xvi–xvii. On the wider issue, see John Dunham, “Manuscripts Used in the
Critical Edition of the Mahābhārata: A Survey and Discussion,” in Essays on the
Mahābhārata, ed. Arvind Sharma (Leiden: E.J. Brill, 1991), 2: “Some libraries have cata-
loged as separate codices manuscripts which are by the same copyist but are of different
parvans of the Mahābhārata.”
51 Smith, The Mahābhārata, 199 (block italics removed).
Upākhyānas and the Harivaṃśa 405
is presumably our Pṛthu Vainya or one of his descendants, and the two
upākhyānas go hand in hand. What Sanatkumāra states in brief in the Vainya-
Upākhyāna is graphically illustrated by the narrative of the Pṛthu-Upākhyāna,
and seems intended to obviate brahman scepticism about the royal preroga-
tive: “The ṛṣis, concerned about lawlessness, entrusted power to the kṣatriyas”
(adharmād ṛṣayo bhītā balaṃ kṣatre samādadhan || Mahābhārata 3.183.25cd).
This emphasis on the brahmans is evident also in a phala passage that
Vaiśaṃpāyana presents before the main body of the Pṛthu-Upākhyāna:
Hear the mystery declared by the seers, your majesty, as it really is—
equivalent to the Veda, and conducive to heaven, fame, long life, and
riches. This is “The Birth of Pṛthu Vainya”; a person who always praises it
gains the respect of the brahmans, and doesn’t regret anything that they
have or haven’t done (kṛtākṛtam).52
Sapta-Upākhyāna
This upākhyāna, which forms the latter part of the Pitṛkalpa (Harivaṃśa 11–19)
in between the accounts of the solar and lunar dynasties (Harivaṃśa 8–10,
20–29),55 is nested within several different dialogues: Vaiśaṃpāyana, narrating
to Janamejaya, is relaying dialogue that took place between Yudhiṣṭhira and
Bhīṣma while the latter was lying on his bed of arrows after the Kurukṣetra war
(Harivaṃśa 11.6). Bhīṣma in turn is relaying what he once heard from Mār
kaṇḍeya, whose narration is based on his encounter with the divine youth
Sanatkumāra.56 In the run-up to the Sapta-Upākhyāna Mārkaṇḍeya explains
to Bhīṣma that what he is about to relate is something he observed first-hand,
as a result of having been given divine eyesight by Sanatkumāra in order to
understand the mysteries of the ancestors (Harivaṃśa 13.71–75). As before,
I summarize the upākhyāna here, chapter by chapter, with the colophon chap-
ter titles:57
Paolo Visigalli [Oxford: Oxbow Books, 2014], 26–27). It is perhaps in terms of the story of
that audience that the whole Mahābhārata can be an upākhyāna (see again n. 30 above).
54 See Brodbeck, The Mahābhārata Patriline, 233–57.
55 In the summary at Harivaṃśa appx 44 there is no independent mention of this upākhyāna,
but only of the Pitṛkalpa. The text reads as follows:
gālavotpattir ikṣvākuvaṃśasyāpy anukīrtanam | appx 44.5 |
pitṛkalpas tathotpattiḥ somasya ca budhasya ca | 6 |.
56 For diagrams and discussion of the dialogical arrangement, see Söhnen-Thieme, “The Set-
ting of Purāṇas,” 444–59. “Mārkaṇḍeya, who is ageless and has seen many cosmic eras, is
certainly a very suitable authority in historical matters … he draws on a higher source of
knowledge: Sanatkumāra, whose gāthās in the name of the ancestors had been quoted
before by Bhīṣma in the śrāddha section of the Anuśāsanaparvan (88.12 ff) … Sanatkumāra
is introduced in the HV [see 1.31 and 12.11] as the first-born son of Brahmā … This is as close
as one can get, in the HV, to Brahmā” (ibid., 458–59).
57 Cf. Dakshina R. Shastri, Origin and Development of the Rituals of Ancestor Worship in
India (Calcutta: Bookland Private Limited, 1963), 277–79. For translations, see Dutt, A
Prose English Translation of Harivamsha, 77–99; Bose, Harivamsha Translated into English
Prose, 59–78; Marcelle Saindon, trans. Le Pitrikalpa du Harivamsha: Traduction, Analyse,
Upākhyānas and the Harivaṃśa 407
the ancestors, they retain their memories (13–16). They are virtuous hunters;
after their parents have died, they renounce and die themselves (17–21). They
are reborn as deer, again retaining their memories; after meditating further on
their circumstances, they give up drinking and die (22–26). Eventually reborn
as cakravāka birds, still with their memories intact, they meditate and do yoga
(27–33). King Vibhrāja of the Nīpas passes through their forest in splendor, and
one of the birds wishes to be like him (34–37).
Harivaṃśa 17 (pitṛkalpaḥ, N M1–3; cakravākacaritam, T2 G1.3.5). Two of the
other birds wish to be the first bird’s ministers; and since they desired such
things, the three birds are cursed, by one of their companions, to get what they
want (1–4). Realizing their mistake, the three petition the remaining four for a
favor (5–7). They are told they will rise to yoga again when they hear a certain
verse, and that the one who becomes the king—that is, the one who, several
remembered births previously, insisted that the cow be dedicated to the ances-
tors—will know the language of every animal (8–11).
Harivaṃśa 18 (pitṛkalpaḥ, N M1.2; pitāputrasaṃvādaḥ, T2 G1.3.5). King
Vibhrāja, having been inspired by the yogic birds, returns to his capital (1–3).
His son Aṇuha marries (4–7). Vibhrāja retires to seek liberation, performing
austerities in the forest where he saw the birds (8–11). He gives his name to
the forest and its lake (12). The birds die and are reborn in Kāmpilya; four
retain their memories still, but three do not (13–14). The three are Aṇuha’s son
Brahmadatta—who knows the language of every animal, and is soon a king
with a yogic queen—and his two ministers (15–23); the four are poor brah-
man brothers (24–26). The four wish to leave their father, but he says they have
not yet done their filial duties (27–28). They tell him a verse that will yield his
fortune if he recites it to the king and his ministers; then they depart (29–32).
Harivaṃśa 19 (pitṛkalpaḥ, Ś1 K Ñ V B Dn Ds D1–3.5 T3.4 G2.3 M1; śrāddhakalpe
pitṛkalpaḥ, M2). Brahmadatta hears an ant pestering its lover, and laughs (1–4).
The queen is upset, thinking he’s laughing at her; he explains that he under-
stands animals, but she objects in disbelief and threatens suicide (5–9).
Brahmadatta petitions Nārāyaṇa for a solution, and is told one is on its way
(10–12). Brahmadatta and his two ministers return to town together; the father
of the four brahmans sees his moment and recites the verse to them (13–17). It
alludes to their births as hunters, deer, and birds (18). The three, stunned,
remember everything, and reward the brahman (19–22). Brahmadatta installs
his son as king and retires to the forest with his wife, who reveals that she was
only pretending not to believe he understood the ants (23–26). The king and
his two ministers resume their yoga, and attain perfection (27–29). Mārkaṇḍeya
concludes the story and extols its benefits (30–33). Vaiśaṃpāyana also con-
Upākhyānas and the Harivaṃśa 409
[118] Yudhiṣṭhira asks Bhīṣma what has become of all those who, willingly
or unwillingly, were slain in the great battle. In reply, Bhīṣma cites the
dialogue of Vyāsa with a creeping insect on a busy road. He asked it from
what it was fleeing in haste, and it answered that a large ox-cart was
approaching and might kill it. Vyāsa asked whether death might not be
preferable to life as an insect; the insect replied that living creatures love
whatever form of life they have. Formerly it had been a wicked Śūdra; the
Śūdra had, however, honoured his mother, and had once paid due hon-
our to a Brahmin guest, and for this reason the insect had retained
memory of its former birth, and had hopes of finding happiness once
more; it asked Vyāsa for his advice. [119] Vyāsa told the insect that he
would use his ascetic power to rescue it; he assured it that it would be
able to achieve an exalted birth. Hearing this, the insect remained
motionless upon the road, and so met its end; several births later it had
arrived at Kṣatriya status. The Kṣatriya returned to Vyāsa, paid him due
respect, and asked what he should do now; Vyāsa answered that he could
now attain brahminhood by giving up his life in battle for the sake of
cows or Brahmins. [120] The Kṣatriya undertook austerities, but Vyāsa
advised him that a Kṣatriya’s observance was to protect his subjects. The
Kṣatriya did so, and not long afterwards he passed away and was reborn
as a Brahmin. Bhīṣma assures Yudhiṣṭhira that those Kṣatriyas who died
at Kurukṣetra have also attained a happy outcome: he should not grieve
for them.61
There are several similarities between this story and that of the Sapta-
Upākhyāna: in both stories the effect of great sinfulness is counteracted, over a
sequence of lifetimes (sometimes featuring voluntary death), as an ongoing
result of a virtuous venerative act (or two, in this case) and the continuity of
memory that follows from it. It is a nice touch that Vyāsa understands the lan-
guage of every animal (sarvabhūtānāṃ rutajñaś, Mahābhārata 13.118.8). Much
could be made of these similarities (and the obvious differences), but for now
it will be enough to note the general effect: the Sapta-Upākhyāna plays on the
theme of a previous upākhyāna. In the following discussion, I focus upon an
issue emergent from the secondary literature.
The story of the Sapta-Upākhyāna is alluded to at Mahābhārata 12.330.38–
39, as mentioned by Yokochi and Söhnen-Thieme;62 and versions are to be
Those ones [i.e., sons of Bharadvāja] who had fallen away from their yoga
spent a very long time amongst the gods; then they were born as sons of
Kauśika—in Kurukṣetra, bull-man!67
This is the only mention in the story of an interval between two births on earth.
Yokochi considers the possibility that this time amongst the gods refers to a
separate birth, but dismisses it as “very unlikely … because their species (jāti)
is not specified there.”68 However, given the locative deveṣu, it would seem
natural to infer birth as gods. And since 14.3 describes the seven dying at the
end of their lives as Bharadvāja’s sons (te sarve saṃyuktāḥ kāladharmaṇā) and
the latter half of 14.4 describes their birth as sons of Kauśika, one wonders how
the intervening “very long time amongst the gods” could possibly not be
another jāti. Are we to envisage suspended animation? Isn’t the realm of the
gods a plane of rebirth? It might seem that Yokochi’s desire to find the recon-
stituted narrative defective and replace it with a narrative of textual history has
forced her interpretation.69
In any case, there is no need to worry about the precise number of births
related. After the seven have died at the end of their lives as deer, we read the
following:
Because of that pure behavior, my boy, they went through purer and
purer births until, devoid of impurity, they became cakravāka birds.70
than as a description of the specific rebirth just described; and the words
aśubhavarjitāḥ … cakravākatvam āgatāḥ mark the limit case at the end of that
process. So if, for example, the story had been said to be a story of twenty-seven
births, then this verse could be summarizing twenty births that intervened
between the birth as deer and the birth as cakravāka birds, but which are not
specified in detail. Reasons for this being said to be a story of seven births are
not hard to find—it matches the number of characters undergoing rebirth in
the story; it matches the number of different kinds of pitṛ described in the
immediately preceding chapter 13; and so on—but once the pattern of rebirth
has been set, there is no need for the stated number of rebirths to match the
number of rebirths actually narrated. All that is required is for there to be
enough rebirths to make the point—amongst various others—that no matter
how pure one’s birth is, one is never out of reach of desire, especially desire for
glorious kingship.
That point is surely of special relevance to Bhīṣma. He has resisted the
temptations of kingship in this life, and so it fits that he is featured here in the
narratorial chain, having carried the story from when he was told it by
Mārkaṇḍeya at a śrāddha rite for his father Śaṃtanu (Harivaṃśa 11.40–41), to
when he now tells it to Yudhiṣṭhira on his deathbed (Harivaṃśa 11.5–6), at the
end of a life of filial duty,72 waiting for the uttarāyaṇa and his next postmor-
tem adventure (Mahābhārata 6.114.86–100; cf. 6.30.23–27 = Bhagavadgītā
8.23–27). But as emphasized fore and aft, the story purports to have beneficial
effects more generally. After the overview has been given in Harivaṃśa 14,
mentioning the gradual yogic perfection achieved by the seven on account of
their homage to the ancestors, we read the following:
And in the same way, your own thoughts will return again and again to
your duty, and absorbed in your habit of yoga you will attain the highest
perfection. For there’s no duty more pressing than yoga duty, you who
know your duty! Practice that most important duty of all, Bhārgava! If
you perform the śrāddha rite, if you prioritize it and are dedicated to it,
eating lightly, your senses conquered, then in the fullness of time you will
get into the habit of yoga.73
72 From this perspective the narration of Bhīṣma’s encounter with Ugrāyudha in Harivaṃśa
15 is not as tangential as it might initially appear, for it emphasizes not only his dedication
to his father, but also his dedication to the family (and thus the ancestors) in general. At
15.45 Bhīṣma says he decided to fight Ugrāyudha “seeing as Vicitravīrya was young and
was totally dependent on me” (vicitravīryaṃ bālaṃ ca madapāśrayam eva ca | dṛṣṭvā).
73 evaṃ dharme ca te buddhir bhaviṣyati punaḥ punaḥ |
yogadharme ca nirataḥ prāpsyase siddhim uttamām || Harivaṃśa 14.9 ||
414 Brodbeck
On that previous occasion these events were visible to me, just so. Heed
them, lofty son of Gaṅgā, and you will then be crowned with fortune. And
others who heed the amazing career of those [seven] will never be born
from any animal womb whatsoever. This upākhyāna is of great import,
and the great seek it out; whoever hears it always keeps their yoga prac-
tice in mind, Bhārata, and achieves tranquility in due course as a result,
and thus arrives at a state of mind that is hard to attain on earth, even for
the siddhas.76
tion of guru Droṇa in the matter of a killing (7.164.66–110). And when Arjuna
hears Bhīṣma’s narration of this story to Yudhiṣṭhira,77 he might recall what
happened when, quailing before the same dastardly deed, he asked Kṛṣṇa what
happens to a yogi who dies before perfecting their yoga: he was told that their
rebirth is such as to facilitate its continuation, and that “perfected over the
course of several lives, they attain the highest state” (anekajanmasaṃsiddhas
tato yāti parāṃ gatim; Mahābhārata 6.28.45cd = Bhagavadgītā 6.45cd).
Syamantaka-Upākhyāna
This upākhyāna forms the last two chapters of Vaiśaṃpāyana’s account of the
lunar dynasty (Harivaṃśa 20–29), which in its latter stages has focused on
Kṛṣṇa’s people, the Yādava-Vṛṣṇis. The episode is mentioned in the summary at
Harivaṃśa appx 44, but the word upākhyāna is not used (kīrtanaṃ kṛṣṇasaṃ
bhūteḥ syamantakamaṇes tathā | appx 44.10). As before, I summarize:78
Harivaṃśa 28 (syamantakalābhaḥ, K1.3; syamantakapratyānayanam, K2
V2 B2 Ds; kṛṣṇasya mithyābhiśaptiḥ, K4; mithyābhiśaptiḥ, Ñ2 V3; sya
mantakopākhyānam, B1 D1–3.6 T2 G2.3 M2; syamantakākhyānam, D4.5;
syamantakamaṇikathanam, T4; syamantakavyākhyānam, G1). Genealogy of
(Andhaka’s son) Bhajamāna’s descendants, including Kṛtavarman and Śata
dhanvan (1–8). Genealogy from Kroṣṭu to the brothers Prasena and Satrājit
(9–11). Prasena obtains the Syamantaka jewel, which has beneficial properties
(12–13); Kṛṣṇa wants but doesn’t take it (14). Prasena, out hunting, is killed by a
lion, and Jāmbavat, a bear, takes the jewel (15–16). Kṛṣṇa, suspected of Prasena’s
murder, sets out to find the jewel (17–18). He follows the trail to Jāmbavat’s cave,
where he hears the jewel being sung about (19–24). Kṛṣṇa enters and fights
Jāmbavat; after twenty-one days his companions leave him for dead and go
home (25–27). Kṛṣṇa conquers Jāmbavat, takes his daughter Jāmbavatī and the
jewel, returns to Dvārakā, and gives the jewel to Satrājit in public, thus clearing
his name (28–31). Genealogy of Satrājit’s descendants, including his daughters,
whom he gives to Kṛṣṇa (32–35). Genealogy of Pṛśni’s descendants, including
Akrūra (36–43). The effect of knowing this chapter (45).
79 Christopher Austin, “The Mystery of the Syamantaka Jewel: The Intersection of Geneal-
ogy and Biography in the Harivaṃśa,” Religions of South Asia 5 (2012): 153–69; see also
Freda Matchett, “The Harivaṁśa: Supplement to the Mahābhārata and Independent
Text,” Journal of Vaishnava Studies 4, no. 3 (1996): 144–45.
80 Simon Brodbeck, “Mercy, my Jewels! Aśvatthāman’s Maṇi and the Syamantaka (and the
Earrings),” paper presented at the Sixth Dubrovnik International Conference on the San-
skrit Epics and Purāṇas (Dubrovnik, August 2011).
81 André Couture, “L’affaire du joyau Syamantaka: Un récit clé pour l’interprétation du
Harivaṃśa,” Journal Asiatique 301, no. 1 (2013): 139–83 (with English summary).
Upākhyānas and the Harivaṃśa 417
The Krishna whom [the author] presents here as part of the dynasty of
Yadu is seen from a purely human point of view, before Hv 30-45 has
revealed him as Vishnu’s earthly manifestation and counterpart.85
In these terms the story would be of interest to King Janamejaya, who is being
aided to reassert his clear paramount status through a robust but clement
sarpasatra-and-aśvamedha. If the Pṛthu-Upākhyāna reassured Janamejaya of
82 Ivan Strenski, “The Syamantaka Gem Story: A Structural Analysis,” Purāṇa 24, no. 2 (1982):
297–337.
83 Madhukar A. Mehendale, “Nirukta Notes IV: Yāska’s Etymology of Daṇḍa,” Journal of the
American Oriental Society 80, no. 2 (1960): 112–15.
84 Austin, “The Mystery of the Syamantaka Jewel,” 154.
85 Matchett, “The Harivaṁśa: Supplement to the Mahābhārata and Independent Text,” 145.
86 Austin, “The Mystery of the Syamantaka Jewel,” 161–62. Brinkhaus evokes something sim-
ilar when he says that the Syamantaka episode “obviously endeavours to turn Kṛṣṇa’s
trickster image in the epic into one of an honest and reliable individual.” Horst Brinkhaus,
“The 16,108 Wives of Kṛṣṇa in the Harivaṃśa,” in Battle, Bards and Brāhmins, ed. John
Brockington (Delhi: Motilal Banarsidass, 2012), 320. What Brinkhaus claims here is not
obvious, and stands in need of explanation (for example, one wonders what Kṛṣṇa would
have done if Śatadhanvan had had the jewel); on the face of it, Kṛṣṇa’s character is con-
tinuous across the whole Mahābhārata.
87 Austin, “The Mystery of the Syamantaka Jewel,” 168.
418 Brodbeck
the brahmans’ support for the existence of his office, the Syamantaka-Upā
khyāna reminds him—through a story of the Vṛṣṇis who he knows destroyed
themselves at Prabhāsa, the remnant coming north into his domains (Mahā
bhārata 16)—of the dreadful in-fighting that can result when that office is not
properly occupied. In this respect, it resembles the story of weak king
Dhṛtarāṣṭra; and we may recall that Dhṛtarāṣṭra is a snake’s name88—thus link-
ing Yudhiṣṭhira’s battle against Kauravas with Janamejaya’s battle against
snakes, and linking Vṛṣṇis with snakes.89
The narrated Syamantaka story is not celebrated as a unit by phala verses in
the way that the Pṛthu-Upākhyāna and Sapta-Upākhyāna are. In Vaiśaṃpāyana’s
narration of the solar and lunar dynasties, a phala verse at the end of a chapter
usually gives a label for that chapter.90 The Syamantaka story has such a verse
at the end of its first chapter:
This, as narrated, is “The False Accusation against Kṛṣṇa.” And those who
accuse falsely can never touch anyone who knows it.91
This phala relates directly to the events portrayed in the chapter. It would seem
to grant Janamejaya, amongst others, invulnerability to false accusation. In
Janamejaya’s case, we could imagine the possible accusation—on the model of
the mongoose accusing Yudhiṣṭhira after the aśvamedha (Mahābhārata 14.92–
93, in the Nakula-Upākhyāna)—that his recent ritual violence was illegitimate
and/or perpetrated through greed.
88 Gösta Johnsen, “Varuṇa and Dhṛtarāṣṭra,” Indo-Iranian Journal 9, no. 4 (1966): 256.
89 See Brodbeck, “Refuge and Reform,” 27–28.
90 This is the practice visible also in Harivaṃśa 1, 2, 3, and (to some extent) 7; see n. 52 above.
There are summative phala verses for chapters 8 (8.48, janma devānāṃ, “The Birth of the
Gods”), 9–10 (10.80, sṛṣṭim ādityasya vivasvataḥ, “The Offspring of Vivasvat Āditya”), 20
(20.47–48, somasya janma, “The Birth of Soma”), 21 (21.37, cyāvanaṃ sthānāt pratiṣṭhāṃ
ca śatakratoḥ, “Indra’s Expulsion from and Return to his Domain”), 22 (22.45, yayāteś
caritaṃ, “The Story of Yayāti”), 23 (23.165–66, pañcavisargaṃ, “The Offspring of the Five,”
incorporating also kārtavīryasya janma, “The Life of Kārtavīrya,” whose phala verse is at
23.163), 24 (23.168, phala given in advance for kroṣṭor … vaṃśaṃ, “The Line of Kroṣṭu,”
known in retrospect as vṛṣṇes trividham … vaṃśaṃ, “The Triple Line of Vṛṣṇi,” at 24.35), 25
(25.17, kṛṣṇasya janma, “The Birth of the Dark One”), 26 (26.28, visṛṣṭiṃ … jyāmaghasya
mahātmanaḥ, “The Offspring of Jyāmagha the Great”), and 27 (27.31, kukurāṇām …
vaṃśaṃ … amitaujasām, “The Line of the Immeasurably Powerful Kukuras”). So, except-
ing the Pitṛkalpa (Harivaṃśa 11–19), every chapter in the lineal narration so far has a phala
and label except chapters 9 and 10, which have one between them.
91 imāṃ mithyābhiśastiṃ yaḥ kṛṣṇasya samudāhṛtām |
veda mithyābhiśāpās taṃ na spṛśanti kadācana || Harivaṃśa 28.45 ||.
Upākhyānas and the Harivaṃśa 419
There is no phala verse at the end of the Syamantaka story’s second chapter,
either for the chapter or for the story, or for the account of the lunar line (chap-
ters 20–29), or for the Sūryavaṃśa–Pitṛkalpa–Somavaṃśa bundle (chapters
8–29) that began soon after the Pṛthu-Upākhyāna, or for the Harivaṃśaparvan
which has often been understood to end just here, before Janamejaya asks his
big question about Viṣṇu (chapter 30; see n. 20 above; Couture begins “l’enfance
de Krishna” at chapter 3092). The absence of any phala verse here might seem
to have the effect of dramatizing Janamejaya’s major interruption at the
outset.
Dhanya-Upākhyāna
… and the killing of Śambara [chapter 99], and the story of the blessing
[Dhanya-Upākhyāna, chapter 100], the majesty of Vāsudeva [chapters
101–5], the fight against Bāṇa, which is treated at length [chapters 106–
13] … 93
As before, I summarize:94
Harivaṃśa 100 (dhanyopākhyānam, K Ñ V B Dn Ds D1.2.4–6; dhanyāścaryo
pākhyānam, D3; āryākhyāne paramārthadarśanam, T1 G3; āścaryopākhyānam,
T2 G1.4 M4; āścaryakathanam, T4 M1–3). Jāmbavatī’s son Sāmba is born, and
Kṛṣṇa comes home after defending Vṛṣṇi interests abroad; Dvārakā prospers
(1–4). Kings visiting Duryodhana in Hāstinapura also come to Dvārakā en
masse, to visit Kṛṣṇa and the Yadus (5–10). Kṛṣṇa hosts them in style, and sto-
ries are told (11–16). There is sudden bad weather and Nārada arrives from
inside a cloud; the bad weather disappears (17–20). Nārada tells Kṛṣṇa that
Kṛṣṇa is a marvel (āścarya) and a blessing (dhanya); Kṛṣṇa replies that he is
indeed, and that he is also accompanied by the dakṣiṇā; Nārada says his
92 André Couture, trans. L’enfance de Krishna: Traduction des chapitres 30 à 78 (éd. cr.) (Paris:
Éditions du Cerf / Québec: Presses de l’Université Laval, 1991).
93 śambarasya vadhaś caiva dhanyopākhyānam eva ca | Harivaṃśa appx 44.55 |
vāsudevasya māhātmyaṃ bāṇayuddhaṃ prapañcitam | 56 |.
94 Cf. Vaidya, “Introduction,” xxvii–xxviii; Vaidya, The Harivaṁśa, vol. 1, 797. For translations,
see Dutt, A Prose English Translation of Harivamsha, 723–28; Bose, Harivamsha Translated
into English Prose, 412–15.
420 Brodbeck
utterance has been completed, and makes to leave (21–24). The assembled
kings, puzzled, ask Kṛṣṇa about these cryptic utterances; Kṛṣṇa asks Nārada to
explain; Nārada sits down and begins (25–31). Nārada once met a massive tur-
tle in the River Gaṅgā, and declared it to be a marvel and a blessing (32–36).
The turtle said that the River Gaṅgā was a marvel and a blessing, not he, and
explained why (37–38). Nārada visited in turn the River Gaṅgā (39–43), the
ocean (44–47), the earth (48–53), the mountains (54–57), Brahmā (58–66), the
Vedas (67–72), and the sacrifices (73–77), in each case accosting them with the
accolade of being a marvel and a blessing, and in each case being referred to
the next. The sacrifices referred him to Viṣṇu, so here he is (78–80). And that’s
why he called Kṛṣṇa a marvel and a blessing; and Kṛṣṇa responded that he is
indeed, and that he is also accompanied by the dakṣiṇā, which was the com-
pletion of the utterance (81–85). Nārada leaves; the kings go home amazed;
Kṛṣṇa and the Yādavas go home too (86–87).
This being the Mahābhārata’s last upākhyāna, it is significant that it con-
cerns the extension of a string of utterance, from a riverine origin, on and on in
a series that ends importantly and precisely with Kṛṣṇa the marvel and the
blessing and, along with him, with the dakṣiṇā. In this way, Nārada’s trek is
deliberately made to stand alongside our own trek as Mahābhārata readers. It
is as if the passage through the upākhyānas were to stand for the passage
through the text as a whole. For the āścarya of this upākhyāna is also the
āścarya of this parvan, the Viṣṇuparvan or Āścaryaparvan,95 which is men-
tioned last and cryptically (as adbhutaṃ mahat and prakīrtitam) in the list
of the one hundred and one Mahābhārata parvans (Mahābhārata 1.2.69, 233;
pp. 393–94 above). Thus the importance of Nārada continuing to the end of his
treasure hunt emphasizes the importance of getting to the end of the Mahā
bhārata; and this can be so despite it being the case that, from here within
Harivaṃśa 100, there is still one full parvan to go (in this sense the Bhaviṣyat
parvan returns to its sequential location as parvan number one hundred and
one), and—wouldn’t you just know it!—exactly eighteen chapters to go.96 As
Biardeau said, “You have to read the whole thing”97—for the dakṣiṇā is distrib-
uted only at the completion of the rite, and only at the completion of the
utterance.
Conclusions
Hiltebeitel has argued, following David Shulman, that this verse would describe
“a digest or abridgment” rather than an earlier text that has since expanded.99
The survey and discussion of the Harivaṃśa’s upākhyānas presented above
in no way supports the idea of such an earlier text. If Mahābhārata 1.1.61 and
the prehistorical interpretation thereof had not already put that idea into our
heads, there seems to be no reason why we would imagine, for example, that
Harivaṃśa 28–29 is a younger part of the Mahābhārata than Harivaṃśa 27,
or that Harivaṃśa 100 is a younger part of the Mahābhārata than Harivaṃśa
101–5. But Mahābhārata 1.1.61 is a puzzling verse nonetheless, because the idea
of a text more or less resembling the one we have, but without the upākhyānas,
is a very bizarre idea whether that text is imagined to be earlier than the one
we have, or later. At least hypothesizing that the shorter text would be ear-
lier makes sense of the presence of Mahābhārata 1.1.61 itself, which then could
presumably have been added after or at the same time as the upākhyānas. If
the abridgement is later than the full text, then either 1.1.61 must refer in the
past perfect tense (cakre, “he made”) to something that will only happen in the
future (i.e., at the time of the completion of the full text, the abridgement has
not yet been made), or the text that was abridged is not the text we have before
us now—in which case the text without upākhyānas might be an abridge-
ment and also part of the prehistory of the text that we have. Although I would
not argue that 1.1.61 was designed to prompt that particular thought, nor do I
suppose that it refers to an actual, more or less 24,000-verse “digest or abridg-
ment that knowers of the Mahābhārata could consult or cite for purposes of
performance from a written text.”100 Rather, I think it is a riddle designed to
prompt the reader to think about upākhyānas in various ways—to think about
the precision of the label, about what exactly it might refer to, and about the
relationship between that material and the material that the label excludes,
in more or less the ways that Hiltebeitel and some of the other contributors to
the larger upākhyāna project have been doing. In other words, that project was
imagined and implicitly proposed by the text’s author/s.
If the idea of a Bhārata without upākhyānas is bizarre, it is probably less
bizarre, and more textually justified (it is explicitly mentioned at Mahābhārata
1.1.61, after all), than the idea of the upākhyānas without the Bhārata; and yet
the way in which I have presented the Harivaṃśa’s upākhyānas in this paper
risks prompting that latter idea. The reader must counter it by remembering
that the contextual dependence of the upākhyānas upon the surrounding text
goes far beyond the connections I have pointed out here, even though I have
pointed out many such connections. That is to say: if one were to try to give
some realistic substance to the idea of the upākhyānas without the Bhārata
100 Ibid. Here, as elsewhere, if one believes what the text says (i.e., if one takes it to be stating
historical fact), one should presumably explain why the text is believed here and not, for
example, when it says that there once was a brahman born from a pot. In this regard, one
does not get into difficulties by interpreting Mahābhārata 1.1.61 in terms of a text prior to
the Mahābhārata (indeed, this seems to be a reasonable interpretation); one gets into dif-
ficulties only if one then thinks that the verse so interpreted could provide support for
any actual theory of the text’s prehistory. Cf. Brodbeck, The Mahābhārata Patriline, 9.
A contrary interpretation of 1.1.61 is not required—and in fact is not useful—in order to
disarm the alleged text-historical implication.
Upākhyānas and the Harivaṃśa 423
by imagining versions of them different from the versions that are in the
Mahābhārata, then they would not be the upākhyānas that the Bhārata with-
out upākhyānas would be without.
Probably more interesting than either of these two ideas (of the Bhārata and
the upākhyānas without each other) is the idea of the upākhyānas as stepping-
stones through the Mahābhārata. In this sense, we can understand the
upākhyānas not only as they relate to the text generally (i.e., to the whole) and
contextually (i.e., to their surroundings at the points where they occur), but
also as they relate to each other as co-items within a series. I have made remarks
in this regard concerning the Pṛthu-Upākhyāna and the Vainya-Upākhyāna,
concerning the Sapta-Upākhyāna and the Kīṭa-Upākhyāna, and concerning
the Dhanya-Upākhyāna as the last in the series. It is also curious that that
Dhanya-Upākhyāna, Harivaṃśa 100, begins with the birth of Sāmba, who is the
son of Jāmbavatī, the bear-maiden taken by Kṛṣṇa at Harivaṃśa 28.28 in the
Syamantaka-Upākhyāna,101 and who was born through a boon granted by Śiva
and Umā as described in Mahābhārata 13.14–16 in the Upamanyu-Upākhyāna.102
Further study of the upākhyānas may throw up further examples of this kind
of connection.
References
Primary Sources
Atharva Veda Sanhita. Edited by Rudolf von Roth and William D. Whitney. Berlin:
Dümmler, 1856.
The Harivaṁśa, Being the Khila or Supplement to the Mahābhārata, for the First Time
Critically Edited. Edited by Parashuram L. Vaidya. 2 vols. Poona: Bhandarkar Oriental
Research Institute, 1969–71.
The Mahābhārata for the First Time Critically Edited. Edited by Vishnu S. Sukthankar,
Shripad K. Belvalkar, Parashuram L. Vaidya, et al. 19 vols. Poona: Bhandarkar Oriental
Research Institute, 1933–66.
The Mahābhārata Text as Constituted in its Critical Edition. General Editor: Ramchandra
N. Dandekar. 5 vols. Poona: Bhandarkar Oriental Research Institute, 1971–76.
101 Also in the Syamantaka-Upākhyāna we hear of Sāmba’s wife Sundarī (Harivaṃśa 28.40–
41).
102 Georg von Simson, “Kṛṣṇa’s Son Sāmba: Faked Gender and Other Ambiguities on the
Background of Lunar and Solar Myth,” in Gender and Narrative in the Mahābhārata, ed.
Simon Brodbeck and Brian Black (London: Routledge, 2007), 235–36.
424 Brodbeck
Secondary Sources
Adluri, Vishwa. “Frame Narratives and Forked Beginnings: Or, How to Read the
Ādiparvan.” Journal of Vaishnava Studies 19, no. 2 (2011): 143–210.
———. “Hermeneutics and Narrative Architecture in the Mahābhārata.” In Ways and
Reasons for Thinking about the Mahābhārata as a Whole, edited by Vishwa Adluri,
1–27. Pune: Bhandarkar Oriental Research Institute, 2013.
———. “Literary Violence and Literal Salvation: Śaunaka Interprets the Mahābhārata.”
Exemplar: The Journal of South Asian Studies 1, no. 2 (2012): 45–68.
Austin, Christopher. “The Mystery of the Syamantaka Jewel: The Intersection of
Genealogy and Biography in the Harivaṃśa.” Religions of South Asia 5 (2012):
153–69.
Bailey, Gregory M. “Brahmā, Pṛthu and the Theme of the Earth-Milker in Hindu
Mythology.” Indo-Iranian Journal 23, no. 2 (1981): 105–16.
Bose, Dhirendra N., ed. Harivamsha Translated into English Prose from the Original
Sanskrit Text. Dum Dum, Bengal: Datta Bose and Co., n.d.
Bowles, Adam. Dharma, Disorder and the Political in Ancient India: The Āpaddharmaparvan
of the Mahābhārata. Leiden: E.J. Brill, 2007.
Brinkhaus, Horst. “The Division into Parvans and the Bhaviṣyaparvan of the Harivaṃśa.”
In Stages and Transitions: Temporal and Historical Frameworks in Epic and Purāṇic
Literature, edited by Mary Brockington, 157–76. Zagreb: Croatian Academy of Sciences
and Arts, 2002.
———. “Manu Vaivasvata as Śrāddhadeva: On the Insertion of the Pitṛkalpa into the
Harivaṃśa.” In Epic Undertakings, edited by Robert P. Goldman and Muneo Tokunaga,
1–11. Delhi: Motilal Banarsidass, 2009.
———. “The 16,108 Wives of Kṛṣṇa in the Harivaṃśa.” In Battle, Bards and Brāhmins,
edited by John Brockington, 315–28. Delhi: Motilal Banarsidass, 2012.
Brockington, John L. “The Spitzer Manuscript and the Mahābhārata.” In From Turfan to
Ajanta: Festschrift for Dieter Schlingloff on the Occasion of his Eightieth Birthday,
edited by Eli Franco and Monika Zin, 75–87. Lumbini: Lumbini International
Research Institute, 2010.
Brodbeck, Simon. “Analytic and Synthetic Approaches in Light of the Critical Edition
of the Mahābhārata and Harivaṃśa.” Journal of Vaishnava Studies 19, no. 2 (2011):
223–50.
———. “Gendered Soteriology: Marriage and the Karmayoga.” In Gender and Narrative
in the Mahābhārata, edited by Simon Brodbeck and Brian Black, 144–75. London:
Routledge, 2007.
Upākhyānas and the Harivaṃśa 425
———. “Janamejaya’s Big Brother: New Light on the Mahābhārata’s Frame Story.”
Religions of South Asia 2, no. 2 (2009): 161–76.
———. The Mahābhārata Patriline: Gender, Culture, and the Royal Hereditary. Farnham:
Ashgate Publishing, 2009.
———. “Mercy, my Jewels! Aśvatthāman’s Maṇi and the Syamantaka (and the Earrings).”
Paper presented at the Sixth Dubrovnik International Conference on the Sanskrit
Epics and Purāṇas. Dubrovnik, August 2011.
———. “Refuge and Reform: Snakes, Gleaners, and Niṣādas in Early Kāvya.” In Puṣpikā:
Tracing Ancient India through Texts and Traditions. Contributions to Current Research
in Indology, Volume 2, edited by Giovanni Ciotti, Alastair Gornall, and Paolo Visigalli.
Oxford: Oxbow Books, 2014.
———. “Solar and Lunar Lines in the Mahābhārata.” Religions of South Asia 5 (2012):
127–52.
———. “The Story of Sāvitrī in the Mahābhārata: A Lineal Interpretation.” Journal of
the Royal Asiatic Society of Great Britain and Ireland (third series) 23, no. 4 (2013):
527–49.
Clercq, de, Eva. “Karman and Compassion: Animals in the Jain Universal History.”
Religions of South Asia 7 (2013): 141–57.
Couture, André. “L’affaire du joyau Syamantaka: Un récit clé pour l’interprétation du
Harivaṃśa.” Journal Asiatique 301, no. 1 (2013): 139–83.
———, trans. L’enfance de Krishna: Traduction des chapitres 30 à 78 (éd. cr.). Avec intro-
duction, annotations et index. Paris: Éditions du Cerf / Québec: Presses de l’Université
Laval, 1991.
Dunham, John. “Manuscripts Used in the Critical Edition of the Mahābhārata: A Survey
and Discussion.” In Essays on the Mahābhārata, edited by Arvind Sharma, 1–18.
Leiden: E.J. Brill, 1991.
Dutt, Manmatha N., trans. A Prose English Translation of Harivamsha. Calcutta: Elysium
Press, 1897.
Edgerton, Franklin. Review of The Mahābhārata for the First Time Critically Edited, by
Vishnu S. Sukthankar. Journal of the American Oriental Society 48, no. 2 (1928):
186–90.
Fitzgerald, James L. “Introduction.” In The Mahābhārata, vol. 7. 11. The Book of the Women.
12. The Book of Peace, Part One, 79–164. Chicago: University of Chicago Press, 2004.
———. “The Sāṃkhya-Yoga ‘Manifesto’ at Mahābhārata 12.289–290.” In Battle, Bards
and Brāhmins, edited by John Brockington, 259–300. Delhi: Motilal Banarsidass, 2012.
Hiltebeitel, Alf. “Among Friends: Marriage, Women, and Some Little Birds.” In Gender
and Narrative in the Mahābhārata, edited by Simon Brodbeck and Brian Black, 110–43.
London: Routledge, 2007.
———. “Between History and Divine Plan: The Mahābhārata’s Royal Patriline in
Context.” Religions of South Asia 5 (2012): 103–25.
426 Brodbeck
———. Dharma: Its Early History in Law, Religion, and Narrative. New York: Oxford
University Press, 2011.
———. “The Geography of the Mahābhārata’s Upākhyānas.” Paper presented at the
Sixth Dubrovnik International Conference on the Sanskrit Epics and Purāṇas.
Dubrovnik, August 2011.
———. “Kṛṣṇa and the Mahābhārata: A Bibliographical Essay.” Annals of the Bhandarkar
Oriental Research Institute 60 (1979): 65–107.
———. “The Nārāyaṇīya and the Early Reading Communities of the Mahābhārata.” In
Between the Empires: Society in India 300 bce to 400 ce, edited by Patrick Olivelle,
227–56. New York: Oxford University Press, 2006.
———. “Not Without Subtales: Telling Laws and Truths in the Sanskrit Epics.” Journal
of Indian Philosophy 33, no. 4 (2005): 455–511.
———. Rethinking the Mahābhārata: A Reader’s Guide to the Education of the Dharma
King. Chicago: University of Chicago Press, 2001.
———. “‘You Have to Read the Whole Thing’: Some Reflections on Madeleine Biardeau’s
Mahābhārata.” In Du texte au terrain, du terrain au texte. Dialogues disciplinaires
autour de l’oeuvre de Madeleine Biardeau. Journée 2011 du Centre d’Études de l’Inde
et de l’Asie du Sud. Paris, 2011. <http://ceias.ehess.fr/docannexe/file/3137/journee_
biardeau_intervention_alf_hiltebeitel.pdf>.
Huntington, Ronald M. “The Legend of Pṛithu: A Study in the Process of Individuation.”
Purāṇa 2 (1960): 188–210.
Johnsen, Gösta. “Varuṇa and Dhṛtarāṣṭra.” Indo-Iranian Journal 9, no. 4 (1966): 245–65.
Matchett, Freda. “The Harivaṁśa: Supplement to the Mahābhārata and Independent
Text.” Journal of Vaishnava Studies 4, no. 3 (1996): 139–50.
Mehendale, Madhukar A. “Nirukta Notes IV: Yāska’s Etymology of Daṇḍa.” Journal of the
American Oriental Society 80, no. 2 (1960): 112–15.
Menon, K.P.A., trans. The Harivamśapurāṇam (Text with English Translation). Delhi: Nag
Publishers, 2008.
Miller, Jeanine. “Predestiny and Evil in Hindu Myth: An Inquiry into the Symbolism of
the Vena-Pṛthu Legend.” In Symbols in Art and Religion: The Indian and the Comparative
Perspectives, edited by Karel Werner, 103–28. Delhi: Motilal Banarsidass, 1991.
Minkowski, Christopher Z. “Janamejaya’s Sattra and Ritual Structure.” Journal of the
American Oriental Society 109, no. 3 (1989): 401–20.
———. “Snakes, Sattras, and the Mahābhārata.” In Essays on the Mahābhārata, edited
by Arvind Sharma, 384–400. Leiden: E.J. Brill, 1991.
Monier-Williams, Monier. A Sanskrit–English Dictionary. Oxford: Clarendon Press, 1964.
O’Flaherty, Wendy Doniger. The Origins of Evil in Hindu Mythology. Berkeley: University
of California Press, 1976.
Pandey, Jagdishwar. “The Story of Brāhmaṇa Brothers in the Harivaṃśa Purāṇa:
A Critical Study.” Journal of the Bihar Research Society 63–64 (1977–78): 723–28.
Upākhyānas and the Harivaṃśa 427
Chapter 14
1 The phrase is Fitzgerald’s (see James L. Fitzgerald, “Negotiating the Shape of ‘Scripture’: New
Perspectives on the Development and Growth of the Mahābhārata between the Empires,” in
Between the Empires: Society in India 300 bce to 400 ce, ed. Patrick Olivelle [Oxford: Oxford
University Press, 2006], 272–73).
2 Rajesh Purohit said this in affirmation of my view of the integral character of the Mahābhārata’s
upākhyānas at the February 2011 conference “Jaya Utsava: Celebrating Living Traditions of
Mahabharata,” Indira Gandhi National Centre for the Arts, New Delhi.
3 Alf Hiltebeitel, “Not Without Subtales: Telling Laws and Truths in the Sanskrit Epics,” Journal
of Indian Philosophy 33, no. 4 (2005): 455–511.
4 M.M. Bakhtin, The Dialogic Imagination, trans. Caryl Emerson and Michael Holquist (Austin:
University of Texas Press, 1981).
5 See Alf Hiltebeitel, Reading the Fifth Veda: Studies on the Mahābhārata. Essays by Alf Hiltebeitel,
vol. 1, ed. Vishwa Adluri and Joydeep Bagchee (Leiden: E.J. Brill, 2011), chapter 4: “Why Itihāsa?
New Possibilities and Limits in Considering the Mahābhārata as History”; Alf Hiltebeitel,
Dharma: Its Early History in Law, Religion, and Narrative (New York: Oxford University Press,
2011), 4–5, 18 n. 25 and Index s.v. “itihāsa”; Alf Hiltebeitel, “The Mahābhārata and the Stories
Some People Tell about Its Tribal and Earliest Histories,” paper presented at the International
this volume, we may look at upākhyānas from the standpoint that upa- con-
notes “near” rather than “sub-”: that is, the term functions not only in terms
of proximity from below (as in “sub”-tales), but in terms of spatio-temporal
nearness to the larger epic ākhyāna. In short compass, by raising these ques-
tions we will be making our way into what the Mahābhārata means by both
the local and the regional. And on a wider scale, the Mahābhārata’s horizons
make it the first Indian text to envision a regional area such as Kurukṣetra and
the upper doab within India (Bharatavarṣa) as a total land and a total peo-
ple. Set in a still wider world, and indeed a vast cosmology,6 we will also be
exploring how upākhyānas help to build the Mahābhārata’s geography into its
cosmography—or, more suggestively, in a formulation I owe to a recent article
by Randy Kloetzli,7 make the whole believable by building its cosmograph into
its geograph. This sense of the nearness of what by itself seems distant also sus-
tains the value of continuing to look primarily at a finite list of units actually
called upākhyānas while also recognizing, as several have done in this volume,
that other “near-tales” can look and act within the larger epic more or less like
upākhyānas as well.
Seminar on the Mahābhārata: Its Historicity, Antiquity, Evolution & Impact on Civilization,
April 26, 2012; Alf Hiltebeitel, “Between History and Divine Plan: The Mahābhārata’s Royal
Patriline in Context,” papers from the May 2010 Cardiff Conference on Genealogy in India,
Religions of South Asia 5 (2011): 103–25.
6 On Bharatavarṣa-India conceived as a total land and people, see Aloka Parasher-Sen, Mlecchas
in Early India, a Study in Social Attitudes towards Outsiders up to ad 600 (New Delhi: Munshiram
Manoharlal, 1991), 91 (dating the Mahābhārata, however, later than Manu); Alf Hiltebeitel,
“The Southern Recension’s Śakuntalā as a First Reading: A Window on the Original and the
Second Reading by Kālidāsa,” in Revisiting Kālidāsa’s Abhijñāna Śākuntalam: Land, Love,
Languages: Forms of Exchange in Ancient India, ed. Deepika Tandon and Saswati Sengupta
(Delhi: Orient BlackSwan Edition, 2011), 22; “On Sukthankar’s ‘S’ and Some Shortsighted
Assessments and Uses of the Pune Critical Edition (ce),” Journal of Vaishnava Studies 19, no.
2 (2011): 97, 102–3 (whether earlier or younger, the Rāmāyaṇa envisions India as a total land
but not as a total people; see Alf Hiltebeitel, “Authorial Paths through the Two Sanskrit Epics,
Via the Rāmopākhyāna,” in Epic Undertakings: Papers of the 12th World Sanskrit Conference,
vol. 2, ed. Robert P. Goldman and Muneo Tokunaga [Delhi: Motilal Banarsidass, 2009],
169–214).
7 Randolph W. Kloetzli, “Ptolemy and Purāṇa: Gods Born as Men,” Journal of Indian Philosophy
38 (2010): 583–623.
430 Hiltebeitel
Since my idea to research this topic began with an inspiration from Kurukṣetra,
that will be a good place to start. What might have prompted someone with
Rajesh Purohit’s intimate knowledge of Kurukṣetra to make such a geographi-
cal proposition about the Mahābhārata’s upākhyānas? I will work here from a
simple fourfold classification of the epic’s upākhyānas: those in Book 1; Book 3;
those from Books 5 to 9 focused on the war; and postwar upākhyānas, all of
which are found in Books 12–14.8 Since I will come to focus mainly on the first
two sets in Books 1 and 3, I will work back to them after offering a few words
about the last two.
Almost all the postwar upākhyānas are narrated either at or in connection
with Kurukṣetra, but only two of them mention it. One of them is the
Sudarśana-Upākhyāna (Mahābhārata 13.2.39c), to be discussed later. The other
is the second so-called Rāma-Upākhyāna (12.48–49), about Rāma Jāmadagnya,
which Kṛṣna tells to Yudhiṣṭhira en route to Bhīṣma’s bed of arrows. Kṛṣṇa
directs Yudhiṣṭhira’s attention to the five Lakes of Rāma, which prompt
Yudhiṣṭhira to ask him about Jāmadagnya’s killing of all the kṣatriyas twenty-
one times over and creating the five lakes at Samantapañcaka from their blood
(8). This upākhyāna thus tells about one of the still extant and most famous
tīrthas at Kurukṣetra.9
Nonetheless, once Bhīṣma lauches his four anthologies of dharma teach-
ings, one might maintain that even where Kurukṣetra is not specifically
mentioned, all the upākhyānas he tells in Books 12 and 13 reflect a Kurukṣetra
perspective, since that is where he speaks from on his bed of arrows. One
suggestive illustration will suffice. As Adam Bowles shows, nuanced differ-
ences can be recognized in the epic’s treatment of the dasyu: its term for a
loosely taxonomized and far-flung population interchangeable at times
8 I first worked from this fourfold division in Hiltebeitel, Dharma: Its Early History in Law,
Religion, and Narrative, 426.
9 See James L. Fitzgerald “The Rama Jāmadagnya ‘Thread’ of the Mahābhārata: A New Sur-
vey of Rāma Jāmadagnya in the Pune Text,” in Stages and Transitions: Temporal and His-
torical Frameworks in Epic and Purāṇic Literature, Proceedings of the Second Dubrovnik
International Conference on the Sanskrit Epics and Purāṇas, Aug. 1999, ed. Mary Brocking-
ton (Zagreb: Croatian Academy of Sciences and Arts, 2002), 96, 104, 118–19; Alf Hiltebeitel,
“Tīrthas, Temples, Āśramas and Royal Courts: Towards an Ethnography of Mahābhārata
Bhakti,” in The Archaeology of Bhakti II: Royal Bhakti, Local Bhakti, ed. Emmanuel Francis
and Charlotte Schmid, 81–126. Pondicherry: IFP & EFEO, 2015), 81–126.
The Geography of the Mahābhārata’s Upākhyānas 431
with others from untouchables and śūdras to tribals and mlecchas.10 Yet it
is also uses the term more particularly for those in border areas within the
orbit of the Kuru realm who pose challenges and opportunities for a Kuru
king, as the newly enthroned Yudhiṣṭhira learns from Bhīṣma mainly in the
Āpaddharmaparvan, which contains Bhīṣma’s advice on how to rule in times
of distress.11 A distinction between mleccha and dasyu emerges in the last unit
of the Āpaddharma: an upākhyāna that James Fitzgerald translates under the
title “The Story of the Ungrateful Brahmin” (Mahābhārata 12.162–67).12 The tale
is set among mlecchas of the north (162.28), but it uses mleccha only that one
time to describe this setting, which, as just noted, can be presumed to mean
north of Kurukṣetra, where Bhīṣma is speaking.13 It provides the only usage
of mleccha in the Āpaddharmaparvan. For the characters in question and
their village, the term is dasyu, used fourteen times among thirty-two usages
10 See Aloka Parasher-Sen, “Naming and Social Exclusion: The Outcast and the Outsider,” in
Between the Empires: Society in India 300 bce to 400 ce, ed. Patrick Olivelle (New York:
Oxford University Press, 2006), , 415–55 on different textual processes and principles
behind the fluid naming of “outcastes” working within caste society and “outsiders”
deemed beyond it, including all the groups named here and many more, most of them
mentioned on both sides of this “binary.”
11 See Adam Bowles, “The Dasyu in the Mahābhārata,” in The Churning of the Epics and
Purāṇas at the 15th World Sanskrit Conference, ed. Simon Brodbeck, Alf Hiltebeitel, Adam
Bowles (New Delhi: Rashtriya Sanskrit Sansthan and D.K. Printworld, in press), 155–72 on
differentiations among dasyus and their own differentiation from “tribals” in areas around
Kuru lands and the upper doab, and on the Āpaddharmaparvan’s special pertinence to
these formulations.
12 James L. Fitzgerald, trans., The Mahābhārata, vol. 7. 11. The Book of Women. 12. The Book of
Peace, Part 1 (Chicago: University of Chicago Press, 2004), 590–602. Cf. Bowles, “The dasyu
in the Mahābhārata.” This discussion also appears in Hiltebeitel, “The Mahābhārata and
the Stories Some People Tell about Its Tribal and Earliest Histories” and reflects my review
in Alf Hiltebeitel, “On Reading Fitzgerald’s Vyāsa,” Journal of the American Oriental Society
125, no. 2 (2005): 241–61 of The Mahābhārata, vol. 7. 11. The Book of Women. 12. The Book of
Peace, Part 1, trans. James L. Fitzgerald (Chicago: University of Chicago Press, 2004), and
my reservations about his translating dasyu as “barbarian.”
13 The Mahābhārata also associates the north with mlecchas; see Mahābhārata 12.200.38–43,
as cited in Hiltebeitel Dharma: Its Early History in Law, Religion, and Narrative, 264–65, on
Greeks (Yauna) along with Kāmbojas (those in “an Iranian area in eastern Afghanistan
that spoke late Avestan”; Michael Witzel, “Brahmanical Reactions to Foreign Influences and
to Social and Religious Change,” in Between the Empires: Society in India 300 bce to 400 ce,
ed. Patrick Olivelle [Oxford: Oxford University Press, 2006], 461]), Gāndhāras, Kirātas, and
Barbaras as “evil-doers” who dwell “without supervision” (niradhiyakṣān) to the north and
did not exist on earth in the Kṛta yuga, but emerged and spread in the Tretā yuga and
contributed to “the terrible twilight time at the end of that yuga.”
432 Hiltebeitel
14 See Mahābhārata 12.131.10–17 on drawing dasyus into the army; 12.133 on a good dasyu,
ruler of the Niṣādas, who are among the prototypical dasyu “tribals.”
15 Fitzgerald has rather loosely “The Conversation between the Seer Viśvāmitra and a
Caṇḍāla in the Barbarian’s Hamlet.” The Mahābhārata, vol. 7, 498.
16 My translation; cf. The Mahābhārata, vol. 7, 541: “barbarian law.”
17 The Mahābhārata, vol. 7, 541; Mahābhārata 140.1–2.
18 The Mahābhārata, vol. 7, 505 and 759; Mahābhārata 12.130.1–2; cf. 139.1–6.
19 See Alf Hiltebeitel, “From Ṛṣidharma to Vānaprastha: the Southern Recension Makeover
of the Mahābhārata’s Umā-Maheśvara Saṃvāda,” in The Churning of the Epics and Purāṇas
at the 15th World Sanskrit Conference, ed. Simon Brodbeck, Alf Hiltebeitel, Adam Bowles
(New Delhi: Rashtriya Sanskrit Sansthan and D.K. Printworld, in press), 14–45; “The
The Geography of the Mahābhārata’s Upākhyānas 433
Similarly, of the eight upākhyānas told in preparation for the war or during
it, all could be said to concentrate on Kurukṣetra, even if only two of them
mention it: the Ambā-Upākhyāna, in which Rāma Jāmadagnya and Bhīṣma
exhaust themselves fighting there over Ambā,20 and the Vṛddha-Kumārī-
Upākhyāna or “Old Maid’s Subtale,” also to be mentioned later. The latter occurs
at Mahābhārata 9.51 and is immediately followed from 9.51.25 through 9.52 by
an account of Kurukṣetra itself, most typically titled the Kurukṣetrapraśaṃsā,
in which the ṛṣis tell Baladeva the epic’s chief story about King Kuru: his plow-
ing of this field and getting Indra’s promise that the most stringent ascetics
and warriors who die in battle at “Kuru’s Field” will go directly to heaven puri-
fied by the wind and its sacred dust.21 Kuru, unlike the prior eponyms Pūru
and Bharata of the Paurava-Bhārata-Kaurava lineage, seems to be an inven-
tion of the Mahābhārata poets—like his mother Tapatī, daughter of the sun,
and his father Saṃvaraṇa.22 Indeed, the two tales the Mahābhārata tells about
King Kuru—his birth story, the Tapatī-Upākhyāna (1.160–63), which I will soon
mention briefly, and his plowing of Kurukṣetra—seem especially dreamlike
and elliptical, which, following Sukthankar,23 I take to be a characteristic of
the Mahābhārata’s baseline text reconstructed in the Poona Critical Edition.
Mahābhārata and the Stories Some People Tell about Its Tribal and Earliest Histories”;
Non-Violence in the Mahābhārata: Śiva’s Summa on Ṛṣidharma and the Gleaners of
Kurukṣetra (London: Routledge, 2016). “Tīrthas, Temples, Āśramas and Royal Courts.”
20 One finds seven mentions of Kurukṣetra from Mahābhārata 5.177.23c–179.17.
21 Hiltebeitel, Dharma: Its Early History in Law, Religion, and Narrative, 555; “The
Mahābhārata and the Stories Some People Tell about Its Tribal and Earliest Histories”;
Non-Violence in the Mahābhārata.
22 See Arthur Anthony Macdonell and Arthur Berriedale Keith, Vedic Index of Names and
Subjects, vol. 2 (Delhi: Motilal Banarsidass, 1967 [1912]), 413: “Saṃvaraṇa is the name of a
Ṛṣi mentioned in one passage of the Rigveda” (5.33.10). Cf. Alexander Lubotsky, A Ṛgvedic
Word Concordance, part 2 (New Haven, CT: American Oriental Society, 1997) 1454: only
usage as a personal name.
23 “The Southern recension impresses us thus by its precision, schematization, and thor-
oughly practical outlook. Compared with it, the Northern recension is distinctly vague,
unsystematic, sometimes even inconsequent, more like a story rather naively narrated, as
we find in actual experience.” V.S. Sukthankar, “Prolegomena,” in The Ādiaprvan for the
First Time Critically Edited (Poona: Bhandarkar Oriental Research Institute, 1933), xxxvi
(author’s italics). See further Hiltebeitel, “From Ṛṣidharma to Vānaprastha” and Non-
Violence in the Mahābhārata.
434 Hiltebeitel
Moving back now to Book 3, it is perhaps well enough known that the Āraṇya
kaparvan builds most (eighteen, to be precise) of its twenty-one upākhyānas
into two main sequences in which in each case the narrations are mostly by
one ṛṣi speaker: in the first series, Lomaśa; in the second, Mārkaṇḍeya. We will
see that our picture of upākhyāna geography widens with these two series.
After Nārada has recounted what Pulastya once told Bhīṣma about the
merits of visiting tīrthas, including the epic’s most detailed description of
Kurukṣetra at Mahābhārata 3.81,24 he introduces Lomaśa to serve as the
Pāṇḍavas’ tīrtha guide, and Lomaśa then tells eight out of nine more-or-less
location-specific upākhyānas, all narrated in the Tīrthayātra-Parvan (3.80–153),
which the Pāṇḍavas (minus Arjuna) and Draupadī hear as they make a clock-
wise pradakṣinā of the subcontinent with Lomaśa. (Arjuna has gone to Kailāsa
and Indraloka to meet Śiva and Indra to obtain divine weapons, and one of the
purposes of the four other Pāṇḍavas and Draupadī’s journey is to bide the time
until they can meet him in the Himalayas when he is ready to come down to
earth.)
Lomaśa begins this sequence with the Agastya-Upākhyāna (3.94–108),
which, while ranging widely,25 begins at Agastya’s hermitage on the Bhāgirathī
(97.26) at the Gate-of-the-Ganges (95.11a), presumably Hardwar. And Lomaśa’s
upākhyāna-telling along the tour ends, after one interruption by Akṛtavraṇa
to narrate the Kārtavīrya-Upākhyāna at Mount Mahendra in Kaliṅga, with the
Yavakrīta-Upākhyāna (3.135–39), which Lomaśa tells higher up the Gaṅgā at the
hermitage of Raibhya,26 where Raibhya’s Veda-knowledge proved to be supe-
rior to Yavakrīta’s.27 From there, Lomaśa leads the Pāṇḍavas further upriver to
24 Thirteen of Book 3’s seventeen mentions of Kurukṣetra occur in this adhyāya. See Hiltebei-
tel, “Tīrthas, Temples, Āśramas and Royal Courts” on “Pulastya’s tīrthayātra” as a “matrix
text” for other Mahābhārata pilgrimage texts in that it is narrated to Bhīṣma before the life
of Kṛṣṇa changes the landscape.
25 Other locations mentioned include Maṇimatī (where Ilvala and Vatāpi live), Vidarbha
(where Agastya gets Lopamudrā from), and of course the Vindhya, which he keeps from
growing as high as Mount Meru.
26 En route, they pass the River Madhuvilā Samaṃgā, “the bathing place of Bharata, … called
Kardamila,” where Indra also bathed after killing Vṛtra, and where Mount Maināka
entered the earth (Mahābhārata 3.135.1–3).
27 Raibhya and Yavakrīta’s father Bharadvāja were friends. Yavakrīta disliked Raibhya’s better
reception by brahmans, and so sought to learn Veda on the Bhagīrathī, where Indra in
brahman guise tried to make a dam with scoops of sand to teach him that the two
tasks were equally fruitless (Mahābhārata 135.33). Indra then promised Yavakrīta and
The Geography of the Mahābhārata’s Upākhyānas 435
a point where the seven-fold Gaṅgā shines forth. There he warns them that
they should guard Draupadī, as they are nearing Kubera’s domain with its
Yakṣas and Rākṣasas (3.140.1–15), and also nearing Mount Kailāsa. This skein of
upākhyānas is now over as they all approach Mount Gandhamādana (141.22).28
It is in the ensuing Gandhamādana narrative that they will be reunited with
Arjuna.
Well along this route, Lomaśa tells the Māndhātṛ- (Mahābhārata 3.126) and
Jantu- (3.127–28) upākhyānas at Kurukṣetra. Māndhātṛ, an Ikṣvāku, got his
name when he suckled Indra’s thumb at birth after his hundred-year gestation
inside his father, who had drunk a potion intended for Māndhātṛ’s mother.
Yudhiṣṭhira now stands at “this most holy spot in the middle of Kurukṣetra”
(puṇyatame deśe kurukṣetrasya madhyataḥ; 3.126.42cd), where Māndhātṛ
offered sacrifices to the gods. Jantu, then, is a brahman who suffered for a
time in hell for officiating at an adharmic sacrifice—that of King Somaka,
who sacrificed one son to obtain a hundred—somewhere on this same ter-
rain.29 Somaka would seem to be the eponym of the Somaka branch of the
Pañcālas, who, like the hundred Kauravas, are destined to be wiped out on
the Kurukṣetra battlefield. Next, Lomaśa speaks on the merits of spending the
night at “the gate of Kurukṣetra” (3.129.11), and then directs the Pāṇḍavas to
Plakṣāvataraṇa on the Yamunā, which is “the gate to the ridge of heaven.” As
one learns during Balarāma’s tīrthayātra, Plakṣāvataraṇa is also near the head-
waters of the Sarasvatī (9.53.10). The Pāṇḍavas do yātsattra or Vrātya things
there, carrying “poles and mortars” and offering up Sāsasvata sacrifices.30 This
is “where King Bharata frequently sent off his black-dappled sacrificial horse,
after he had obtained the whole earth by dharma”; here Yudhiṣṭhira can “gaze
upon the worlds and see Arjuna off on above” (3.129.13–19). They are still near
Bharadvāja full sight of the Vedas, but Bharadvāja saw this as a source of pride for
Yavakrīta that would be his death. Yavakrīta then rapes Raibhya’s son Parāvasu’s wife, and
dies by a sorceress. Bharadvāja cremates Yavakīta and dies in the fire, cursing Raibhya.
Raibhya’s sacrificial patron King Bṛhaddyumna then holds a session with Raibhya’s two
sons as priests. Coming home at night, Parāvasu kills Raibhya in the forest, mistaking him
for an animal, and tells his brother Arvāvasu he must finish the rite. After further turn-
about, the gods favor Arvāvasu and explain to Yavakrīta why his Veda knowledge was infe-
rior to Rabhya’s: he learned it the easy way and not from a guru.
28 Heading there, Subāhu, king of the mountain men (Kuṇindas), hosts them (Mahābhārata
141.26–28).
29 See Mahābhārata 3.128.1 ff.: a site for sacrifices by Prajāpati, Ambarīṣa, Yayāti, etc.; 128.17.
30 On Plakṣāvataraṇa/Plakṣa Prāśravaṇa tīrtha, see Hiltebeitel, Rethinking the Mahābhārata,
143–55; on yātsattras and Vrātyas along Baladrāma’s and others’ routes, ibid., 121–55, 159–
60, 166, 170–71.
436 Hiltebeitel
Kurukṣetra, for next day on the Sarasvatī, Lomaśa tells them, “This is Prajāpati’s
altar, five leagues around, the field of that great-spirited sacrificer Kuru itself”
(22–23).
Book 3’s second main upākhyāna sequence is less topographical than
Lomaśa’s. It occurs when all five Pāṇḍavas and Draupadī have come down
from Gandhamādana by a well-detailed route to the Kāmyaka Forest. There,
while they are receiving a visit by Kṛṣṇa and Satyabhāmā, Mārkaṇḍeya arrives
(Mahābhārata 180.39). Kṛṣṇa—with the Pāṇḍavas and the brahmans’ permis-
sion—asks Mārkaṇḍeya to tell all of them “holy tales of the past” (180.42–44).
Nārada then drops in close behind Mārkaṇḍeya (45-47), knowing that the age-
less sage would speak, and encourages him to begin. As he is about to do so,
Yudhiṣṭhira celebrates that Keśava is there to listen (181.1–3), and Mārkaṇḍeya
tells Yudhiṣṭhira that the Pāṇḍavas have come to earth from the world beyond
for the sake of surakārya (39), “the work of the gods.”31 All this is stagesetting
for the Mārkaṇḍeya-Sāmasya-Parvan (3.179–221), in which Mārkaṇḍeya now
tells a brief opening story—not billed as an upākhyāna—about how brahmans
are immune from death (3.282). This prompts Yudhiṣṭhira to give Mārkaṇḍeya
the topic of the greatness of brahmans (3.183.1), on which Mārkaṇḍeya tells
the Vainya-Upākhyāna (3.183) as illustration, 32 and next how Sarasvatī praises
brahmans (3.184), before he seemingly diverges from this topic to tell the
Matsya-Upākhyāna (3.185) in which Brahmā, not Viṣṇu, becomes the Fish that
saves Manu, getting him to build an ark. This introduces four adhyāyas that
are surely the centerpiece of this upaparvan, which might itself explain why
none of the stories and predictions Mārkaṇḍeya tells in this span are called
upākhyānas. These are his revelations (1) about kalpas and caturyugas; (2)
about swimming during a pralaya into the baby Kṛṣṇa’s mouth, after which
he tells the Pāṇḍavas and Draupadī they should take refuge in the same Kṛṣṇa
31 This is one of many points at which one hears about the “work of the gods” and/or the
Pāṇḍāvas’ or others’ divine parentage. Like most of these, it is not mentioned in short-lists
of such revelations designed to demonstrate the alleged peripheralty of these themes to
the epic. See, e.g., Simon Brodbeck, “Husbands of the Earth: Kṣatriyas, Females, and
Female Kṣatriyas in the Strīparvan of the Mahābhārata,” in Epic Undertakings: Papers of
the 12th World Sanskrit Conference, vol. 2, ed. Robert P. Goldman and Muneo Tokunaga
(Delhi: Motilal Banarsidass, 2009), 37–38 and n. 17, and, for comment, Hiltebeitel, Dharma:
Its Early History in Law, Religion, and Narrative, 573.
32 Atri and Gautama vie for the patronage of King Vainya (not a king of any place men-
tioned), and get into an argument over whether Vainya (Atri’s position) or Indra (Gauta-
ma’s position) is the “Provider.” The arbiter selected is Sanatkumāra who favors Atri’s view
and thus also Vainya, who gives Atri an abundance of prizes, “for I think you are omni-
scient!” (32).
The Geography of the Mahābhārata’s Upākhyānas 437
seated among them; (3) about the future adversities of the Kaliyuga; and (4)
about how the brahman Kalki Viṣṇuyaśas will restore the Kṛtayuga. Indeed,
the identification of the Fish as Brahmā in an upākhyāna just before this non-
upākhyāna “bhakti run” suggests the possibility that Mārkaṇḍeya is treating
the Fish story as transitional to his more important revelations about Kṛṣṇa
and Kalki. Mārkaṇdeya has in any case clearly built matters up to a lesson on
cosmography rather than geography, of which there is only minimal interest
in these stories, upākhyānas and otherwise. Thus even though Yudhiṣṭhira
launches Mārkaṇḍeya’s next tale with what looks like a reminder that what he
wants to hear about is still “the lordliness of Brahmans” (190.1), Mārkaṇḍeya
resumes telling upākhyānas that carry his cosmological interest along through
the next three of them—the Maṇḍūka- (3.190);33 Indradyumna- (3.191);34 and
Dhundhumāra-Upākhyānas (3.192–95)35—before ending the sequence with
the more down-to-earth and whimsically ethical Pativratā-Upākhyāna (3.196–
206), most of which takes place in Mithilā.
Mārkaṇḍeya’s string of six upākhyānas is interrupted now by Vyāsa,
who drops in on the Pāṇḍavas at Kāmyaka Forest to tell them the Mudgala-
Upākhyāna (Mahābhārata 3.246–47), about which I will say a bit in closing. The
Pāṇḍavas will then ask Mārkaṇḍeya to tell them two more famous upākhyānas
before Book 3 is over: the Rāma-Upākhyāna, the one about Rāma Dāśarathi
(3.257–76), and the Sāvitrī-Upākhyāna (3.277–83). Speaking now strictly in
terms of upākhyāna sequentiality, one will notice that just as Lomaśa’s string
33 The Ikṣvāku king Parikṣit marries a frog-princess whom he meets in a pool where he goes
to bathe in the woods. He must never show her water. But she finds water and resumes
being a frog. The king orders the killing of all frogs until he gets her back from the frog
king Āyu, who gives back his daughter Suśobhanā, who “has this bad streak: many are the
kings she has deceived before.” In giving her to this Parikṣit, Āyu curses her to have
unbrahmanic sons for having been deceptive (38–42). The eldest son then won’t give
Vāmadeva back his Vāmya horses, and this brahman shows his superiority through sev-
eral episodes. With this, if not before, the “greatness of brahmans” theme seems to have
petered out.
34 King Indradyumna was bounced from heaven because no one knew him there; his “fame”
was lost. He approached Mārkaṇḍeya to see if he remembered him and was told, “We are
not alchemists [rāsāyanikās].” They finally find a turtle, Akūpāra, in a Himalayan lake who
remembers him. Indradumna created the lake by the stamping of the cows he gave away
there in his fire sacrifices.
35 Dhundhu, son of the asuras Madhu and Kaiṭabha, bothered Utaṅka by making earth-
quakes in the desert. The Ikṣvāku king Kuvalāśva, pervaded by the force of Viṣṇu, killed
Dhundhu, which gave him the name Dhundhumāra. Viṣṇu had killed Madhu and
Kaiṭabha on his uncovered thigh after Brahmā was born from Viṣṇu’s navel. An Ikṣvāku
genealogy is given down through Kuvalāśva and his remaining three sons—his other
21,000 sons recalling the 60,000 sons of Sagara killed by Kapila (Mahābhārata 195.26).
438 Hiltebeitel
Damayantī then turns north and walks for three days and nights before coming
upon a “circle of hermitages” (93d) “looking like a heavenly park” (56-57), pop-
ulated by ṛṣi-muni ascetics “the likes of Vasiṣṭha, Bhṛgu, and Atri,” who “lived
on water or off the wind, or fed on leaves” (58–59). Damayantī’s ordering of the
two words śata and śṛṅga when she calls the mountain śṛṅga-śatair reverses
and evokes the Himalayan Śataśṛṅga Mountain of Book 1, where a mix of
earthly and celestial ṛṣis protected Pāṇḍu, Kuntī, and Madrī—but especially
the two women, one of whom, Kuntī, holds the secret to the birth of sons for
the impotent Pāṇḍu. The solicitous Śataśṛṅga Ṛṣis, clearly attuned to the epic’s
divine plan, then soon escort the Pāṇḍavas, Kuntī, and the remains of Pāṇḍu
and Mādrī to Hāstinapura to introduce the boys to the Kurus there as the sons
of Pāṇḍu, before vanishing on the spot.39 Similarly, when the Ṛṣis in the Nala-
Upākhyāna now heed Damayantī’s call to the hundred-peaked mountain, they
arrange for her to find them so that they can reassure her that she will find
Nala—only then to vanish, along with their hermitages, leaving her wondering
whether she had seen them and their hermitages only in a dream (93).
What the poets have done here, then, is to build the main story’s Śataśrṅga
cosmography from the Ādiparvan into the Nala-Upākhyāna. The Mahābhārata’s
main story does this kind of ṛṣi-fusing at moments when earthly and celes-
tial ṛṣis sidle their way together into its divine plan. Another such instance
that involves celestial ṛṣis heading to the Kuru court occurs when a host of
them—including Rāma Jāmadagnya, who must be taking leave from Mount
Mahendra—descends from the sky to greet Kṛṣṇa on his pre-war embassy to
Hāstinapura (Mahābhārata 5.81.26–64). I will mention this trope of ṛṣi-fusing
again in connection with the Śakuntalā-Upākhyāna, the first Ādiparvan sub-
tale (according to the Critical Edition).
39 See for discussion Hiltebeitel, Dharma: Its Early History in Law, Religion, and Narrative,
393-409; “The Mahābhārata and the Stories Some People Tell about Its Tribal and Earliest
Histories.”
40 See Simon Brodbeck, The Mahābhārata Patriline: Gender, Culture, and the Royal Heredi-
tary (Aldershot: Ashgate Publishing, 2009), 28–29. Each of the two upākhyānas is further
the lead-in to one of the epic’s two dynastic genealogies: the śloka vaṃśa coming after
“Śakuntalā,” and the prose vaṃśa after “Yayāti” and the Uttara-Yāyāta.
The Geography of the Mahābhārata’s Upākhyānas 441
dynasty’s third eponym, Kuru, already mentioned, whom the Ādiparvan also
introduces in the Tapatī-Upākhyāna (1.160–63), which is named after Kuru’s
mother Tapatī. But the landscape and time-frame of the Tapatī-Upākhyāna is
best discussed not in the context of dynastic geography or history but in rela-
tion to the two intermingled upākhyānas about ṛṣis that immediately follow it:
the Vasiṣṭha- (1.164–68.173) and Aurva-Upākhyānas (1.169–72). These five, plus
the aforementioned Sunda-Upasunda-Upākhyāna (1.201–4) will be the main
Ādiparvan subtales under scrutiny from here on in this essay. The six will be
taken up in the order just given, which is the order in which they appear in the
Critical Edition. Nonetheless, it is worth keeping in mind that the full eleven
upākhyānas of Book 1 could be seen to convey a certain tightening of cosmo-
graphic, geographic, and especially itihāsic momentum from first to last.41 For
whereas the first two deal with kings who establish the Kuruvaṃśa’s rule over
the entire world, the last one coincides with the burning of the Khāṇḍava
Forest that clears the land for a new branch of that dynasty, the Pāṇḍavas, to
found their new capital at Indraprastha, traditionally identified with Delhi.42
With regard to the Śakuntalā-Upākhyāna’s geo-history, I must call attention
to some modifications and fine-tunings made quite early by the redactors of
the Southern Recension, which I will often refer to as elsewhere by S. I have
written an article on S’s “first reading” of Śakuntalā that touches on the Yayāti-
Upākhyāna as well.43 The Śakuntalā-Upākhyāna begins with Janamejaya saying
that now that he has heard about how the Gods, Dānavas, Rākṣasas, Gandharvas,
and Apsarases took on their aṃśāvataraṇam, he wants Vaiśaṁpāyana to tell
41 Of the other four upākhyānas not mentioned in this paragraph, Mahābhiṣa- (Mahābhārata
1.91) anticipates Gaṅgā’s down-to-earth appearances and the geography they entail up and
down the river, including her decision to enter history by loving the Bhārata/Kuru line
(1.92.12c–13; 93.45–46; see Hiltebeitel, Dharma: Its Early History in Law, Religion, and Narra-
tive, 352); Aṇimāṇḍāvya- (1.101) follows after the dynastic account has reached the birth of
Vidura, but mentions no particular locations when an unnamed king’s men chase a bunch
of thieves into an ashram (101.9); Vyuṣitāśva- (1.112) tells of an unregistered lunar dynasty
king; and Pañcendra- (1.189.1–34), by which Vyāsa explains why Draupadī should marry all
five Pāṇḍavas, begins in Naimiṣa Forest (1a), but would seem to shift at verse 9 to somewhere
on the Gaṅgā where the sacrificing gods see golden lotuses floating by, which Indra traces to
their upriver source where it is possible to see Śiva playing dice on a Himalayan peak (14).
Note that the ṛṣis who come at the bidding of the impaled silent Muni in Aṇīmāṇḍavya-
return at night in the form of śakuna birds (101.13-15) to ask about his predicament—mark-
ing, as in Śakuntalā-, the interplay between geosphere and cosmosphere. But in Śakuntalā-,
the birds are śakunta birds and are not ṛṣis.
42 See Alf Hiltebeitel, “Among Friends: Marriage, Women, and Some Little Birds,” in Gender
and Narrative in the Mahābhārata, ed. Simon Brodbeck and Brian Black (London: Rout-
ledge, 2007), 117–26.
43 Hiltebeitel, “The Southern Recension’s Śakuntalā as a First Reading.”
442 Hiltebeitel
him “from the beginning [āditaḥ] … how the Kuru vaṃśa came into being”
(Mahābhārata 1.62.2). Vaiśaṁpāyana then introduces Duḥṣanta44 as a Paurava
vaṃśakāra or “lineage maker” who was a “herdsman of the earth to her four
horizons” (pṛthivyāś caturantāyā goptā). He enjoyed not only “the entire earth
with all four quarters” (caturbhāgaṃ bhuvaḥ kṛtsnam sa bhuṅkte) but “also the
countries that are surrounded by the ocean” (samudrāvaraṇāṃś cāpi deśān),
and his pleasure extended as a “scourge of his enemies as far as all forest tribes
and barbarians” (ā mlecchāṭavikān sarvān … ripumardanaḥ) (62.3c–5b).45 Note
that Janamejaya, whom Vaiśaṁpāyana immediately calls Bharatasattama or
best of Bharatas (62.3d), asks about the Kuruvaṃśa and is taken back to the
earlier Paurava line that reaches Duḥṣanta, who, it seems, is the one credited
with being the vaṃśakāra within this one line. Clearly, just as the poets find
ways to suggest the fusion of celestial and earthly ṛṣis, they find ways to estab-
lish continuity within a single Paurava-Bharata-Kuru dynastic line.
Duḥṣanta goes hunting. S’s first five interpolations are leisurely, yet sug-
gest the Southern Recension’s game plan for “Śakuntalā” in toto. After leaving
Duḥṣanta’s brutal hunt intact through a first forest, S’s first insertion comes
as Duḥṣanta approaches a second forest. This insertion and the second intro-
duce the idea that this transitional second forest, and the third where he
finds Kaṇva’s hermitage, are resorted to by Siddhas, Cāraṇas, Apsaras, and
Gandharvas. The third Southern Recension insert then says of Kaṇva’s āśrama
that it is “frequented by hosts of Maharṣis” (64.22/681*). A fourth insertion then
accounts for practitioners of all four Vedas there. And the fifth describes its
resident brahmans as specializing in all kinds of Vedic sciences.46 The middle
interpolation about Maharṣis thus erases the distinction between the celestial
hosts mentioned in the first two inserts, and the worldly brahmans in the last
two. Just as it did on the Śataśṛṅga Mountain, this signals that both types are
found in Kaṇva’s hermitage.
44 I use the Critical Edition’s spelling Duḥṣanta, though as I indicate in Hiltebeitel, “The
Southern Recension’s Śakuntalā as a First Reading,” 18, n. 2. I regard it as an emendatation
of a more likely Duṣyanta.
45 On the compound mlecchāṭavikān, used only one other time in the Poona Critical Edition
in connection with the imminent fall of Duryodhana, the “last Kaurava,” see Hiltebeitel
forthcoming C § A. Since according to the Bhṛgu-Bharadvāja Saṃvāda (see n. 8 above)
mlecchas or barbarians do not appear until the Tretā yuga, this might strain the idea that
the idyll would imply a Kṛta yuga (on which see Madeleine Biardeau, “Śakuntalā dans
l’épopée,” Indologica Taurinensia 7, Dr. Ludwig Sternbach Felicitation Volume, Part I [1979]:
115–25), though it could just be that Duṣyanta is a king who makes the yuga a Kṛta.
46 See Sukthankar, “Prolegomena,” xxxviii.
The Geography of the Mahābhārata’s Upākhyānas 443
S thus brings out that Kaṇva is an exemplar of both types of ṛṣis, as of course
he already is in the baseline story. That is, he is a ṛṣi suited to make interven-
tions on behalf of dharma and a divine plan. Indeed, we have just noted two
episodes that involve a mix of celestial and earthly ṛṣis heading to Hāstinapura:
one, to escort the Pāṇḍavas and Kuntī there to meet the astonished Kauravas,
the first among them being Dhṛtarāṣṭra, Pāṇḍu’s brother and for now the
reigning king; the other to join Kṛṣṇa on his pre-war embassy there. The most
striking parallel and the one proximate to “Śakuntalā” in the Ādiparvan is the
first, for Śakuntalā too will have an escort to Hāstinapura, which Kaṇva him-
self will arrange, and Śakuntalā has to introduce Bharata at Hāstinapura to
another equally caught-off-guard “Kuru” king, her husband-in-denial Duḥ
ṣanta. But Kṛṣṇa’s embassy also presents parallels, for in this episode it will
also be a question of ṛṣis who will bear witness to what goes on in the Kuru
court. Indeed, by the time of the Mahābhārata’s main story, Kaṇva, like Rāma
Jāmadagnya, has clearly become one of the ṛṣis who is cued into the epic’s
divine plan. He is most likely47 among those who descend from the skies to
see Kṛṣṇa as he approaches Hāstinapura as the Pāṇḍavas’ last “peace” envoy,
who then descend again to take seats in the Kuru court to hear and witness
what Kṛṣṇa will say and do there (Mahābhārata 5.92.40–45). In any case, he
is there with Rāma Jāmadagnya, and as soon as Jāmadagnya tells Duryodhana
the Dhambhodbhava Upākhyāna (5.94), Kaṇva launches two tales of his own.
First, he tells Duryodhana about Indra’s charioteer Mātali—an “ancient his-
tory” (itihāsam-purātanam; 5.95.11b) rather than an upākhyāna. This is about
Mātali’s search of various earthly and subterranean worlds including finally
Bhogāvatī, the world of Snakes, for a son-in-law to marry to his daughter
(5.95–103). Kaṇva prefaces this story by mentioning the indestructibleness of
Brahmā and of Nara and Nārāyaṇa (95.1–3). And he ends on the point that, just
as Garuḍa’s pride was humbled by Viṣṇu, so will Duryodhana’s be by Vāsudeva,
whom Duryodhana should rather recognize as the one he should turn to as the
“savior” of his lineage (vāsudevena tīrthena kulaṃ rakṣitumarhasi, 103.34cd)—
which only provokes Duryodhana to derisive laughter (36). Undeterred, Kaṇva
then tells his story of Gālava (5.104–21)—likewise called an itihāsam-purātanam
(5.104.7ab) but also a “great narrative” (mahad-ākhyānam, 121.22ab)—that
further features Garuḍa, now taking Gāḷava on tour over the four directions,
before he brings King Yayāti back into focus from Book 1, along with novel-
ties about Yayāti’s daughter Mādhavī. As the Śakuntalā-Upākhyāna appears
47 He is not mentioned in the short list of such ṛṣis who appear at first along with Rāma
Jāmadagnya (Mahābhārata 5.81.26–28), but it is clearly a question of these and countless
others; see Hiltebeitel, Dharma: Its Early History in Law, Religion, and Narrative, 620–21.
444 Hiltebeitel
within the overarching momentum of the aṃśāvataraṇa, one could think that
Kaṇva’s protection of Śakuntalā in it is already part of a divine plan.48
But for now, as Duḥṣanta comes to the third forest, he finds that “a sovereign
beauty reigned over it” (lakṣmyā paramayā yutam; Mahābhārata 1.64.5).49 He
has arrived at Kaṇva’s hermitage on the banks of the Mālinī River, which “added
to the loveliness of this forest of austerities” (64.18).50 “Thereupon the illustri-
ous warrior drew nigh to the [śrī-posssessing]51 hermitage that … was the
image of the world of the Gods [devaloka]” (64.19). This verse describes how
Vaiśaṁpāyana says the hermitage looked to Duḥṣanta, and his starting with its
being like devaloka probably goes with its being śrīmān. But as Biardeau
astutely says, “Duṣyanta first thinks he is in Devaloka …, but it is rather a
Brahmaloka …, for the place resonates with Veda recitation and all preoccupy
themselves with sacrificial tasks.”52 Biardeau offers only a shorthand explana-
tion of why she speaks of Brahmaloka here. But let us consider a more
cosmologically attuned account of that place by André Couture as he com-
ments on chapters 39 and 40 of the Critical Edition of the Harivaṃśa. Couture’s
debts to Biardeau as her former student and her continuing reader are acknowl-
edged and profound here, and his work on the Harivaṃśa is important in
working out that text’s relations with the Mahābhārata.
The “Viṣṇuite bhakti” of the Harivaṃśa, says Couture,
48 See Hiltebeitel, “The Southern Recension’s Śakuntalā as a First Reading,” on one of S’s few
insertions into the story of Śakuntalā’s birth that she tells Duḥṣanta, quoting what her
mother Menakā said before abandoning her:
“You, luminous one, are the destructress of the foremost tapas of the great Ṛṣi. Therefore
I will go to heaven, having come for the purpose of the work of the Gods” (devakāryārtham;
66.8/603* lines 3-4).
As I say there, this devakārya would probably “include more than just the benefit of one
god like Indra, who was worried about Viśvāmitra’s tapas. We may thus infer that the baby
Menakā addresses will carry on the devakārya by engendering the Bhārata dynasty.”
49 Note how the Buffalo image of a king out of control, and needing a Mahiṣī, now alternates
to an image of his royal complement being Śrī-Lakṣmī.
50 Cf. 3.80.64 (p. 375) on Kaṇva’s hermitage, said there to be near Agastya’s Lake and where
Yayāti fell.
51 See note to 1.64.5. Van Buitenen does not complicate his translation, but śrīmān is signifi-
cant here in describing the āśrama. See also 64.34.
52 Biardeau, “Śakuntalā dans l’épopée,” 116. Biardeau cites as usual the Vulgate; see 64.30, 34,
and 40. Note further S’s attention to the weird munis of Kaṇva’s ashram with their varied
diets at App. 1, No. 47 lines 13-16.
The Geography of the Mahābhārata’s Upākhyānas 445
where all celebrate together the same eternal rites; but not a world of
indifference, because in celebrating the ancient rites prescribed by the
Veda, a Brahmin, in respecting his svadharma, gives homage to Viṣṇu.53
This is not to say that the Harivaṃśa provides an eternal place to do one’s
svadharma only for brahmans, “those who dedicate themselves to asceticism
according to the Vedānta.”54 For when Kṛṣṇa lifts Mount Govardhana to pro-
tect the cowherds, he opens the still higher world of Goloka to all, beginning
with those who pursue the svadharma of cowherds. And still higher than that,
“there remains the ascetic residence reserved for Viṣṇu himself.”55 It seems that
the Harivaṃśa inserts Goloka into what Couture calls this “new cosmology,”56
otherwise seeking to provide a more definitive theological and soteriological
structure to elements it knows from the Mahābhārata, including Kṛṣṇa’s teach-
ings on svadharma in the Bhagavadgītā.57 This is background to what Couture
has to say about Brahmaloka that is most interesting with respect to the way
that Kaṇva’s hermitage might represent it in the Śakuntalā-Upākhyāna:
53 André Couture, trans. L’enfance de Krishna: Traduction des chapitres 30 à 78 (édition cri-
tique) du Harivaṃśa (Paris: Les éditions du Cerf; and Montreal: Presses de l’Université
Laval, 1991), 144 n. 1. Couture draws here on Nīlakaṇṭha and comments that the chapter
(1.49 in the Vulgate) has the purpose of showing “que même les habitants du Brahmaloka
honorent (bhajanti) Viṣṇu. Ils ne vivent pas dans l’indifférence, en s’imaginant que c’est ce
qu’exige leur devoir.”
54 Couture, trans. L’enfance de Krishna, 51.
55 Ibid., citing HV 62.24–32.
56 Couture, trans. L’enfance de Krishna, 51, mentioning that “Il y a peu de texts qui parlent
explicitement de ce fameux Goloka,” but showing precedent for it in the Mahābhārata’s
myth of King Pṛthu (Mahābhārata 12.29.129–36; 12.59), to whose story the Harivaṃśa con-
secrates its fifth and sixth chapters (51–53).
57 Couture presents the Harivaṃśa as having thoroughly absorbed the Gītā’s teachings on
svadharma (Couture, trans. L’enfance de Krishna, 41, 46, 56–57, 67–68, 87 n. 21 [karmay-
oga], 98 n. 21, 407, 427)—something that I believe the Mahābhārata has not done so thor-
oughly with regard to the Gītā’s cosmologically tuned alignment of svadharma with
karmayoga; see Hiltebeitel, Dharma: Its Early History in Law, Religion, and Narrative, 462,
518–34, 536 n. 32.
446 Hiltebeitel
One profit we can draw from this lengthy quotation is the backing it gives to
my proposal that we understand Mārkaṇḍeya’s representation of Manu’s Fish
with Brahmā as interior to a relation of devotion between Beings and Kṛṣṇa-
Janārdana. Another is that its evocation of Brahmaloka as a hidden mysterious
āśrama can clarify the comings and goings of the celestial ṛṣis who fabricate a
parallel forest āśrama in which to appear before Damayantī. But for now, turn-
ing back to Kaṇva’s āśrama, the Nālinī River is described at length embracing it,
culminating in a verse that compares the site to the “dwelling place [sthānam]
of Nara and Nārāyaṇa, adorned by the Gaṅgā,” while the hermitage itself, which
Duḥṣanta knows to be that of Kaṇva-Kāśyapa, whom he has come to see
(Mahābhārata 1.65.8ab), is compared to Citraratha’s park (64.20–25; 65.8). It is,
of course, unlikely that Duḥṣanta has been either to Nara and Nārāyaṇa’s her-
mitage (presumably Badarī) or to Citraratha’s park. He does, however, now
enter Kaṇva’s hermitage alone with such ruminations attributed to him, while
telling his escort to wait behind for him (1.64.26–27; 65.1). He finds the place
humming with Vedic sounds and bustling with Vedic business, which is exten-
sively described at the end of an adhyāya.
Once in the āśrama, Duḥṣanta finds it empty but for the beautiful Śakuntalā.
She tells him her “reverend father” (pitā me bhagavān) Kaṇva is away gathering
fruits (Mahābhārata 65.9ab; 67.6ab). Pressed by Duḥṣanta to confirm his sense
that she cannot be a brahman’s daughter and must be a kṣatriya’s, she tells him
the story—which she once overheard Kaṇva tell a visiting ṛṣi (65.19)—of how
Indra persuaded the Apsaras Menakā to seduce Viśvāmitra, and Śakuntalā
became their daughter, all of which took place in the vicinity of the current
conversation: “on a lovely tableland in the Himalayas [prasthe himavato ramye]
by the river Mālinī” (66.8cd). There on the Mālinī’s bank, Menakā abandoned
Śakuntalā at birth and went back up to the assembly of Indra (9); and there
King, you see the faults of others that are small like mustard seeds and
you look but do not see your own, the size of pumpkins! Menakā is one of
the Thirty Gods. The Thirty come after Menakā! My birth is higher than
59 See, however, the more condensed and perhaps more romantic verse where Śakuntalā
re-summarizes matters for Duhṣanta: “The Apsaras Menakā gave birth to me on a flank of
the Himalaya [himavatah pṛṣṭhe] and abandoned me piteously and went as if I were
another’s child” (Mahābhārata 68.69; cf. 68.73, where Duḥṣanta repeats these words call-
ing Menakā a slut, and then calling Śakuntalā the same).
60 S also has Kaṇva predict to Śakuntalā that when Bharata is born after her three-year preg-
nancy (Mahābhārata 1.68.2; 58ab), he will be a cakravartin—a prediction that Vaiśaṃpāyana,
in the baseline, saves for the end of the story (69.45–47).
61 1.624*, lines 4-7, after 68.2; 1.649* after 68.59ab.
62 See Hiltebeitel, “The Southern Recension’s Śakuntalā as a First Reading.”
448 Hiltebeitel
yours, Duḥṣanta! You walk on earth, great king, but I fly the skies. See how
we differ, like Meru and a mustard seed! I can roam to the palaces of great
Indra, of Kubera, of Yama, of Varuṇa: behold my power, king! (Mahābhārata
1.69.1–4)
Yet although she says she can go to palaces of the four Lokapālas, she remains
in place and closes out her speech on a down-to-earth note: “Even without you,
Duḥṣanta, my son shall reign over this four-cornered earth crowned by the
king of mountains (śailarājāvataṃsakam | caturantām imām urvīm; Mahābhā
rata 1.69.27), to which S has her recall Indra’s prophesy at Bharata’s birth that
he will become a cakravartin. S then gives her these final words: “A witnessless
woman of small fortune, I will go as I have come” (1.625*, l 4; 670*, ll 1 and 4).
She departs, clearly on foot, with S adding that she took Bharata with her, and
in the same interrupted verse, the disembodied heavenly voice bears witness
to Duḥṣanta that she has spoken the truth (69.28 and 671*).
As noted, S now reverses the back-to-back order between “Śakuntalā” and
“Yayāti.” Since putting “Yayāti” before “Śakuntalā” makes the “Vedic” past more
linear, S probably reverses the order from an interest in advancing the epic’s
own genre identification as itihāsa, “history.”63 But N’s baseline could have had
its prior reasons to historically prioritize Bharata. Bharata, unlike his ancestors
Yayāti or Pūru or his descendant Kuru, has from Yajur Veda and Brāhmaṇa texts
on, been linked to a prototypical royal name that has unified Vedic peoples,64
and, from the Mahābhārata on, linked that name with a still larger Vedic land,
Bhāratavarṣa.
Moving on to Yayāti, I will keep mainly to the upākhyāna in his name
(Mahābhārata 1.70–80), and say less about its two sequels: the Uttara-Yāyāta
(1.81–88) that follows it immediately; and Book 5’s Gālava narrative, already
briefly mentioned. Each has geographical interest,65 but it is the Yayāti-Upā
khyāna that is positioned informatively not only within the aṃśāvataraṇa
63 For the Mahābhārata, itihāsa is more specific than itihāsapurāṇa, and the epic’s genre is
not kāvya, on which Romila Thapar, Śakuntalā: Texts, Readings, Histories (London:
Anthem Press, 2002), 5–8 is misleading.
64 In the rājasūya; see Jan Heesterman, The Ancient Indian Royal Consecration: The Rājasūya
Described According to the Yajus Texts and Annotated (The Hague: Mouton & Co, 1957), 70;
Hiltebeitel, “The Southern Recension’s Śakuntalā as a First Reading,” 22 n. 20.
65 A movement from lower heavenly worlds to higher ones is developed further in the
Uttarayāyata at 1.84.13 ff., and as Sathaye notes in this volume, Yayāti’s grandsons rule later
kingdoms that would have been of interest to epic audiences. And “Gālava” begins with
Garuḍa showing Gālava all four regions from the sky, before he finds his way to the four
grandsons.
The Geography of the Mahābhārata’s Upākhyānas 449
(which is also the case for the Uttara-Yāyāta), but (in the Critical Edition) just
after the Śakuntalā-Upākhyāna. I also attend to only two matters of Yayāti-
Upākhyāna geography: the apparent proximity of the asura king Vṛṣaparvan’s
city to Yayāti’s capital; and Yayāti’s dispensation of lands to his five sons.
Until one gets to the latter scene on which the Yayāti-Upākhyāna closes,
Yayāti’s story presents a geography of two kingdoms. These are Yayāti’s king-
dom among humans and Vṛṣaparvan’s among asuras. A third kingdom, Indra’s
among gods, is there too, but less in play. The key ṛṣi is the asuras’ imperious
chaplain, guru, and sorcerer, Kāvya Uśanas or Śukra, and it is the fortunes of this
brahman’s daughter Devayānī that come to be of interest in all three kingdoms,
but especially in Vṛṣapavan’s and Yayāti’s. In their two kingdoms, the main
locations are their two purams, and between them lies a forest that is, once
again, “like Citraratha’s” (Mahābhārata 1.73.4b). There Yayāti meets Devayānī
when he pulls her up by hand from a well—an action that she later recalls
as equivalent to his having married her when he meets her there a second
time, both meetings having occurred while he was hunting. Van Buitenen cre-
atively waffles in translating puram. He wants both purams at least for a while
to be folkloric “castles.” Thus he first makes Vṛsaparvan’s puram his “castle”
when Śarmiṣṭḥā, Vṛṣaparvan’s daughter, leaves it to become Devayānī’s slave
(1.75.19d).66 Likewise, Yayāti’s puram is his “castle” when, just after he has mar-
ried Devayānī, he returns happily to it with her (76.35c; 77.1).67 Yet Vṛṣaparvan’s
puram has from the beginning been where Śukra dwells “in the city of the lord
of the asuras” (asurendrapure; 1.71.16c). And in the end Yayāti too has “departed
from his city (purāt)” to undertake a forest life (1.80.25c), once he has gotten his
“city and country peoples’” (paurajānapadais; 80.24a) leave to go.
Two questions emerge from these three main locations in this phase of the
story: Where is Yayāti’s puram? And where is it in relation to Vṛṣaparvan’s and
the forest between them? The location of Yayāti’s capital is not made explicit
in Book 1 (that is, in either the Yayāti-Upākhyāna or the Uttara-Yāyāta), but
Kaṇva mentions that it was Pratiṣṭhāna in the Gālava-Ākhyāna in Book 5
(Mahābhārata 5.112.9c). Yet one has no reason to think it would have been any-
where else in Book 1. For early in the Uttara-Yāyāta, Yayāti tells Indra that he
gave his central inheritance to Pūru in terms that befit Pratiṣṭhāna: “All this
country between the Ganges and the Yamunā is yours. You shall be king in the
middle of the earth” (1.82.5a-c; Van Buitenen trans.). When the Pāṇḍavas hear
66 And again when Devayānī, satisfied, agrees to go back to it (Mahābhārata 1.75.24a); and
when Kāvya Uśanas, also placated, goes back into the “castle” with both of them (75.25c).
67 And again when Yayāti repairs to it after getting Uśanas to mitigate his curse, and asks first
his eldest son Yadu to give him his youth (Mahābhārata 1.79.1).
450 Hiltebeitel
68 Each king, including Ilvala, greets Agastya and the kings who one after another join him,
“at the border of his realm” (viṣayānte; Mahābhārata 3.96.2c, 13d, 97.1d; viṣayasyānte,
96.7c). I mention the possibility that having the Ikṣvaku third may suggest Ayodhyā, but if
so, the text leaves it only a suggestion, which may be significant in this epic representation
of Vedic times. On Vedic Ikṣvākus and their eastward movements, see Michael Witzel,
“The Vedas and the Epics: Some Comparative Notes on Persons, Lineages, Geography, and
Grammar,” in Epics, Khilas, and Purāṇas: Continuities and Ruptures, Proceedings of the
Third Dubrovnik International Conference on the Sanskrit Epics and Puranas, September
2002, ed. Petteri Koskikallio (Zagreb: Croatian Academy of Sciences and the Arts, 2005),
21–80.
The Geography of the Mahābhārata’s Upākhyānas 451
though she is a brāhmaṇī, turns out to have the same dark complexion of the
earth as Draupadī. As noted, however, the gods interest in her fortune is not
direct, and the main ṛṣi in the story, her father, has to be neutralized because he
favors the asuras. If the Śakuntalā-Upākhyāna hints at devakārya or the work-
ings of a divine plan, the Yayāti-Upākhyāna suggests that Yayāti is more on his
own than that any gods or ṛṣis might be guiding him. Part of Yayāti’s charm is
that he always seems to be making things up as he goes along. Yet as the
Agastya-Upākhyāna shows, in Yayāti’s ancient “Vedic times,” other human
kings had demon neighbors.69 It is just that whereas their lands and kingdoms
were little defined, Yayāti’s occupied the center of the earth.
This brings us to the second matter of geographical importance in the Yayāti-
upākhyāna, his apportioning of the earth among his five sons. On this, Georges
Dumézil made an argument that the “Vedic times” represented in the Yayāti
story would once have been actual, since he viewed the full Yayāti dossier to be
laced not only with para-Vedic elements but still older Indo-Iranian and Indo-
European ones, through which he considered Yayāti as a type of legendary
“first king.” Even though I no longer find convincing Dumézil’s, Wikander’s, or
any other scholar’s arguments for para-Vedic elements in the Mahābhārata’s
main story,70 such an argument could still carry for an old upākhyāna about a
first king. I believe that in the case of the Yayāti-Upākhyāna, Dumézil’s careful
study means that we must still take that possibility seriously.71 Still, let us note
that there is one element that Dumézil makes no attempt to align with Vedic,
Indo-Iranian, or Indo-European prototypes. It is the very one we have found
the Mahābhārata itself assigning to “Vedic times”: the proximity of neighborly
69 The Yayāti- and Agastya-Upākhyānas are also connected by the power of revival—even of
one who has digested someone (Kāvya Uśanas’s disciple Kaca; other brahmans but not
Agastya) who will burst open the eater (Kāvya Uśanas; Ilvala) to come out alive. In both,
the power belongs originally to demons (in “Yayāti” through their guru), who have to be
deprived of it. One difference is that whereas Kāvya Uśanas imparts his secret to his brah-
man disciple Kaca so that it can be used to revive Uśanas as well, Ilvala has apparently not
told his brother Vātāpi. But Vātāpi has no chance to come out alive thanks to Agastya’s hot
digestive tract.
70 See Hiltebeitel, Dharma: Its Early History in Law, Religion, and Narrative, 379 n. 96; “The
Mahābhārata and the Stories Some People Tell about It—Part 1,” Exemplar: The Journal of
South Asian Studies 1, no. 2 (2012): 2–26 and “The Mahābhārata and the Stories Some Peo-
ple Tell about It—Part 1,” Exemplar: The Journal of South Asian Studies 2, no. 1 (2013): 1–14,
and Hiltebeitel, “The Mahābhārata and the Stories Some People Tell about Its Tribal and
Earliest Histories,” on Dumézil’s view that Dharma’s incarnation in Yudhiṣṭhira is a late
retouch of an older Vedic connection between Yudhiṣṭhira and Mitra.
71 As I viewed Dumézil’s findings about Yayāti and Mādhavī in Alf Hiltebeitel, “Dumézil and
Indian Studies,” Journal of Asian Studies 34 (1974): 136–37.
452 Hiltebeitel
The whole land between the Ganges and the Yamunā is yours; you your-
self will be king over the center of the earth; your brothers will be lords of
the outlying regions.72
With his interpretation of Yayāti as a first king, Dumézil wants these “outlying
regions” to be, or to originally have been, the four cardinal directions. That is,
he wants Pūru’s four older brothers, each of whom has rejected Yayāti’s request
to exchange their youth for his old age, to be given an inheritance of a quarter
of the earth. But much depends on how antya is translated in the last pāda. Van
Buitenen takes it similarly: “your brothers shall rule the outer regions.”73 We
have at least seen anta (if not antya) used quite appositely, in the sense Dumézil
wants, in the Śakuntalā-Upākhyāna, where Vaiśaṁpāyana introduces Duḥṣanta
as a “herdsman of the earth to her four horizons” (pṛthivyāś caturantāyā goptā;
Mahābhārata 1.62.3c; Van Buitenen trans.).74 But Yayāti could also be saying,
“your brothers will be lords over ‘peripherals’” in the sense of “the lowest,” since
these are more specific meanings of antya.75 Such is its sense when Duryodhana,
having collected the tribute at Yudhiṣṭhira’s Rājasūya, says, “The Himalayas
and oceans and marshes that produce all the gems serve like the lowliest
72 Georges Dumézil, The Destiny of a King, trans. Alf Hiltebeitel (Chicago: University of Chi-
cago Press, 1973), 27 (Dumézil’s italics).
gaṅgāyamunayor madhye kṛtsno ‘yaṃ viṣayas tava |
madhye pṛthivyās tvaṃ rājā bhrātaro ‘ntyādhipās tava || (Mahābhārata 1.82.5)
73 J.A.B. van Buitenen, trans. The Mahābhārata, vol. 1. 1. The Book of the Beginning (Chicago:
University of Chicago Press, 1973), 197 (my italics).
74 See also among Śakuntalā’s parting words to Duḥṣanta before he hears from the heavenly
voice: “my son shall reign over the four-cornered earth” (caturantām imām urvīm; Mahābhā
rata 1.69.27).
75 Monier-Williams’s primary definition of antya is “last in place, in time, or in order; … low-
est in place or condition, undermost, inferior, belonging to the lowest class.” Monier
Monier-Williams, A Sanskrit-English Dictionary (Oxford: Clarendon Press, 1964), 44.
The Geography of the Mahābhārata’s Upākhyānas 453
Therefore, Turvaśu, your offspring will face extinction. Fool, you shall rule
over people whose customs and laws are corrupt and whose walks of life
run counter to decency, the lowest ones [antyeṣu] who feed on meat.
They will lust after the wives of their gurus and couple with beasts; evil
barbarians that follow the laws of cattle are they whom you will rule!
(Mahābhārata 1.79.11-13; Van Buitenen trans.)
Yayāti is of course cursing the four oldest sons who will not exchange their
youth for his senility, and Turvaśu in particular will be linked with the Yavanas
(80.20). Yayāti is not assigning them four directional lordships they might actu-
ally want.
Dumézil, who often wants to reconstruct an ancient civility and nobility for
the ārya and the Indo-Europeans,76 seems to suppress this demeaning sense
of antya. It “could well be,” he says, “that, from vedic times, Yayāti, through his
sons, was regarded as the one who was responsible for the distribution of men
among the ‘Fifths’ of the earth, that is, the four cardinal points and the center.”77
He relies here on ṚV 8.10.5, addressed to the Aśvin twins: “May you be in the
east or the west, O possessors of high good, may you be among Druhyu, Anu,
Turvaśa or Yadu, I call you, come.” The mention of east and west together is,
according to Dumézil,
76 See some discussion in Hiltebeitel, Dharma: Its Early History in Law, Religion, and Narra-
tive, 385–86.
77 Dumézil, The Destiny of a King, 19.
454 Hiltebeitel
the arya nationality), analyzed into its typical divisions, fills out the local
divisions of the space it occupies.78
But Dumézil is very much aware that there are difficulties in taking Yayāti’s
main usage to imply the four cardinal directions:
In fact, despite this beautiful declaration opposing the center and the
periphery, the countries where the four bad sons are sent seem to be con-
centrated in the northwest. This probably is related to displacements of
peoples and changes in their horizon between the time of the hymns and
that of the epic. This śloka thus prolongs the old doctrine, while śloka
3533 [this should be 3433] ‘actualizes’ the doctrine geographically in an
inadequate, degenerate fashion.79
From Yadu are born the Yādavas, from Turvasu the Yavanas, from Druhyu
the Bhojas, and from Anu the Mlecchas (Dumézil trans., 1973, 18).
Clearly, Dumézil does not actually suppress the “inferior” meaning of antya; he
rather displaces it to a degeneration or degradation of the myth itself, in which
all five of Yayāti’s sons would have peopled the earth with āryas.
Dumézil also never mentions that the world-centered capital of the “whole
land between the Ganges and the Yamunā,” which Yayāti bequeaths to Pūru, is
at Pratiṣṭhāna. That too would have to have been a relocation from the north-
west, though not a degradation of the myth. For if the rulerships he assigns to
his four oldest sons surrounded this center, it could not be that far to their east.
Here Dumézil could have turned for support to O.P. Bharadwaj, who finds a
way to locate the Pratiṣṭhāna of Yayāti’s ancestor Purūravas in the “Sarasvatī-
Dṛṣadvatī doab” and Kurukṣetra area. Bharadwaj traces the etymologies of
several local names there back to a hypothetical Pratiṣṭhāna “without claim-
ing the list to be exhaustive,” and argues that, with its prefix prati- meaning
“against or in opposition to,” Pratiṣṭhāna could denote “a town founded against
or in opposition to an earlier town” and thereby describe “a king’s second or
new seat of Government.” From this vantage point he asks “the crucial ques-
tion” of whether Purūravas’s Pratiṣṭhāna can be spotted “anywhere on the
78 Ibid., 19–20.
79 Ibid., 137 n. 57, citing D.D. Kosambi,“The Vedic Five Tribes,” Journal of the American Orien-
tal Society 87 (1967): 38–39.
The Geography of the Mahābhārata’s Upākhyānas 455
Sarasvatī,” and affirms that it can be located at a town named Patran, and pro-
nounced Pātaḍān, “not very far from the western bank of the Ghaggar which
is undoubtedly the original Ṛgvedic Sarasvatī.”80 Although all of this is implau-
sible none of it is impossible. But even if one granted that the Yādavas, Bhojas,
Yavanas, and Mlecchas could denote “degraded” āryas, I still do not see any way
to get this degradation to hark back to four lands identified with the four cardi-
nal directions. Rather, it seems that Yayāti curses all four to be identified with
only one proverbially bad direction, which can be said of northwestern lands
in both epics.81 At best, this could still be evidence that the Yayāti-Upākhyāna
retains this orientation as a Vedic reminiscence while relocating the center to
Pratiṣṭhāna.
This brings us to the Tapatī-Upākhyāna, which, as noted, tells of the birth of
Kuru, a dynastic successor of Yayāti and Pūru, Duḥṣanta and Bharata. The Tapatī-
(Mahābhārata 1.160–63), Vasiṣṭha- (1.164–68, 173), and Aurva-Upākhyānas
(1.69–72) are told one after the other, with the last two overlapping, in the
Citraratha Upaparvan (1.153–73). The Upaparvan begins just after the Pāṇḍavas
have finished up their business with the Rākṣasa Baka at Ekacakrā, and ends
just before the next upaparvan (1.174–85) takes them to Pañcāla for Draupadī’s
svayaṃvara. A brahman comes to Ekacakrā and tells stories, and is asked about
the Pañcāla court. He retells the story of Droṇa and Drupada (see 1.121–22 and
1.128) up to the births of Draupadī and Dhṛṣṭadyumna, and Droṇa’s acceptance
of Dhṛṣṭadyumna as his pupil. Kuntī then says that Ekacakrā has gotten boring,
they ought to go to Pañcāla. On the way they meet Vyāsa who tells them that
Draupadī is destined to be their wife. Reaching the Somaśravāyana Ford on the
Gaṅgā at night, they trespass on the territory of the Gandharva Aṅgāraparṇa
Citraratha, who threatens them. I am not certain whether this is a down–to-
earth version of Citraratha’s Park. In any case, Arjuna challenges and defeats
him with an āgneya missile, and Citraratha’s wife prays for mercy, which
Yudhiṣṭhira grants (158.1–30). The Gandharva gives up his name and bestows
on Arjuna a magic of Vision, which came from Manu to Soma to Viśvāvasu
to Citraratha, and he gives each Pāṇḍava a hundred horses, wherepon he and
Arjuna make a friendship-alliance (30–55). Citraratha says that the Pāṇḍavas
were vulnerable because they have neither fires nor a priest; they should find
a purohita. He also tells Arjuna that he knows the five brothers as having six
progenitors, their divine parents plus Pāndu; and he keeps calling Arjuna by
the name Tāpatya, which leads Arjuna to formulate his leading question that
prompts Citraratha to tell him the Tapatī-Upākhyāna (159).
As with Duṣyanta and Yayāti before him, everything begins for Saṃvaraṇa
when he goes hunting. Though by Saṃvaraṇa’s time, we might imagine that
the future “Kurus,” who would seem to be named after his son, might have
again moved their capital, the epic does not help us to make such an identifi-
cation.82 But once again, there is no name for Saṃvaraṇa’s city/capital (see
Mahābhārata 1.160.20; 163.15–20), or of the mountain toward which he heads.
The main novelty seems to be that his horse dies on this mountain’s slope (par-
vatopavane; 160.21d) or plateau (giriprasthe; 160.26a; 162.10c), which is also left
vague. As with so many of these upākhyānas, where geography is invoked, the
point is to fit the geograph to the cosmograph. How is a king to meet the Sun’s
daughter, or get in touch with her again after she has bedazzled him first on her
own doing? The second problem is solved by Saṃvaraṇa’s purohita Vasiṣṭha.
While Saṃvaraṇa watched from his position of prayer, Vasiṣṭha strode up from
the mountain plateau to see and converse with the Sun (162.16)! The upākhyānas
foreground a narrative world where the main Mahābhārata story’s geograph
would be believable. With this, Tapatī joins the run of divine mothers who
have entered and, with Draupadī, are about to enter, the Mahābhārata patri-
line. The upākhyāna ends with Tapatī giving birth to Kuru (163.23ab), which
leaves it somewhat surprising that before this, Saṃvaraṇa is already repeatedly
called a Kuru (160.12b; 161.3c; 163.6d) and even Kauravaśreṣṭha, “the best of
Kauravas” (161.41c).
No connection is made during the Tapatī-Upākhyāna with the fact that
Vasiṣṭha is Saṃvaraṇa’s purohita and did just what was needed to bring
Saṃvaraṇa and Tapatī together. One must wait for Arjuna to make this con-
nection at the beginning the Vasiṣṭha-Upākhyana, asking to hear more about
“the blessed seer who was the house-priest of our forebears” (Mahābhārata
1.164.4). The Gandharva then makes that his next point (164.11–14). It is curious
that Vasiṣṭha serves the Kuruvaṃśa in the Tapatī-Upākhyāna. The Vasiṣṭha-
Upākhyāna quickly makes it clear that the Mahābhārata knows him as
82 Such a geography would reflect the one described by Michael Witzel (“The Development
of the Vedic Canon and Its Schools: The Social and Political Milieu,” in Inside the Texts—
Beyond the Texts: New Approaches to the Study of the Vedas, ed. Michael Witzel [Cam-
bridge, MA: Department of Sanskrit and Indian Studies, Harvard University, 1997],
247–345) for the origins of the Kuru state; see Hiltebeitel, “The Mahābhārata and the
Stories Some People Tell about Its Tribal and Earliest Histories.”
The Geography of the Mahābhārata’s Upākhyānas 457
purohita of the Ikṣvākus (1.164.9–10). Perhaps things were also getting boring
in Ayodhyā.
The Vasiṣṭha-Upākhyāna now gets into the story of King Gādhi and his son
Viśvāmitra, from Kanyakubja/Kanauj (1.165.3–4). It is presumably in a forest
not far from Kanauj that Vasiṣṭha’s cow Nandinī routs Viśvāmitra’s anārya
troops (165.35–37). The cow produces mlecchas now right there, in the doāb
heartland and in olden times, to prevent Viśvāmitra from abducting her by
fighting his army. If one prioritizes early elements of the Yayāti-Upākhyāna to
the Vasiṣṭha-Upākhyāna, which might make sense since one can date the
Vasiṣṭha- and Viśvāmitra-maṇḍalas of the Ṛgveda later than its other family
books, which do mention at least the five peoples if not their having any con-
nection with Yayāti,83 one could deduce that mlecchas have now advanced
from the northwestern borders into central India. Note too that three of the
four mleccha contingents that are said all in one verse (35) to come from
Nandinī’s piss and shit—the Śakas, Pahlavas, and Yavanas—are the most his-
torically au courant mlecchas with the epic’s likely time of composition. And
the fourth group mentioned with them, the Śabaras, are a tribal population
from central India itself.84 A more diverse Indic group then comes from
Nandinī’s foam in the next verse (36)—all being, however, mlecchas (37).
Next we come to the story of Kalmāṣapāda, the Ikṣvāku king whom Vasiṣṭha’s
son Śakti curses to become a cannibal in some other forest where the Ikṣvāku
king is hunting. During their mutual vituperation, Viśvāmitra is a hidden lis-
tener who uses his knowledge of this curse to contrive the killing of Śakti and
then Vasiṣṭha’s other ninety-nine sons—who are all set up by Viśvāmitra for
Kalmāṣapāda to devour after Śakti has cursed him to become a cannibal. When
Vasiṣṭha has lost his hundred sons and tried many times to kill himself, his first
attempt is to jump from Mount Meru, hitting his head on a rock (166.41)! This
could be in the same vein as his striding up to see Sūrya in the Tapatī-
Upākhyāna. But one may also recall that as part of what follows from the
Mahābhiṣa-Upākhyāna, the Vasus once came too near to Vasiṣṭha on Mount
Meru, and that there too there was some funny business about Dyaus being the
ringleader in stealing Vasiṣṭha’s cow, which led to Dyaus’s incarnation as
Bhīṣma.85 Meanwhile, back down to earth, after his ineffective leap from
Mount Meru and two further failed attempts at suicide, the first by fire in a
great forest (mahāvane), and the second in an unnamed (but in terms of what
follows, let us guess the “western”) ocean (Mahābhārata 166.42–45), Vasiṣṭha
jumps into a river that leaves him “freed from the fetters [pāśair vimuktaḥ]”
with which he had bound himself to drown, and names the river the Vipāśā
(Beas), after which he makes one more try to kill himself while wandering
toward the Himālaya, where another anecdote explains why he gave the name
Śatadru (Sutlej) to the river he jumps in there, which, considering Vasiṣṭha to
be like fire, “ran off in a hundred directions [śatadhā vidrutā[” (167.6–9). The
epic has thus moved from Vasiṣṭha cosmological preeminence on Meru to his
quasi-cosmogonic naming of two rivers of the Vedic Panjab.86
Vasiṣṭha finally decides to remain among the living when he hears the voice
of Veda recitation coming from the womb of his daughter-in-law (snuṣā)
Adṛśyantī, who is carrying the precocious child of Śakti in her twelfth year of
pregnancy (Mahābhārata 167.14). The child is Parāśara, named after this
incident,87 who remains in the womb far longer than it takes his son Vyāsa
(who was born knowing the Veda on the day he was conceived), and four times
as long as it took Vyāsa to have “made the Mahābhārata.”88 In this light it can
hardly be incidental that Adṛśyantī now tells Vasiṣṭha that only he can ward off
the cannibal Kalmāṣapāda (Śakti’s devourer), since Vasiṣṭha is “the first of all
scholars of the Veda” (sarvavedavidāṃ vara; 167.20b). Kalmāṣapāda uses the
same epithet shortly after this (168.13) when requesting a boon from Vasiṣṭha
on behalf of the Ikṣvākus,89 which will be that Vasiṣṭha come back with him to
Restrain your mind from this evil destruction of all the worlds [pāpāt sar-
valoka parābhavāt]. For none of the barons or any of the seven worlds
[lokāh sapta] offended our might and mortification, son. (Mahābhārata
1.170.20c-21b; Van Buitenen trans.)
The inset Aurva-Upākhyāna also ends by taking us to one of the outer cosmo-
logical worlds when Vasiṣṭha tells Parāśara that Aurva’s Bhṛgu ancestors
convinced Aurva to release his wrath (manyu) into Varuṇa’s realm, the ocean,
where it became what Vedavids know as the great Horse’s Head that spits fire
while eating or drinking the ocean’s waters (Mahābhārata 171.21–23)—the
same Horse’s Head whom the Nārāyaṇīya’s quasi-upākhyāna91 about Hayaśiras
knows as a Veda-spouting form of Nārāyaṇa.
92 See Mahābhārata 1.189.1–34; Alf Hiltebeitel, The Ritual of Battle: Krishna in the Mahābhā
rata (Albany: State University of New York Press, 1990 [1976]), 144–92; Hiltebeitel, Rethink-
ing the Mahābhārata, 118–19; Hiltebeitel, “The Mahābhārata and the Stories Some People
Tell about It,” Parts 1 and 2; and Hiltebeitel, “The Mahābhārata and the Stories Some Peo-
ple Tell about Its Tribal and Earliest Histories.”
The Geography of the Mahābhārata’s Upākhyānas 461
those who go into hiding until “treasure-filled Earth saw sacrifice and Veda-
study halt, kings and Brahmins perish,” rites and festivals cease, plowing and
cattle-tending end, and her cities and hermitages razed, such that “Earth,
bestrewn with bones and skeletons, became a loathsome sight” (22–24). Their
mission accomplished such that the universe (jagat) is in disarray up to the
moon, sun, stars, and constellations, “the two Daityas, having thus cruelly con-
quered all the directions [diśo], without a rival left, made their dwelling place
at Kurukṣetra” (27).
This is said, I presume as a point of emphasis, in the last verse of an
adhyāya, whereupon the next one begins with the devarṣis and siddhas, who
have seen this massacre, going “out of pity for the universe” (jagataḥ kṛpayā;
Mahābhārata 203.2d) to Brahmaloka to report matters to Brahmā. He is sur-
rounded there by some of the highest gods—Mahādeva, Agni, Vāyu. Candra
and Āditya, Dharma, and Parameṣṭhin (203.4)—and by strange high-spirited
ṛṣi-types: not only by siddhas and brahmarṣis (203.3c) but by Vaikhānasas,
Vālakhilyas, Vānaprasthas, Beam-Drinkers (marīcipāḥ), Unborn Ones (ajāḥ),
the Undistracted (avimūḍhāḥ), Firewombs (tejogarbhāḥ), and other tapasvins
(5). In consultation with Mahādeva Sthāṇu, Parameṣṭhin, Indra, Dharma, and
others, Brahmā summons Viśvakarman to create Tilottamā, who, in circum-
ambulating the gods before departing, makes the gawking Shāṇu into a virtual
caturmukhaliṅgam (“four-faced liṅga”) and Indra into the “thousand-eyed”
Sahasrākṣa (26).93 As she leaves, “all the gods and supreme ṛṣis thought that
her perfect beauty had already done the gods’ work [kṛtam ityeva tat kāryam]”
(203.29).94
How long Sunda and Upasunda stayed at Kurukṣetra and what they did
there goes unmentioned. The denouement has the two asuras, now corrupt
with power and pleasures, return “one day to play on a rock plateau on a ridge
of the Vindhya” (vindhyasya pṛṣṭhe sama śilātale; Mahābhārata 204.6ab) with
their women, engaging in wine, song, and dance. There Tilottamā makes her
appearance, and the brothers clobber each other to death with maces (gadās),
seeing which, their women and the rest all retreat in fear and despair (viṣāda)
to Pātāla (204.20)! Brahmā then dismisses everyone from Brahmaloka, and
puts Indra in charge of the triple world.
93 “Thus the Great God Sthāṇu became of yore four-faced, and the slayer of Vala thousand-
eyed” (Van Buitenen trans.).
94 Literally, “the Gods and Supreme Ṛṣis thought that her beauty had [already] done what
was to be done,” that is, the devakārya or “the work of the gods.” On this recurrent term,
see Hiltebeitel, Dharma: Its Early History in Law, Religion, and Narrative, chapter 12.
462 Hiltebeitel
97 See The Mahābhārata, vol. 7, 769 n. to 12.148.12: “The Sarasvatī bounds the northern side of
‘Kuru’s Field,’ and the Dṛṣadvatī the southern; see MBh 3.81.175 and MBh 9.36.41 and 48, which
locates Kurukṣetra on the southern bank of the Sarasvatī.”
98 See Stephanie Jamison, Sacrificed Wife/Sacrificer’s Wife: Women, Ritual and Hospitality in
Ancient India (New York: Oxford University Press, 1996), 248 on this upākhyāna; Hiltebei-
tel, Rethinking the Mahābhārata, 145 n. 55 on Vṛddhakanyā Tīrtha and its heroine.
464 Hiltebeitel
99 That is, there is nothing about his mother Sudarśanā being a putrikā for King Duryod-
hana. See Brodbeck, The Mahābhārata Patriline.
100 As the form taken by Sarasvatī there, according to 9.37.24c–25 see Brodbeck, The
Mahābhārata Patriline, 147. On the Sudarśana-Upākhyāna, see Brodbeck, The Mahābhā
rata Patriline, 147–48; Jamison, Sacrificed Wife/Sacrificer’s Wife, 153–56, 247–48.
The Geography of the Mahābhārata’s Upākhyānas 465
The last two upākhyānas about unusual human beings at Kurukṣetra are
gleaners’ tales—the Mudgala- and Nakula-Upākhyānas—but they are differ-
ently oriented. In the Mudgala-Upākhyāna, in response to Yudhiṣṭhira’s
question, which of the two counts for more in the afterworld, dānadharma or
tapas (Mahābhārata 3.245.26), Vyāsa speaks in favor of giving so long as one
gives rightfully obtained wealth (32), which leads him to tell about the ṛṣi
Mudgala who gave unstintingly to guests what little he garnered from living
righteously off what rice he gleaned from harvested fields at Kurukṣetra
(246.3a). Yet when an envoy of the gods tries to interest him in ascending to
heaven, telling Mudgala he will find there “the law-minded, the masters of self,
the serene and controlled and unenvious, those accustomed to dānadharma,
and champions with their scars showing” (247.14), Mudgala rejects heaven in
favor of “the eternal and supreme perfection that is marked by extinction.”101
Mudgala is a solitary straight-shooter like the Vṛddhakumārī, even if he wants
something higher than she does, and does not have to deal with any one-night
interruptions.
In the Nakula-[Mongoose-]Upākhyāna, the Mahābhārata’s last, a half-
golden blue-eyed mongoose, soon revealed to have been Dharma in disguise,
tells Yudhiṣṭhira that his aśvamedha was not worth the barley grains of a
gleaner whose vow was to live off what he got from picking over the ground
like a pigeon “on the dharma-field Kurukṣetra chosen by many who know
dharma” (dharmakṣetre kurukṣetre dharmajñair bahubhir vṛte; Mahābhārata
93.2ab).102 This uñchavṛtti- or gleaner-brahman still fed his guests before he fed
his wife, son, or daughter-in-law, and these three likewise observed his vow
along with him, when Dharma, in still another disguise, appeared as a hungry
brahman to test the gleaning brahman’s “pure gift” (śuddha dāna; 93.57).103
Like the Sudarśana-Upākhyāna, this one tells about a household or family.
These families of Kurukṣetra do not pursue a solitary’s exit from the world to
another but an arduous dharma in this one.
Needless to say, one could say more about these four upākhyānas just as a
set, not to mention their wider thematic repercussions with gleaners104 and
101 Mahābhārata 3.247.43cd: jagāma śāśvatīṃ siddhiṃ parāṃ nirvāṇalakṣaṇam; see J.A.B.
van Buitenen, trans., The Mahābhārata, vol. 2. 2. The Book of the Assembly Hall; 3. The Book
of the Forest (Chicago: University of Chicago Press, 1975), 705.
102 The first pāda echoes Bhagavadgītā 1.1; the second could also be translated “covered by
many who know dharma.” Cf. Mahābhārata 14.92.7, 21 also mentioning the gleaner’s resi-
dence at Kurukṣetra.
103 See Hiltebeitel, “Not Without Subtales,” 492; Dharma: Its Early History in Law, Religion,
and Narrative, 436.
104 Note how the beginning of the Uttara-Yāyāta includes a phase for Yayāti to practice glean-
ing (Mahābhārata 1.81.13c) toward the end of the first thousand years of the long ascetic
466 Hiltebeitel
with Dharma taking on disguises. But what I would like to suggest in closing is
this: what the epic poets do with Kurukṣetra is describe a place that is familiar
to them. When they speak of such unusual men and women—no less oddballs
than those in Brahmaloka, but in different ways—they may actually be writing
a Mahābhārata ethnography out of their own experience there. I have been
pursuing this notion in other recent essays,105 and for now, let me just cite
T.P. Mahadevan, with whom I have been putting my head together on it: “The
story seems to begin with the Pūrvaśikhā Brahmans, with their original homes
in the Kuru-Pāñcāla area. … They may well have been part of Hiltebeitel’s inter-
generational committee of ‘out of sorts’ Brahmans, the Śrotriya or Uñchavṛtti
Brahmans, giving us the first Brahmanical redaction of the epic, … or a text
very close to it, ca. third bce.”106 Actually, I was never bold enough to think that
gleaners could have been among the composers, and am wary of the phrase
“first Brahmanical,” as if there must have been something earlier. And I still
would prefer a second to first bce date. But the Pūrvaśikhā geography looks full
of promise.
References
Primary Sources
The Mahābhārata for the First Time Critically Edited. Edited by Vishnu S. Sukthankar,
Shripad K. Belvalkar, Parashuram L. Vaidya, et al. 19 vols. Poona: Bhandarkar Oriental
Research Institute, 1933–66.
The Mahābhārata, vol. 1. 1. The Book of the Beginning. Translated by J.A.B. van Buitenen.
Chicago: University of Chicago Press, 1973.
The Mahābhārata, vol. 2. 2. The Book of the Assembly Hall; 3. The Book of the Forest.
Translated by J.A.B. van Buitenen. Chicago: University of Chicago Press, 1975.
The Mahābhārata, vol. 3. 4. The Book of Virāṭa; 5. The Book of the Effort. Translated by
J.A.B. van Buitenen. Chicago: University of Chicago Press, 1978.
regime that gets him just to Indraloka. On gleaners, cf. Hiltebeitel, Dharma: Its Early His-
tory in Law, Religion, and Narrative, 611–14, contrasting earthly gleaners and odd-type
celestial Ṛṣis in the Umā-Maheśvara-Saṃvāda much as I contrast the strange types here
found at Kurukṣetra and in Brahmaloka. See further Hiltebeitel, “From Ṛṣidharma to
Vānaprastha” and Non-Violence in the Mahābhārata.
105 See Hiltebeitel, Non-Violence in the Mahābhārata and “Tīrthas, Temples, Āśramas and
Royal Courts.”
106 Thennilapuram Mahadevan, “The Three Rails of the Mahābhārata Textual Tradition,”
Journal of Vaishnava Studies 19, no. 2 (2011): 41.
The Geography of the Mahābhārata’s Upākhyānas 467
The Mahābhārata, vol. 7. 11. The Book of Women. 12. The Book of Peace, Part 1. Translated
by James L. Fitzgerald. Chicago: University of Chicago Press, 2004.
Secondary Sources
Bakhtin, M.M. The Dialogic Imagination. Translated by Caryl Emerson and Michael
Holquist. Austin: University of Texas Press, 1981.
Bharadwaj, O.P. Studies in the Historical Geography of Ancient India. Delhi: Popular
Prakashan, 1986.
Biardeau, Madeleine. Le Mahābhārata: Un récit fondateur du brahmanisme et son inter-
prétation, 2 vols. Paris: Éditions du Seuil, 2002.
———. “Śakuntalā dans l’ épopée.” Indologica Taurinensia 7, Dr. Ludwig Sternbach
Felicitation Volume, Part I (1979): 115–25.
Bowles, Adam. “The Dasyu in the Mahābhārata.” In The Churning of the Epics and
Purāṇas at the 15th World Sanskrit Conference, edited by Simon Brodbeck, Alf
Hiltebeitel, Adam Bowles, 155–72. New Delhi: Rashtriya Sanskrit Sansthan and
D.K. Printworld, in press.
Brodbeck, Simon. “Husbands of the Earth: Kṣatriyas, Females, and Female Kṣatriyas in
the Strīparvan of the Mahābhārata.” In Epic Undertakings: Papers of the 12th World
Sanskrit Conference, vol. 2, edited by Robert P. Goldman and Muneo Tokunaga, 33–63.
Delhi: Motilal Banarsidass, 2009.
———. The Mahābhārata Patriline: Gender, Culture, and the Royal Hereditary. Aldershot:
Ashgate Publishing, 2009.
Bronkhorst, Johannes. “Āśramas, Agrahāras, and Monasteries.” Paper presented at the
Fifth Dubrovnik International Conference on the Sanskrit Epics and Purāṇas.
Dubrovnik, August 11–16, 2008.
Couture, André, trans. L’enfance de Krishna: Traduction des chapitres 30 à 78 (édition
critique) du Harivaṃśa. Paris: Les éditions du Cerf; and Montreal: Presses de
l’Université Laval, 1991.
Danino, Michel. The Lost River: On the Trail of the Sarasvatī. New Delhi: Penguin, 2010.
Dumézil, Georges. The Destiny of a King. Translated by Alf Hiltebeitel. Chicago: University
of Chicago Press, 1973.
———. Mythe et épopée, I: l’Idéologie des trois fonctions dans les épopées des peuples
indo-européens. Paris: Gallimard, 1968.
Fitzgerald, James L. “Negotiating the Shape of ‘Scripture’: New Perspectives on the
Development and Growth of the Mahābhārata between the Empires.” In Between
the Empires: Society in India 300 bce to 400 ce, edited by Patrick Olivelle, 257–86.
Oxford: Oxford University Press, 2006.
———. “The Rama Jāmadagnya ‘Thread’ of the Mahābhārata: A New Survey of Rāma
Jāmadagnya in the Pune Text.” In Stages and Transitions: Temporal and Historical
Frameworks in Epic and Purāṇic Literature, Proceedings of the Second Dubrovnik
468 Hiltebeitel
International Conference on the Sanskrit Epics and Purāṇas, Aug. 1999, edited by
Mary Brockington, 89–132. Zagreb: Croatian Academy of Sciences and Arts, 2002.
Heesterman, Jan. The Ancient Indian Royal Consecration: The Rājasūya Described
According to the Yajus Texts and Annotated. The Hague: Mouton & Co, 1957.
Hiltebeitel, Alf. “Among Friends: Marriage, Women, and Some Little Birds.” In Gender
and Narrative in the Mahābhārata, edited by Simon Brodbeck and Brian Black, 110–43.
London: Routledge, 2007.
———. “Authorial Paths Through the Two Sanskrit Epics, Via the Rāmopākhyāna.” In
Epic Undertakings: Papers of the 12th World Sanskrit Conference, vol. 2, edited by
Robert P. Goldman and Muneo Tokunaga, 169–214. Delhi: Motilal Banarsidass, 2009.
———. “Between History and Divine Plan: The Mahābhārata’s Royal Patriline in
Context.” Papers from the May 2010 Cardiff Conference on Genealogy in India, edited
by Simon Brodbeck and James Hegarty. Religions of South Asia 5 (2011): 103–25.
———. “Bhīṣma’s Sources.” In Vidyārṇavavandanam: Essays in Honor of Asko Parpola,
edited by Klaus Karttunen and Petteri Koskikallio, 261–78. Studia Orientalia. Finnish
Oriental Society, no. 94. Helsinki, 2001.
———. Dharma: Its Early History in Law, Religion, and Narrative. New York: Oxford
University Press, 2011.
———. “Dumézil and Indian Studies.” Journal of Asian Studies 34 (1974): 129–37.
———. “From Ṛṣidharma to Vānaprastha: The Southern Recension Makeover of the
Mahābhārata’s Umā-Maheśvara-Saṃvāda.” In The Churning of the Epics and Purāṇas
at the 15th World Sanskrit Conference, edited by Simon Brodbeck, Alf Hiltebeitel,
Adam Bowles, 14–45. New Delhi: Rashtriya Sanskrit Sansthan and D.K. Printworld,
in press.
———. “The Mahābhārata and the Stories Some People Tell about It—Part 1.” Exemplar:
The Journal of South Asian Studies 1, no. 2 (2012): 2–26.
———. “The Mahābhārata and the Stories Some People Tell about It—Part 2.” Exemplar:
The Journal of South Asian Studies 2, no. 1 (2013): 1–14.
———. “The Mahābhārata and the Stories Some People Tell about Its Tribal and Earliest
Histories.” Paper presented at the International Seminar on the Mahābhārata: Its
Historicity, Antiquity, Evolution & Impact on Civilization, April 26, 2012.
———. “The Nārāyaṇīya and the Early Reading Communities of the Mahābhārata.” In
Between the Empires: Society in India, 300 bce to 400 ce, edited by Patrick Olivelle,
227–56. New York: Oxford University Press, 2006.
———. Non-Violence in the Mahābhārata: Śiva’s Summa on Ṛṣidharma and the Gleaners
of Kurukṣetra. London: Routledge, 2016.
———. “Not Without Subtales: Telling Laws and Truths in the Sanskrit Epics.” Journal
of Indian Philosophy 33, no. 4 (2005): 455–511.
———. “On Reading Fitzgerald’s Vyāsa.” Journal of the American Oriental Society 125,
no. 2 (2005): 241–61.
The Geography of the Mahābhārata’s Upākhyānas 469
———. “On Sukthankar’s ‘S’ and Some Shortsighted Assessments and Uses of the Pune
Critical Edition (ce).” Journal of Vaishnava Studies 19, no. 2 (2011): 87–126.
———. Reading the Fifth Veda: Studies on the Mahābhārata. Essays by Alf Hiltebeitel,
vol. 1. Edited by Vishwa Adluri and Joydeep Bagchee. Leiden: E.J. Brill, 2011.
———. Rethinking the Mahābhārata: A Reader’s Guide to the Education of the Dharma
King. Chicago: University of Chicago Press, 2001.
———. The Ritual of Battle: Krishna in the Mahābhārata. Albany: State University of
New York Press, [1976] 1990.
———. “The Southern Recension’s Śakuntalā as a First Reading: A Window on the
Original and the Second Reading by Kālidāsa.” In Revisiting Kālidāsa’s Abhijñāna
Śākuntalam: Land, Love, Languages: Forms of Exchange in Ancient India, edited by
Deepika Tandon and Saswati Sengupta, 17–37. Delhi: Orient BlackSwan Edition, 2011.
———. “Tīrthas, Temples, Āśramas and Royal Courts: Towards an Ethnography of
Mahābhārata Bhakti.” In The Archaeology of Bhakti II: Royal Bhakti, Local Bhakti,
edited by Emmanuel Francis and Charlotte Schmid, 81–126. Pondicherry: IFP & EFEO,
2015.
Jamison, Stephanie. Sacrificed Wife/Sacrificer’s Wife: Women, Ritual and Hospitality in
Ancient India. New York: Oxford University Press, 1996.
Kloetzli, W. Randolph. “Ptolemy and Purāṇa: Gods Born as Men.” Journal of Indian
Philosophy 38 (2010): 583–623.
Kosambi, D.B. “The Vedic Five Tribes.” Journal of the American Oriental Society 87 (1967):
33–39.
Koskikallio, Petteri, ed. Epics, Khilas, and Purāṇas: Continuities and Ruptures. Proceedings
of the Third Dubrovnik International Conference on the Sanskrit Epics and Purāṇas,
September 2002. Zagreb: Croatian Academy of Sciences and Arts, 2005.
Lubotsky, Alexander. A Ṛgvedic Word Concordance. 2 Parts. American Oriental Series,
Vols. 82–83. New Haven, CT: American Oriental Society, 1997.
Macdonell, Arthur Anthony and Arthur Berriedale Keith. Vedic Index of Names and
Subjects, 2 vols. Delhi: Motilal Banarsidass Publishers, 1912.
Mahadevan, Thennilapuram. “On the Southern Recension of the Mahābhārata,
Brahman Migrations, and Brāhmī Paleography.” Electronic Journal of Vedic Studies 15,
no. 2 (2008): 1–143.
———. “Textual Support for Early Integrity of the Mahābhārata from its Southern
Recension.” Paper delivered at the 14th World Sanskrit Conference, Kyoto, 2009.
———. “The Three Rails of the Mahābhārata Textual Tradition.” Journal of Vaishnava
Studies 19, no. 2 (2011): 23–69.
Mitchiner, John E. The Yuga Purāṇa: Critically Edited with an English Translation and a
Detailed Introduction. Bibliotheca Indica, Calcutta: The Asiatic Society, 1986.
Monier-Williams, Monier. A Sanskrit-English Dictionary. Oxford: Clarendon Press, 1964.
470 Hiltebeitel
Parasher-Sen, Aloka. Mlecchas in Early India, a Study in Social Attitudes towards Outsiders
up to ad 600. New Delhi: Munshiram Manoharlal, 1991.
———. “Naming and Social Exclusion: The Outcast and the Outsider.” In Between the
Empires: Society in India 300 bce to 400 ce, edited by Patrick Olivelle, 415–55. New
York: Oxford University Press, 2006.
Sukthankar, Vishnu S. “Epic Studies VI: The Bhṛgus and the Bhārata: A Text- Historical
Study.” Annals of the Bhandarkar Oriental Research Institute 18, no. 1 (1936): 1–76.
———. “Introduction.” In The Mahābhārata for the First Time Critically Edited, vol. 3.
the Āraṇyakaparvan. i–xxxviii. Pune: Bhandarkar Oriental Research Institute, 1942.
———. On the Meaning of the Mahābhārata. Bombay: Asiatic Society, 1957.
———. “Prolegomena.” In The Mahābhārata for the First Time Critically Edited, vol. 1.
The Ādiparvan. i–cx. Poona: Bhandarkar Oriental Research Institute, 1933.
Thapar, Romila. Śakuntalā: Texts, Readings, Histories. London: Anthem Press, 2002.
Thomas, Lynn. “Paraśurāma and Time.” In Myth and Myth-Making, edited by Julia Leslie,
63–85. Richmond, Surrey: Curzon, 1996.
Witzel, Michael. “Brahmanical Reactions to Foreign Influences and to Social and
Religious Change.” In Between the Empires: Society in India 300 bce to 400 ce, edited
by Patrick Olivelle, 457–99. Oxford: Oxford University Press, 2006.
———. “The Development of the Vedic Canon and Its Schools: The Social and Political
Milieu.” In Inside the Texts—Beyond the Texts: New Approaches to the Study of the
Vedas, edited by Michael Witzel, 247–345. Cambridge, MA: Department of Sanskrit
and Indian Studies, Harvard University, 1997.
———. “Early Sanskritization. Origins and Development of the Kuru State.” In Recht,
Staat und Verwaltung im klassischen Indien. The State, the Law, and Administration in
Classical India, edited by B. Kölver, 27–52. Munich: R. Oldenbourg, 1997.
———. “The Vedas and the Epics: Some Comparative Notes on Persons, Lineages,
Geography, and Grammar.” In Epics, Khilas, and Purāṇas: Continuities and Ruptures,
Proceedings of the Third Dubrovnik International Conference on the Sanskrit Epics
and Puranas, September 2002, edited by Petteri Koskikallio, 21–80. Zagreb: Croatian
Academy of Sciences and the Arts, 2005.
Index Index 471
Index