bayt al- ikma 133

Except for a few lines quoted in Tārīkh-ī Bibliography

Bayhaq, which contain a verse of al-Bayhaqī’s
Editions of AL-MA ĀSIN
lampooning al-Bu turī (ed. usaynī, 267;
Ed. Friedrich Schwally, Giessen 1902 (indexed
ed. al-Hādī, 295), his poetry has been by Oskar Rescher, Index und Stellennachweise
neglected and probably lost. His only sur- zu Fr. Schwally’s Baihaqī , Stuttgart 1923) ;
viving work, Kitāb al-ma āsin wa-l-masāwī, ed. Mu ammad Kāmil al-Na sānī, Cairo
1906 (based on Schwally’s edition) ; ed.
which was compiled during the caliphate of Mu ammad Abū l-Fa l Ibrāhīm, Cairo 1961;
al-Muqtadir (r. 294–320/908–32), earned ed. Dār ādir, Beirut n.d. (bowdlerised); ed.
him his reputation (though only in modern Mu amad Sawīd, Beirut 1988 (bowdlerised);
times). This work, which features a number ed. Adnān Alī, Beirut 1999; ed. Abd al-
Rāziq Isbir, Beirut 2005.
of political-religious subjects in its first
section, is an anthology of already existing Sources
adab discourses—anecdotes, lengthy nar- Alī b. Zayd al-Bayhaqī, Kitāb-i tārīkh-i Bayhaq,
ratives and traditions, specimens of prose ed. Qārī usaynī (Hyderabad 1968), 261–7;
Alī b. Zayd al-Bayhaqī, Ta rīkh Bayhaq, Ara-
genres, poetry—and even the introduction bic trans. Yūsuf al-Hādī (Damascus 2004),
is based on quotations from the Kitāb al- 291–6; Ibn al-Faqīh al-Hamadhānī, Kitāb
hayawān (“Book of animals”) of al-Jā i . It al-buldān, ed. Yūsuf al-Hādī (Beirut 1996),
constitutes a turning point in the history of 11–2, 369; Ibn al-Rūmī, Dīwān Ibn al-Rūmī,
ed. usayn Na ār (Cairo 1973–79), 1:411–2,
adab literature, as it marks the beginning 2:581–2, 3:1209, 1251–7, 4:1435–8, 1535–6,
of the literary genre called al-ma āsin wa-l- 1554–5, 1669, 1871–2, 6:2537–9, 2543–4;
masāwī (Gériès, Un genre, 71–101, 149–152; Yāqūt, Mu jam al- udabā , ed. I sān Abbās
EI 2 ). Al-Bayhaqī’s originality is manifest in (Beirut 1993), 3:1029; al-Tha ālibī, Tatim-
mat Yatīmat al-dahr fī ma āsin ahl al- a r, ed.
the meticulous application of his chosen Mufīd Mu ammad Qumay a (Beirut 1983),
method of organising the materials—on 4:108.
the basis of ma āsin versus masāwī (merit,
good, positive, virtuous, proper versus Studies
Ibrahim Gériès, Un genre littéraire arabe. Al-Ma āsin
fault, bad, negative, vicious, improper)—in wa- l-masāwī, Paris 1977; Ibrahim Gériès,
juxtaposed and contrasting chapters, whose al-Ma āsin wa-l-masāwī, EI2; GALS, 1:249;
titles bear the names of the various topics Rhuvon Guest, Life and works of Ibn Er-Rūmī
treated, and in which he presents his ideas, (London 1944), 29, 133; Said Boustany, Ibn
ar-Rūmī. Sa vie et son œuvre (Beirut 1967), index;
conceptions, and attitudes towards them. Beatrice Gruendler, Medieval Arabic praise poetry.
Al-Bayhaqī’s book reflects the practical Ibn al-Rūmī and the patron’s redemption (London
philosophy of the third/tenth-century and New York 2003), 43; Abd al- amīd
adīb (littérateur), and reveals him to be a Jīda, al-Hijā inda Ibn al-Rūmī (Beirut 1974),
198–9; Franz Rosenthal, A history of Muslim
aydī Shī ī, an ethical person with refined historiography (Leiden 1968), 321–2; Henri
tastes, and someone well versed in adab Massé, Ibn al-Fa īh, EI2.
materials—both prose and poetry, and
especially the works of al-Jā i (Gériès, Un Ibrahim Gériès
genre, 79–101). The book, which preserves
materials from earlier works now lost, was
used extensively by the unknown author Bayt al- ikma
of al-Ma āsin wa-l-a dād, falsely attributed
to al-Jā i (Gériès, Un genre, 102–10), but Bayt al- ikma (“the House of Wis-
was neglected by the majority of medieval dom”) was the palace library of the
Arab writers and biographers. early Abbāsid caliphs, mentioned in the
134 bayt al- ikma

sources only in connection with al-Rashīd Fihrist (ed. Gustav Flügel, 2 vols., Leipzig
(r. 170–193/786–809) and al-Ma mūn (r. 1871–2, repr. Beirut 1964), on which
196–218/812–833). The idea, developed in some of the later sources are largely de-
twentieth-century scholarship, that the Bayt pendent. The term bayt al- ikma alternates
al- ikma was a bureau for the large-scale with khizānat al- ikma—Sahl b. Hārūn (d.
translation of Greek books into Arabic, 215/830), for example, is cited as both
operating along the lines of a modern āhib bayt al- ikma, 10, and ā ib khizānat
research institute or even a university, is al- ikma, 120—and sometimes the institu-
entirely incorrect. While we have little tion is referred to merely as khizāna (5 [bis],
information about the real nature of this 19). In the Fihrist these terms are most fre-
library, it is clear that it had more to do quently associated with the caliphs Hārūn
with collecting and preserving books of al-Rashīd and, especially, al-Ma mūn. The
pre-Islamic Iranian and early Arabic lore construction in which this is expressed is
than with transmitting Greek science. either an i āfa (khizānat al-Ma mūn, “al-
The expression bayt al- ikma (as well as Ma mūn’s storehouse [of books],” 5) or
the alternate expression, khizānat al- ikma) a prepositional phrase with li- (khizānat
is apparently the Arabic translation of a al- ikma lil-Ma mūn , “the storehouse of
Middle Persian term for libraries of the wisdom of al-Ma mūn,” 274). Courtiers
Sāsānian kings. A Middle Persian account of al-Mutawakkil (r. 232–47/847–61) in
from the sixth century C.E. states that the the next generation—al-Fat b. Khāqān
Sāsānids and their predecessors kept copies (d. 247/861; 116, 143) and Alī b. Ya yā
of books of religion and science in a ganj b. al-Munajjim (d. 275/888–9; Yāqūt,
(treasury, storehouse), a word equivalent Irshād al-arīb, ed. D. S. Margoliouth, Ox-
to Arabic khizāna (Shaki, 114–25). amza ford 19222, 5:467)—are also designated as
al-I fahānī (d. after 350/961) reports having their own bayt or khizānat al- ikma of
that in pre-Islamic Iran books containing an unsurpassed number of books, showing
recastings in verse of Persian historical that the terms refer to a library in the con-
lore, warfare, and romances were stored ventional sense. In view of its association
in “houses of wisdom” (buyūt al- ikma) with the caliphs al-Rashīd and al-Ma mūn
for the Sāsānian kings (al-Amthāl al- ādira and of the Sāsānian origin of both the
an buyūt al-shi r, cited by Gregor Schoeler, terms and the institution, it seems beyond
2:308). Ādāb al-mulūk, a book on royal de- reasonable doubt that the references are to
portment deriving from Sāsānian sources a palace library.
and ascribed to al-Sarakhsī (d. 286/899), The Fihrist provides the following in-
provides information on the role of the formation about this library when it is
palace library (bayt al- ikma) in connec- mentioned in association with the names
tion with the king’s study of royal history of these two caliphs. Among their holdings
(Rosenthal, 109). that are mentioned explicitly are books
The Arabic term was probably coined described as having “old-fashioned copy-
in early Abbāsid times, in the second half hand” (qadīm al-naskh, 21), one, presumably
of the second/eighth century. Our source in Arabic, allegedly written in the hand of
of information on this matter is almost ex- Abd al-Mu alib b. Hāshim, the grand-
clusively the late fourth/tenth-century book father of the prophet Mu ammad (5),
catalog of al-Nadīm (written 376/987), al- another written in the imyarite script (5),
bayt al- ikma 135

and still another in the Sūdānī script (19). in a much later source to have ordered a
The activities that were carried out in the book on the biographies of Persian kings
library included book copying ( Allān al- (siyar al-mulūk) to be brought to him from
Shu ūbī, 105)—clearly as a means to enrich the Bayt al- ikma (preface of the Nihāyat
the collections—and book binding (Ibn Abī al-arab fī akhbār al-Furs wa-l- Arab of pseudo-
l- arīsh, 10). Mentioned as affiliated with A ma ī, cited by Alī 1951, 143). It is only
the library are Sahl b. Hārūn (10, 120, 125) under al-Ma mūn that we hear of men
and Salm (120, 243, 268, 305) as directors with a different profile affiliated with the
or librarians ( ā ib), and, as associated em- caliph’s library, namely the mathematician
ployees, Abū Sahl al-Fa l b. Nawbakht (fl. and astronomer Mu ammad b. Mūsā al-
c. 158–193/775–809; 274), Sa īd b. Hārūn Khwārazmī, the astrologer Ya yā b. Abī
(120, 125), and Mu ammad b. Mūsā al- Man ūr al-Munajjim, and the mathemati-
Khwārazmī (d. c.232/847; 274), as well as cians known as the Banū Mūsā. We do not
Ya yā b. Abī Man ūr al-Munajjim and the know what became of the library thereaf-
Banū Mūsā (Ibn al-Qif ī, Ta rīkh al- ukamā , ter, but al-Nadīm was able to identify copies
ed. Julius Lippert (Leipzig 1903), 441–2), of books from this khizānat al- ikma when
it being stated explicitly with regard to the he wrote his Fihrist in 376/987.
Banū Mūsā that they were “registered” The library-director Salm, though of
there (athbatahum) by al-Ma mūn. Persian background, is recorded in the
The men mentioned as affiliated with Fihrist as having been involved, as a mem-
this library were for the most part Irani- ber of a committee, in the translation of
ans, and in a few instances it is expressly Ptolemy’s Almagest (267f.) under commission
recorded in the Fihrist that they were by the Barmakid Ya yā b. Khālid (Hārūn’s
involved in translating books from Persian vizier from 169/786 to 187/803). In a
into Arabic, as was the case with Abū colophon of a manuscript containing the
Sahl al-Fa l b. Nawbakht (274) and Salm translation of the earliest extant Arabic
(120); the latter is also mentioned as hav- paraphrase of Aristotle’s logic, Salm is said
ing prepared, like Ibn al-Muqaffa and also to have been involved, again together
Sahl b. Hārūn—both well known shu ūbīs with others, in its translation for Ya yā
and Middle Persian experts—summaries b. Khālid (Kraus, 1–20). The capacity in
and extracts of Kalīla wa-Dimna (305). It is which Salm was involved in these projects
thus clear that the function of this library is not clear. It is improbable that he knew
under the early Abbāsids was similar to Greek or Syriac, in which case either the
that under the Sāsānians, that is, the pres- translations of both these works were done
ervation of the Persian heritage, although from Middle Persian versions, or Salm per-
now in Arabic translations, to which there haps merely edited or polished the versions
was apparently added the corresponding prepared by the translators from Greek or
function of collecting and preserving Arab Syriac. In any case, the available evidence
traditions: in addition to “old” books from does not indicate that these projects took
the pre-Islamic and early Arabian tradition place in the Bayt al- ikma as part of its
mentioned above were books on Arab his- regular activities; the mention of the Bayt
tory and warfare said to have been commis- al- ikma in these instances is merely in the
sioned by al-Man ūr (Gutas, Greek Thought, title identifying Salm. A similar reference
57 n. 49), and Hārūn al-Rashīd is imagined to Salm as part of a committee sent by
136 bayt al- ikma

al-Ma mūn to Byzantium to collect Greek that were widely read as authoritative in
manuscripts (243) is to be discounted as part because of the author’s expertise in
legendary. Salm was in charge of the the history of Arabic medicine—Meyerhof
Abbāsid palace library under Ya yā b. repeated and elaborated this imaginative
Khālid al-Barmakī, as is attested not only interpretation of the Bayt al- ikma as a
in the Fihrist but also, independently, in Ibn full-fledged academy and institute of trans-
Abd Rabbihi’s (d. 328/940) al- Iqd al-farīd lation, founded by al-Ma mūn in 830 or
(ed. A mad Amīn, A mad al-Zayn, and 832, where all the Greek manuscripts of
Ibrāhīm al-Ibyārī, Cairo 1940–53, 2:127, the caliph were kept and in which a team
where the printed name Sulaymān is of translators worked under the direction
clearly an error for Salm); and the Barma- of unayn b. Is āq. Very nearly the same
kids had already been removed from power picture of the Bayt al- ikma appears in
ten years before al-Ma mūn’s accession. Dominique Sourdel’s article in EI 2. Later
Furthermore, this report in the Fihrist is publications describe the Bayt al- ikma
part of the fictitious account that credits as a full-fledged college of the sciences or
the translation movement to al-Ma mūn’s humanities. There is, however, no evidence
dream of Aristotle (Gutas, 95–104). for these assertions, and the brief facts
It appears that it was from reports mentioned above constitute almost the
such as these concerning Salm that these entirety of the information that we possess
arose in twentieth-century scholarship the on the subject.
myth that the Bayt al- ikma of the early
Abbāsid caliphs was an academy and a Bibliography
school for the study of the ancient sciences, Jawād Alī, Mawārid Ta rīkh al- abarī II,
and a centre for the translation of Greek Majallat al-Majma al- Ilmī al- Irāqī 2 (1951),
works into Arabic in which unayn b. 135–90; Paul Kraus, Zu Ibn al-Muqaffa ,
RSO 14 (1934) 1–20; repr. in his Alchemie,
Is āq (d. 260/873) was active and which Ketzerei, Apokryphen im frühen Islam, ed. Rémi
was founded by al-Ma mūn in 217/832. Al- Brague (Hildesheim 1994), 89–108; Max
though there was some earlier speculation Meyerhof, New light on Hunain ibn Ishaq
about it, it was DeLacy O’Leary’s Arabic and his period, Isis 8 (1926), 685–724, repr.
in Fuat Sezgin (ed.), Galen in the Arabic tradi-
thought and its place in history (London 1922) tion. Texts and studies (Frankfurt 1996), 3:1–40;
that first linked the Bayt al- ikma with the Franz Rosenthal, From Arabic books and
Nestorian physicians of Baghdad (including manuscripts, XVI. As-Sarakhsī (?) on the
unayn), asserted that it was founded by Appropriate Behavior for Kings, JAOS 115/1
(1995), 109a; Gregor Schoeler, Arabische Hand-
al-Ma mūn, and gave 832 as the official schriften (Wiesbaden 1990), 2:308; Mansour
date of its establishment—all without Shaki, The Dēnkard account of the history
citing any sources. Following O’Leary’s of Zoroastrian scriptures, ArO 49/2 (1981),
unfounded assertions, and inspired by 114–25.
For critical discussions of the evidence see
G. Bergsträsser’s publication (Leipzig 1925) Marie-Geneviève Balty-Guesdon, Le Bayt al-
of unayn’s bibliographic Risāla of Galenic ikma de Baghdad, Arabica 39 (1992) 131–50;
translations, Max Meyerhof published an Françoise Micheau, The scientific institutions
article that was responsible for the propa- in the medieval Near East, in Encyclopedia of
the history of Arabic science, vol. 3, Technology,
gation of this myth (Meyerhof, 685–724). alchemy, and life sciences, ed. Roshdi Rashed
In this and subsequent publications in (London and New York 1996), 986–8; P. S.
German, English, and French—articles van Koningsveld, Greek manuscripts in the
bookbinding 137

early Abbasid empire. Fiction and facts about in leather cases with a vertical opening
their origin, translation, and destruction, BO (Illustration 1). In the Maghrib and other
55/3–4 (1998) 345–71; Dimitri Gutas, Greek
thought, Arabic culture. The Graeco-Arabic transla- parts of Africa, Qur ān manuscripts as well
tion movement in Baghdad and early Abbāsid society as the most popular prayer books were
(2nd–4th/8th–10th centuries) (London and New kept in leather bags that were decorated
York 1998), 53–60. or covered with embroidered velvet. An
Dimitri Gutas – Kevin van Bladel ornamental cord made the bags easier to
transport.
We can distinguish two types of binding
Bookbinding used in the Muslim world. The first was
used for Qur ān manuscripts in an oblong
Several types of bookbinding, Ar. tasfīr format written in the varieties of script
or tajlīd, were used in the Islamic world. collectively known as Kufic and produced
The art is known through a number of until the fifth/eleventh century; these
technical treatises but mainly via the study manuscripts were encased in a way that was
of surviving examples, the dating and lo- apparently distinctive to copies of the
calisation of which are complex matters, Qur ān. Georges Marçais and Louis Poinssot,
given that the time at which a book was in their 1948 study, described bindings of
bound is not necessarily the same as when this type found in al-Qayrawān. Other dis-
it was written. Not all Islamic manuscripts coveries followed, associated with Qur ān
were bound, and in cases in which binding manuscripts preserved in Damascus (but
did not occur shortly after copying, a bind- moved to Istanbul) and in an ā . This
ing may have been added upon acquisition type, thought to resemble Coptic bindings
or donation to an institution, or to replace from the second/eighth to the third/ninth
a damaged binding; re-use of bindings also century, was apparently known throughout
occurred quite frequently. the Islamic world, as the geographically
These bindings, meant mainly to protect, wide-ranging examples suggest. With this
covered the most precious manuscripts, and sort of binding, the wooden covers were
the ornamental possibilities were developed encased in leather and attached to the text-
rapidly. In an inventory of al-Qayrawān block by sewing strings that held the
Mosque, in modern-day Tunisia, drawn up quires together or by being glued to the
in 691–2/1292–3, we find descriptions of end sheets. A rim or wall, also of leather,
Qur ān bindings, but nothing is mentioned enclosed the box on the three open sides;
of bindings for other texts. Indications in a leather strap attached to the lower board
other documents suggest, to the contrary, was wound around a hook fixed to the edge
that some manuscripts were not bound at of the upper cover to close the volume.
all. Several techniques were used in the The inside of the cover was lined with
Maghrib and the Middle East. With the parchment. The covers were decorated
exception of Christian-Arabic manuscripts, by means of stamping or by a technique
some of which were bound in the Byzan- using strings, a method of ornamentation
tine tradition, the techniques used in the also found on bindings used for Coptic
Muslim world differ from those employed and Syriac manuscripts. With the string
in the West. The bindings, themselves pre- technique, a string was first inserted into
cious and fragile, were at times inserted a groove carved into the wooden board