The Conflict of Interpretations and the Limits of Pluralism

Author(s): Paul B. Armstrong

Source: PMLA, Vol. 98, No. 3 (May, 1983), pp. 341-352
Published by: Modern Language Association
Stable URL: http://www.jstor.org/stable/462275
Accessed: 27/07/2010 22:45

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/action/showPublisher?publisherCode=mla.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact [email protected].

Modern Language Association is collaborating with JSTOR to digitize, preserve and extend access to PMLA.

http://www.jstor.org
 PAUL B. ARMSTRONG

The Conflictof Interpretationsandthe Limitsof Pluralism

C AN ANY interpretation lay definitive main objectives are, first, to explain why permis-
claim to correctness? This question has sible readings may differ and, second, to show
occasioned one of the most important that criteria for validity nonetheless act as con-
disputes in literary theory over the last several straints on acceptability and regulate claims to
years. Sharp disagreement divides those who legitimacy even when unresolvable conflicts di-
argue that interpretation is limitless from those vide interpretations.
who hold that textual meaning is singular and
ultimately discoverable. Taking as their slogan I
Nietzsche's contention that there is no truth but
only an array of interpretations, the radical To discover why permissible readings can dis-
relativists insist that any work allows innumer- agree, we must go back to the foundations of
able readings. In its most extreme form, this interpretation and examine the role of belief in
argument asserts that all interpretations are understanding. The first premise of hermeneutics
necessarily misinterpretations-that no criteria is that interpretation is basically circular. The
exist, within a text or outside it, for judging any classic formulation of the hermeneutic circle
reading the "right" one. (I have in mind, ob- holds that we can only comprehend the details
viously, the Yale deconstructionists and their of a work by projecting a sense of the whole, just
mentor, Jacques Derrida, but Norman Holland as, conversely, we can only achieve a view of the
and Stanley Fish hold similar views.)' The whole by working through its parts. All interpre-
anarchistic, nihilistic implications of this position tation consequently requires acts of faith-
are more disturbing than the often playful, puck- beliefs that compose parts into a whole, hy-
ish attitude of its adherents might suggest. The potheses that we check, modify, and refine by
dangers of denying interpretive limits have moving back and forth between aspects of any
prompted others to insist that meaning is deter- state of affairs and our sense of its overall con-
minate. The monists support their claim variously, figuration. Hence Spitzer's assertion that inter-
with appeals to the author's intention, norms in pretation depends on an "inner click"-a "divi-
the work itself, or plain old common sense (see nation" of the relation between part and whole
Hirsch, Validity; Wellek; and Reichert, Making (7, 19). Hence, too, Iser's description of read-
Sense). But they are united by their opposition ing as a process of "consistency-building," an
to a pluralistic view of interpretation that would ongoing quest for patterns that establish co-
allow for different, equally "correct" readings. herence among the elements of a text ("Reading
The rigidity of the monists is as unacceptable, Process," esp. 283-90; see also Iser's Act 118-
however, as the nihilism of the radical rela- 34). From the title page on, we ceaselessly
tivists. Neither position can account for the and silently use the details to project hypotheses
paradox that characterizes the actual practice of about the whole, conjectures that are at first
our discipline: we have legitimate disagreements vague and provisional. Then we employ these
about what literary works mean, but we are also guesses to make sense of the work's parts-just
able to say that some readings are wrong, not as everything new we come across helps us to
simply different. Contemporary criticism needs a refine and amplify our overarching construct (or
theory of limited pluralism to explain this para- leads us to overturn it if anomalies persistently
dox and to chart a middle way between the crop up and the parts refuse to fit).
anarchists and the absolutists. In what follows I This version of the hermeneutic circle has
try to sketch the outlines of such a theory.3 My three important implications for the relation be-

341
342 The Conflict of Interpretations and the Limits of Pluralism
tween theory and practice. First, because inter- involves beliefs more fundamental than the
pretation always requires guesswork, no rules hypotheses that align parts into wholes. Every
can guarantee successful hypotheses in advance. interpretive approach has its own anticipatory
Even the most sophisticated theorists and the understanding of literature, one that reflects its
most practiced critics have stared blankly at a most basic presuppositions. Phenomenology al-
page, waiting for its configurations to suggest ready sees human being as an incarnate con-
themselves. Beginning students may dream of sciousness directed toward objects, and so it in-
one day becoming expert enough to comprehend terprets works as constructs of consciousness
a novel or a poem automatically-without the that display a world. Structuralism grasps
hesitancy, confusion, and uncertainty of experi- human being in advance as a mind governed by
menting with guesses-but the more experienced a linguistic logic of binary oppositions, and so it
they become as interpreters, the more they re- construes myths and other texts as logical mod-
alize that exegesis cannot escape trial and error. els that attempt to resolve contradictions. As
This observation suggests my second point: a Bultmann notes, "All understanding, like all in-
theory of interpretation is not a machine for terpretation, is . . . continually oriented by the
cranking out readings. Practitioners of any manner of posing the question and by what it
method must start anew and try out guesses aims at [by its Woraufhin]. Consequently, it is
every time they take up a work. Experience never without presuppositions; that is to say, it
teaches because past acts of interpretation pro- is always directed by a prior understanding of
vide practice in guessing. But different texts the thing about which it interrogates the text"
demand different hypotheses. My third point, (239; trans. slightly modified).'' The character-
consequently, is that no theory of interpretation istic hypotheses projected by a method of inter-
can guarantee persuasive, effective readings.4 pretation are practical embodiments of more
Any method, no matter how promising, can lead basic beliefs about human being, the being of the
to more or less convincing interpretations. object interrogated, and the being of the world
Heidegger reformulated the hermeneutic cir- as a whole. Psychoanalysis, Marxism, phenom-
cle to bring out its inherent temporality. As he enology, structuralism-each has a different
explains in Being and Time, understanding re- method of interpretation because each has a
quires expectations. We can only interpret some- different metaphysics, a different set of convic-
thing, he argues, if we have already grasped it tions that make up its point of departure and
through a "fore-seeing" (Vor-sicht) that pro- define its position in the hermeneutic field. A
jects and delimits a range of meanings it might Freudian, for example, who believes that human
have. Our interpretations turn these possibilities beings are sexual animals and that literary works
into actualities. To interpret is thus to lay out are disguised expressions of repressed libidinal
(Aus-legen is Heidegger's term) an anticipatory desires, will arrange textual details in configura-
understanding that has cleared the way for tions that differ from those of a Marxist critic,
fuller, more explicit, and more refined acts of who believes that human beings are social, his-
construal (188-95). Phrased in the traditional torical creatures and that art reflects class inter-
language of the hermeneutic circle, Heidegger's ests. To embrace a type of interpretation is to
point is that our preliminary sense of the whole make a leap of faith by accepting one set of
gives us a particular set of expectations that then presuppositions and rejecting others.
direct our attention and that the subsequent ex- My illustrations so far have been kinds of
plication of details checks, modifies, and fills in. interpretation that were originally extraliterary,
To project a hypothesis is to anticipate a pos- and I have chosen them because they clearly
sible future. The surprise we sometimes experi- demonstrate that a method's practical her-
ence in reading illustrates Heidegger's argument. meneutic hypotheses reflect deeper metaphysical
We would not have been surprised if we did not convictions. But my argument holds as well for
have expectations, a prior understanding that methods that seem purely literary-like, say, the
turned out to be unreliable. New Criticism. Although the New Critics advo-
Heidegger's notion of anticipatory under- cated close reading free of preconceptions and
standing suggests that making sense of literature usually eschewed philosophical speculation,
 Paul B. Armstrong 343
sticking instead to the concrete requirements of own dialectic of blindness and insight-a ratio
explication, it is now generally acknowledged of disguise and disclosure that stems from its
that their interpretive practice rested on a spe- presuppositions. To accept a method of interpre-
cific set of beliefs about the human world.6 tation is to enter into a wager-to gamble,
For example, although Brooks's contention namely, that the insight its assumptions make
that "the language of poetry is the language of possible will offset the risk of blindness.7
paradox" (1) may seem a purely literary state- Conflicting modes of interpretation can be
ment summarizing the results of numerous ex- classified by what they aim at and how they take
egeses, his poetics actually elaborate principles their aim. For example, Paul Ricoeur divides the
laid down in Eliot's famous essay "The Meta- hermeneutic field into "archaeological" and
physical Poets," which claims that "a degree of "teleological" methods (Conflict 21-22).
heterogeneity of material compelled into unity Archaeological interpretation is a hermeneutics
by the operation of the poet's mind is omnipres- of unmasking. For approaches of this kind-
ent in poetry" (243). For Brooks paradox is which include psychoanalysis, Marxism, and
the poetic vehicle that molds disparate, often structuralism-meaning is never on the surface;
contradictory materials into oneness. But for rather, the surface is a disguise, a mask that
Eliot poetry's powers of integration have more must be demystified to uncover the meaning be-
than aesthetic significance; he invokes them to hind it. The rule for reading is suspicion. For
counter a general cultural crisis caused by the teleological approaches, such as phenomenology
prevalence of fragmentation, dissociation, and and the New Criticism, the rule is trust. Meaning
unassimilated complexity in all spheres of life. is to be found not behind the text but beyond-
Brooks's masterful analyses of poetic paradox in the goals, possibilities, and values that literary
are not purely literary, then; they carry a hidden works and other cultural objects testify to or try
freight of ideological commitment insofar as his to point toward. The appropriate interpretive at-
allegiance to unity in multiplicity endorses El- titude is therefore not suspicion but openness to
iot's views about modernism. This state of affairs revelations.
is unavoidable, however, because there is no in- There are further divergences between rival
terpretation without presuppositions about the hermeneutics according to the kind of arche (or
being of the work and of the world. If we seek to origin) and telos (or end) that they attempt to
understand without preconceptions, we do not disclose. Among the great modern practitioners
escape them; we reproduce them in our inter- of suspicious interpretation, Nietzsche unmasks
pretations, but without recognizing them for texts and institutions to show the will to power
what they are-our own assumptions, not inde- they disguise; Freud unearths the regressive pull
pendent facts in the text. (Heidegger makes a of infantile fixations and unconscious desires;
similar point [191-92].) Marx demystifies the seemingly independent
The presuppositions of any interpretive cultural superstructure to uncover its origins in
method are both enabling and limiting. They the economic base. Similarly, although phe-
give us a vantage point from which to construe nomenology and the New Criticism both ap-
the work-a specific place of observation, with- proach a work more to reveal its values than to
out which knowing would be impossible. They disclose its deceptions, they differ radically in
also furnish a set of expectations with which to their understanding of the object of interpreta-
pose questions to a work that would otherwise tion. For the New Critic, the literary work is a
remain silent, and they provide guidance and self-sufficient structure of norms that an inter-
inspiration as we begin to make guesses. But pretation may approximate but cannot fully re-
presuppositions are at the same time limiting be- alize. For the phenomenologist, the work is not
cause in opening up a work in a particular way so much an objective structure as a meeting of
they close off other potential modes of access. subjectivities-the consciousness of the reader
Every interpretive approach reveals something bringing to life authorial acts of consciousness
only by disguising something else that a compet- that lie dormant in the black marks preserving
ing method with different assumptions might them.
disclose. Every hermeneutic standpoint has its Interpretations conflict with each other when
344 The Conflict of Interpretations and the Limits of Pluralism
they embody opposing presuppositions. To what philosophically rigorous, methodologically self-
extent, then-if at all-can their disagreements conscious efforts to amalgamate opposing strat-
be resolved? We should distinguish first between egies. A weak, watered-down criticism may re-
"weak" and "strong" disagreements. When the sult. The most powerful approaches to
opposing critics operate within a shared body of interpretation often owe the depth of their in-
assumptions (two Marxists, for example, or two sight to the radical one-sidedness of their beliefs.
phenomenologists), their disagreement reflects For example, after describing the potential for
the heterogeneity of their interpretive school, reductionism in psychoanalytic criticism,
different ways of taking advantage of the possi- Ricoeur nonetheless warns that "Psychoanalysis
bilities made available by the same assumptions, cannot be reproached for [its] narrowness; it is
or a better as opposed to a worse application of its raison d'etre" (Conflict 14; page references in
the same procedures. Their battles may be next two paragraphs also refer to Conflict).
vociferous, but I call their disagreement "weak" Among contemporary theorists of her-
because it is not a dispute about fundamental meneutics, Ricoeur has in fact made the most
convictions. rigorous, sustained attempt to achieve what he
"Strong" disagreements may begin with dif- calls "a true arbitration among the absolutist
ferences about how to construe a particular text, claims" of opposing systems of interpretation
but they ultimately go back to divergences be- (15).8 He seeks to show that the theory behind
tween the basic presuppositions underlying the each method is justified by the particular area of
opposing methods. Structuralism and Marxism human experience it singles out as its special
differ in their approach to Greek literature, for province. In his view, "each hermeneutics dis-
example-Levi-Strauss viewing Oedipus as an covers the aspect of existence which founds it as
attempt to resolve a logical contradiction be- method" (19). We can then reconcile opposing
tween two explanations of human reproduction methods, he argues, by resolving their theories
(210-13) and Marx regarding classical myth- into a unified image of human being-that is, by
ology as an effort to establish an imaginary showing how the different modes of existence
mastery over nature when backward economic they focus on belong to the "coherent figure of
conditions prevent material control (149-51). the being which we ourselves are" (23).
But this disagreement between interpretive strat- Ricoeur's project suffers from several prob-
egies reflects a more fundamental conflict of lems, however, which may explain why he de-
belief-between the Marxist's conviction that scribes his dream of reconciliation as a "prom-
human beings are social creatures whose nature ised land" that the philosopher, "like Moses,"
changes with their daily practice over history can only "glimpse. . . before dying" (24). The-
and across cultures and the structuralist's ories of interpretation not only disagree about
assumption that human beings are linguis- the defining aspect of human beings (who are
tic animals defined by their unchanging capacity variously viewed as primarily linguistic, psycho-
to order the universe by binary oppositions logical, or social) but also hold conflicting, ir-
(Levi-Strauss holds that all versions of the reconcilable beliefs about the same aspect of
Oedipus story, from Sophocles' play to Freud's existence (witness the disagreement among
theory, are variants of the same mythic structure Freud, Jung, Lacan, and Sartre, who respec-
because all center on the same contradiction). tively regard the unconscious as libidinal, arche-
An eclectic response to hermeneutic conflict typal, linguistic, or nonexistent). And even if a
ignores the implications of "strong" disagree- method does elevate one aspect of existence to a
ment. A critic who borrows freely from many privileged status, its assumptions about that area
different methods runs the risk of introducing may have implications for some other area that
self-contradictions into his or her assumptions. are incompatible with the presuppositions of an-
Some presuppositions exclude each other, as the other hermeneutic. Structuralism's views about
dispute between structuralism and Marxism language and Marxism's beliefs about society,
shows. Or, when not mutually exclusive, the for example, lead to irreconcilable disagreement
assumptions of two approaches may harmonize over whether human nature is fundamentally
more or less well. A danger threatens even universal or radically historical. Furthermore,
 Paul B. Armstrong 345
theories of interpretation sometimes resist con- is not), so that the "identity" of "a work" is
finement within the limited realms of relevance typically a multiplicity correlated to the many
to which Ricoeur would restrict them. Consider, possible ways of construing "it."
for example, the prolific writings in which Freud Wayne Booth makes a proposal that seems
applies his psychological assumptions to politics, more modest than Reichert's: "It would no
art, anthropology, and religion. Such encyclo- doubt be impossible to determine precisely how
pedic explanatory ambitions are only to be ex- much we share in the texts we dispute about, but
pected, inasmuch as any presuppositions about we surely could not go on disputing at all if a
human being, even if made at first in a limited core of agreement did not exist" ("Limits"
realm, have wide-ranging metaphysical ramifica- 412-13).10 The metaphor of a shared "core"
tions. is unfortunate, however, because it implies an
Merleau-Ponty claims that "All . . . views are autonomous essence of textual identity. We need
true provided that they are not isolated" and not assume radical textual sameness, though, to
argues that "We must seek an understanding explain the possibility of literary discussion. To
from all . . . angles simultaneously" (xix). In exchange views, we only need points of com-
criticizing Ricoeur's dream of hermeneutic unifi- parison and contrast, of overlap and divergence;
cation, however, I have tried to show that some the name of the text, for example, and substan-
"truths" cannot be brought into agreement, be- tial agreement about the register of characters,
cause they rest on irreconcilable presupposi- the basic elements of the plot, and the language
tions. Some methods of interpretation are, of are all that we require to debate different con-
course, more compatible than others. Indeed, an ceptions of Hamlet. There are usually areas of
important project for a philosophy of interpreta- agreement between even the most widely diver-
tion would be to investigate the extent to which gent interpretations, but these only establish the
hermeneutics can converge because their pre- possibility of discussion; they are not conclusive
suppositions are reconcilable-and to identify evidence of the text's autonomous essence. I can
the lines across which antagonists face each recognize a novel or a poem in a botched student
other without the possibility of merger. But we essay, for example, but that does not mean that
cannot arrive at "the truth" about the meaning the student and I fundamentally agree about the
of a literary work by reconciling opposing posi- text's meaning or that we both see the "same"
tions in the conflict of interpretations. text.
One frequently encountered rejoinder from Some who acknowledge the possibility of mul-
the monists is that differing interpretations must tiple interpretations consider it to be a distin-
have some degree of identity because they derive guishing feature of art. The attempt to identify
from the same text. John Reichert insists, for the being of art is notoriously difficult, however
example, that "multiple interpretations do not -not least because competing hermeneutics in-
multiply the number of things which they are sist on incompatible definitions."1 Interpretive
interpretations of" ("Reply" 166). He is right, disagreement is not restricted to aesthetics,
of course; but he is also wrong, because he mis- though, and art is not unique in lending itself to
takes the mode of existence of the literary work. a variety of readings. Disciplines can be clas-
A text is not an independent object that remains sified, in fact, according to whether they tend
the same regardless of how it is construed. It is toward monism or pluralism. Some fields-most
not autonomous but "heteronomous." While a of the natural sciences, for example-have a
literary work transcends any individual interpre- high degree of unanimity about permissible as-
tation, it exists only in and through its "con- sumptions and desirable explanatory goals, but in
cretizations"-so that it will cease to exist in other disciplines-including psychology, eco-
any meaningful sense if it is no longer read.9 nomics, and philosophy, in addition to literary
Different interpretations concretize the work studies-basic disagreements about such matters
differently, and its identity is the synthesis of lead to fundamental differences of interpreta-
changing construals across history and over the tion. The inability to reconcile opposing inter-
field of conflicting modes of understanding-a pretations is a basic fact of professional and
synthesis that may not be complete (and usually pedagogical life in the humanities, but it is a
346 The Conflict of Interpretations and the Limits of Pluralism
problem with broader epistemological and insti- and-fro movement between a guess about the
tutional horizons.12 whole and the parts it seeks to make sense of
need not become a vicious circle if we regard
II our hypotheses as at most provisional and re-
If the differences between rival readings cannot main open to indications that they need revision.
always be resolved, we are left with a number of According to the test of inclusiveness, a hy-
troubling questions about validity. Must we re- pothesis becomes more secure as it demonstrates
sign ourselves to relativism-with all interpreta- its ability to account for parts without en-
tions considered equal and with no grounds countering anomaly and to undergo refinements
available for choosing among them? Are there and extensions without being abandoned. The
no standards of correctness to distinguish be- interpretive act may be "irrational" in the sense
tween legitimate and mistaken readings? (My that it requires inspired guessing, but it is "ra-
students' version of this issue is, What entitles tional" in the sense that our hypotheses must be
Mr. Armstrong to put grades on our papers?) able to withstand self-conscious examination.
These questions restate the traditional philo- Although an interpretation that is not inclu-
sophical concern that the hermeneutic circle may sive is illegitimate, an inclusive reading is not
be a vicious one. A self-confirming circularity necessarily the only "correct" one. The elements
would seem to threaten belief not only in the of a text can be construed and made coherent in
hypotheses that align parts into a coherent whole a variety of ways. Different interpretive methods
but also in the presuppositions that define any based on different presuppositions can pass the
hermeneutic standpoint. How can we avoid get- test of inclusiveness with equal success. The
ting trapped in a vicious circle where our hy- absolutist E. D. Hirsch and the relativist Stanley
potheses about the whole are vindicated by Fish agree about very little, but they do agree
evidence that they themselves have shaped in about this point. Hirsch argues that the standard
making sense of the parts? If any type of inter- of inclusiveness "cannot, in fact, either reconcile
pretation can only discover what its presupposi- different readings or choose between them. As
tions open up to its view, then how can we be a normative ideal, or principle of correctness, it
sure that these beliefs are trustworthy? is useless" (Validity 227). Hirsch exaggerates,
In answering these questions, I will argue that however. The test of inclusiveness is useful for
we commonly invoke tests for validity that act as excluding bad guesses, even though it cannot
constraints on interpretation and mark a bound- conclusively settle all conflicts between inter-
ary between permissible and illegitimate read- pretations or put an end to "strong" disagree-
ings. But I will also demonstrate that the limits ment. Fish claims that any interpretive frame-
of these tests prevent them from establishing any work can find a way of accounting for an
single interpretation as the "right" one. I con- apparent anomaly. Whenever someone confronts
tend, in other words, that literary criticism is him with a counterexample, he explains, he
inherently pluralistic but that it is nonetheless looks "immediately for ways to demystify or de-
what Stephen Toulmin calls "a rational enter- construct it" and to interpret it in a manner con-
prise," with standards and restrictions built into sistent with his presuppositions-and, he says
its proceedings, not a field of anarchistic free modestly, "I always succeed" ("Reply" 178).
play where anything goes (133-99).13 Fish does not realize how close he is here to
The first test for the validity of a reading is vicious circularity, but his boast shows that the
inclusiveness. If understanding is a matter of fit- test of inclusiveness cannot produce a single,
ting parts into a whole, then that belief about definitive interpretation.
their relations will be superior which can en- A second test-intersubjectivity-can supple-
compass the most elements in the configuration ment and strengthen the standard of inclusive-
it projects. A part that refuses to fit is a falsify- ness, but it too is of limited effectiveness. As I
ing anomaly. Critics and teachers invoke the have explained at length, belief plays an integral
standard of inclusiveness, for example, when role in understanding. We should therefore re-
they praise a reading for its "scope" and member the lines from Novalis that Conrad
"depth" (or damn it for lacking them). The to- chose as the epigraph for Lord Jim: "It is cer-
 Paul B. Armstrong 347
tain my conviction gains infinitely, the moment sonableness of a presupposition, the ability of a
another soul will believe in it" (ix). Since inter- set of assumptions to rally a community of be-
pretation is essentially an act of belief, our read- lievers to its side is a test of its probable worth.
ing becomes more credible if others assent to it Conversely, assumptions that cannot win en-
or at least regard it as reasonable. Conversely, dorsement probably deserve to die. Within a
the disagreement of others may signal that our community of believers, the faithful accept,
interpretation is invalid because unsharable. As reject, and rank interpretations (psychoanalytic
Peirce points out, "Unless we make ourselves critics, for example, passing judgment on one
hermits, we shall necessarily influence each oth- another's work on the basis of how well each
er's opinions; so that the problem becomes how puts their shared assumptions into practice).
to fix belief, not in the individual merely, but in Across the boundaries between communities,
the community" (235). combatants attempt to prove the merits of their
The weakness in the test of communal agree- beliefs by winning converts; and when one side
ment hinges on the need to persuade others to recognizes the other as a worthy opponent, it
assent to our convictions about the meaning of a acknowledges the credibility of the alternative
text or about the best way to understand litera- position (structuralism, for example, implicitly
ture. Intersubjective validation necessarily en- grants the legitimacy of phenomenology by en-
tails the use of rhetoric. As an act of power tering into a serious debate with it). Thus, while
designed to move others in a certain direction, monistic agreement about the truth does not re-
rhetoric is open to abuses that may undermine sult from this second test for legitimacy, we see
the fair application of credibility as a test for again that accepting pluralism need not mean
legitimacy. In a modern restatement of the clas- abandoning all standards and procedures for
sical concern about the potential distortions of validation.
rhetoric, Jauss distinguishes between Uberzeu- A third test is efficacy-the evaluation of a
gen, the genuine work of convincing others that hypothesis or a presupposition on pragmatic
our reasons and beliefs deserve credence, and grounds, to see whether it has the power to lead
Uberreden, the use of tactics to overpower or to new discoveries and continued comprehen-
trick others into granting assent (1:50-51). sion. In one sense this test simply restates
The line between these two methods is often diachronically what the test of inclusiveness de-
hard to draw, but it stands as a warning that scribes synchronically. We should doubt the
communal agreement does not always prove efficacy of a hypothesis if repeated anomalies
validity. interrupt the progress of interpretation, since
If the ability to win assent suggests that a such blockages suggest that our guess is not in-
belief is legitimate, then complete and universal clusive. In another sense, though, the test of
agreement should be the ultimate indication that efficacy introduces something new and impor-
an interpretation is correct. This reasoning, in tant because it extends as well to the most fun-
any case, underlies Spitzer's contention that the damental presuppositions of an interpretive
proper goal of literary criticism is a consensus method. William James argues that "we have the
omnium-a state of perfect intersubjectivity, right to believe at our own risk" any assumption
with all minds at one about the meaning of a "that is live enough to tempt our will" (Will
work (38). This ideal is not finally attainable, 32). But he also warns that the actions ensuing
however, because conflict between irreconcilable from our convictions invariably have conse-
methods prevents unanimity. "Strong" dis- quences and that these consequences may re-
agreement about the fundamental presupposi- bound to cast doubt on what we believe. The
tions that should guide understanding stands in presuppositions on which any hermeneutic takes
the way of universal consensus. its stand are not immune from practical testing.
This impasse introduces some interesting They must continually justify themselves by
complications to the test of intersubjective their efficacy. If they repeatedly fail to lead to
agreement, but it does not eliminate the criterion persuasive, inclusive readings, we may conclude
altogether. Although a majority vote by itself that the problem lies not with the interpreters'
does not decisively prove or disprove the rea- limited skills but with the assumptions them-
348 The Conflict of Interpretationsand the Limits of Pluralism
selves. We put our presuppositions into play greater claim to validity because of the past per-
when we follow their guidance in interpreting a formance or the specialized training of those
work. But we also put them at risk, inasmuch as who hold them. Teachers, for example, who ex-
our encounter with a text may defy our expecta- ercise the authority of grading student papers,
tions and challenge our assumptions. have successfully passed through apprenticeships
The importance of holding ourselves open to (taking oral exams, writing dissertations) that
experiences that may prove our presuppositions have judged them qualified to receive the rights
ineffective leads Peirce to warn against the pre- and powers of professional responsibility. This
mature, dogmatic fixation of belief. He describes claim to authority must continue to prove itself
three roads to dogmatism that deserve attention in the classroom and in various professional
because they provide a negative counterpart to forums or else students and colleagues may deny
the three positive tests for correctness-a set of it. As Toulmin points out, the institutional au-
"Thou shalt not"s to accompany the inter- thority granted leading members of a profession
preter's affirmative obligations in the pursuit of is a provisional mandate that they receive (and
legitimate readings. Still, despite the value of can forfeit) because of their intellectual au-
these warnings, they need to be qualified in ways thority in the field (261-81).14
that will, I think, shed further light on the rela- The match between institutional authority
tion between belief and validity. and intellectual justification is not always per-
Peirce warns first against what he calls "the fect, but institutional sanction does not have
method of tenacity"-"a steady and immovable the exclusively deleterious effect that Peirce as-
faith" that holds stubbornly to its convictions in signs to it. While it may discourage potentially
spite of any and all evidence against them (235, promising but renegade innovations, its con-
234). This method runs the risk of falling into servative force also prevents an excessively rapid
the double trap of solipsism and a self-confirm- overthrow of methods of understanding that
ing, vicious circularity. It disregards dissent have established their worth by their efficacy and
from the community and blinds itself to poten- their success in winning adherents. Authority
tially falsifying anomalies. But Peirce's warning may rigidify into dogmatic tyranny, but it wards
must be modified, for unless we hold to our be- off the anarchy of perpetual revolution.
liefs with a certain amount of tenacity we may Finally, Peirce's strictures against "the a
abandon a promising hypothesis prematurely, priori method," though they, too, provide a use-
before giving it a fair chance by struggling to ful warning, overstate the case. This method de-
solve the problems it raises. Similarly, to give fends its fundamental assumptions by arguing
our hypotheses and assumptions a just trial, we that they are "agreeable to reason." Its standard
must show a degree of persistence in defending of judgment is not "that which agrees with ex-
them against opposing readings and rival ap- perience, but that which we find ourselves in-
proaches. Interpretation requires a delicate bal- clined to believe" (238-39). Peirce protests that
ancing act between excessive stubbornness and such a procedure removes belief from the testing
overhasty capitulation. ground altogether, allows the interpreter uncon-
Peirce's warning against "the method of au- trolled license, and reduces inquiry to a matter of
thority" stands in need of similar amendment. "taste" and "fashion." Nevertheless, there is
According to Peirce, this method settles opinion something a priori about accepting any set of
by appealing to institutional sanction-to what presuppositions as our hermeneutic point of de-
the leaders of the community and its accepted parture. In deciding where to place our al-
ways of understanding allow to be true. The legiances as critics, we must choose among a
danger here is a tyrannical communal solipsism variety of equally defensible alternatives. Clear-
that makes itself immune to the challenge of dis- headed thinking or an appeal to common experi-
sent and denies opposing views an unbiased ence will not in itself show us the one right road
hearing (235-38). But once again authority has to take. Although the presuppositions and the
benefits that Peirce overlooks. In the business results of any method must prove their worth
of interpretation, not all practitioners are equal. in the ways I have described, the choice of a
Some views are legitimately presumed to have a hermeneutic standpoint is inherently somewhat
 Paul B. Armstrong 349
arbitrary. A different decision can always be cause our critical commitments can be analyzed
justified with equal cogency. and debated. If every method discloses some
By calling our choice of hermeneutic al- things at the cost of disguising others, then the
legiance "arbitrary,"I do not mean that it does merits and risks of its hermeneutic "wager" can
not matter. I mean only that since it is a choice be examined and discussed. A critic may prefer
we always have alternatives. Obviously, because some kinds of insight and tolerate some areas
of its wide-ranging epistemological and meta- of blindness; once again, what ratio between the
physical implications, the choice is deeply sig- two to accept is a matter of choice, but not of
nificant. In choosing a hermeneutic standpoint, unreasoned choice. Stanley Fish may be right
we decide how we are going to conduct our- that "one man's reason is another man's ir-
selves, with what kinds of objects and aims, in relevance" ("Reply" 175).16 But interpreters
what sort of critical universe. Even the monist are accountable to the community for the conse-
Hirsch concedes that "The choice of an inter- quences of their commitments, and some wagers
pretive norm is not required by the 'nature of the stand up to critical public scrutiny better than
text,' but, being a choice, belongs to the domain others do. The beliefs that constitute the theory
of ethics"; he insists, however, that "Unless implicit in any method may be ethical a prioris,
there is a powerful overriding value in disregard- but they must still attempt to justify themselves
ing an author's intention (i.e., original mean- through public debate and philosophical reflec-
ing), we who interpret as a vocation should not tion. Critics can and do argue, for example,
disregard it" (Aims 7, 90; emphasis omitted).15 about whether to hold with the Freudians that
The difficulty that Hirsch wrongly minimizes, human beings are primarily narcissistic, relent-
though, is that powerful overriding values are lessly seeking fulfillment of unconscious desires
always at stake in the decision between opposing in defiance of the ego's feeble attempts to tame
strategies of interpretation. Indeed, by retreating the pleasure principle to the exigencies of reality,
to the ground of ethics, Hirsch must allow the or to agree with the phenomenologists that hu-
legitimacy of a choice he would prefer to con- man beings are defined by consciousness and
demn. Barthes and Foucault invoke the value of freedom, by an ability to transcend given situa-
liberty in attacking the notion of a controlling tions and to explore the possibilities open to
author, a concept they regard as a debilitating choice.
restriction on the text's capacity for signification. Gerald Graff complains about the inconclu-
Liberty is certainly a "powerful" ethical value, siveness of such arguments: "The notion that
and indeed many would consider it an "overrid- choices determine norms rather than obey them
ing" one. does away with the idea that there are certain
Ultimate ethical disputes of this kind always norms that ought to be chosen by societies and
arise when we reach the fundamental principles thus precipitates a radical cultural relativism"
involved in "strong" hermeneutic disagreements. (Literature 38). Because norms are not given
Interpreters always base their choice of values transcendentally and eternally but are developed
on an "ought" rather than on an "is"-or, more through social agreement, Graff's demand for
precisely, they decide what to assume "is" the criteria independent of human choices can never
status of literary works and the human world on be met. The question of what we "ought" to
the basis of what "ought" to be believed about believe can only be decided by discussion and
them. Every hermeneutic standpoint has an a argument within the community. While such ex-
priori foundation because it rests on an ethical changes do not lead to agreement about a single,
decision about what it is better to believe- indubitable truth, they do introduce testing and
whether, for example, human beings are essen- evaluation, thereby rescuing the field from the
tially historical agents or instruments of a uni- anarchy of total relativism. Literary criticism is
versal logic (to recall one last time the debate a pluralistic universe, but there are limits to its
between Marxism and structuralism). pluralism.17
Literary criticism is a "rational enterprise,"
however, not only because tests for validity act University of Virginia
as constraints on its proceedings but also be- Charlottesville
350 The Conflict of Interpretations and the Limits of Pluralism

Notes
1 The Yale school includes Harold Bloom, Paul de (although it would permit these positions to continue
Man, Geoffrey Hartman, and J. Hillis Miller; see their practicing their local brands of criticism). The distinc-
anthology, Deconstruction and Criticism. See also the tion between "global" and "local" levels of discourse
books by Holland and by Fish. These theorists, of is important, however, because the global can help
course, have many crucial differences that I cannot clarify local issues while local considerations can ex-
explicate in detail here. The deconstructionists, for plain the practical significance of global alternatives.
example, are much more diverse than many of their 5There is an extensive literature on the role of
critics realize. My ultimate concern, however, lies not presuppositions in interpretation. Any list of the most
with the distinctions among recent critical positions but interesting examples would have to include-in addi-
with the basic epistemological question of relativism- tion to the theories of Heidegger, Bultmann, and
an issue with a long history before it entered the Ricoeur-Collingwood's notion of "the logic of ques-
contemporary critical scene. tion and answer," Gadamer's defense of "prejudice,"
2 Reichert calls himself a pluralist (Making Sense and Kuhn's concept of "paradigms."
xi), but his allegiance to monism is made unmistak- 6 For example, Graff describes the New Criticism as
ably clear in the argument of his book and in the the purveyor of "modernist assumptions about lan-
dispute that he and Fish fought in the pages of Critical guage, knowledge, and experience" (Literature 5). See
Inquiry (Summer 1978, Autumn 1979, and Summer also his Poetic Statement and Lentricchia.
1980). 7 My terms "blindness" and "insight" may call to
3 Given the importance of this task, it is not sur- mind the title of the well-known book by de Man,
prising that others have tackled it before me. My title although he defines them somewhat differently. His
alludes to the two most important attempts to explain argument is not that every hermeneutic insight comes
and justify a limited hermeneutic pluralism: Paul at the price of a specific mode of blindness but that
Ricoeur's Conflict of Interpretations and Wayne Booth's "Critics' moments of greatest blindness with regard
Critical Understanding: The Powers and Limits of to their own critical assumptions are also the moments
Pluralism. In the course of my argument I relate my at which they achieve their greatest insight" (109).
theory to their views. From the point of view of my theory, these moments
4 For the sake of clarity, I should point out that I are of special interest and importance because at such
use the word "theory" in two related but different times the interpreter unwittingly criticizes his or her
senses: "global" and "local." The context always in- presuppositions by seeking to transcend their limits-
dicates which is in force. A "global" theory attempts even while reaffirming allegiance to them and acknowl-
to explain the foundations of literature or the basic edging their necessity by refusing to recognize that
dimensions of the act of interpretation as such. My they are deficient.
proposals about the limits of pluralism are "global." 8 See also Ricoeur's Freud, a monumental critique of
A "local" theory of interpretation articulates the as- psychoanalysis, and Ihde's study of Ricoeur's philoso-
sumptions and habitual ways of operating that define a phy.
particular method of understanding. In this sense there 9The terms "heteronomous" and "concretization"
are Marxist and psychoanalytic theories of interpreta- derive from Ingarden; see esp. 336-43. Here and in the
tion that correlate to concrete critical practices. But next paragraph my assertions about the mode of exis-
the border between the two types of theory is some- tence of the literary work are necessarily brief. Space
times fuzzy-and necessarily so. Every local theory does not permit me to offer a full-scale ontology of
of how best to do the work of interpretation has global literature. My purpose is only to suggest the counter-
ramifications that touch aesthetic, metaphysical, and arguments that can be made to the typical, seemingly
epistemological issues. Conversely, global theoretical conclusive appeal to the identity of the text as a con-
statements of the sort "the work of art is x" or "inter- straint on interpretation.
pretation is inherently y" commonly endorse some 10Also see Booth's discussion of "common knowl-
kinds of criticism and condemn others. For example, edge" and his distinction between "understanding" and
Iser's Act of Reading simultaneously presents a global "overstanding" in Critical Understanding, esp. 241-50.
theory and a local one; it comprehensively describes 11 A good account of this problem can be found
the reading process but also makes a case for a cer- in Pratt's incisive refutation of the distinction between
tain mode of interpretive activity. "ordinary" and "literary" language (esp. 3-78). Also
My theory of hermeneutic conflict attempts to main- see Ellis' discussion of the impossibility of achieving a
tain a neutral stance that would allow any number of single, universal definition of the essence of literature
critical procedures. But obviously there is no such thing (esp. 24-53).
as neutrality, even at the global level, inasmuch as 12 The classic formulation of the general claims of
many different ways of organizing the field of literary pluralism is, of course, William James's Pluralistic
studies are possible. My "neutral" theory contests the Universe. For an important defense of a pluralistic
claims, for example, of both absolutism and relativism epistemology see Pepper.
 Paul B. Armstrong 351
13
According to Toulmin, however, literary criticism invoked as a criterion to resolve critical disputes be-
would not qualify as a "discipline." With an atypi- cause it is frequently the subject of dispute. Second,
cal and unfortunate lapse into monism, he reserves although Hirsch defends authorial intention as the only
this term for fields in which all (or by far most) standard that can make literary studies a progressive
of the investigators are committed to a single ex- discipline, the coherent and goal-oriented program of
planatory goal and to a common agenda of problems research he desires can be achieved by any group of
that must be overcome to attain the ideal state of investigators who jointly pursue the consequences of a
knowledge. certain set of assumptions. But because the interpreta-
14The dangers and benefits of authority have re- tion of literature welcomes many kinds of presupposi-
cently been the subject of considerable critical debate. tions, the field has a variety of competing research
On one side of the question, Gadamer, for example, programs, not a single agenda of problems to be
argues that authority is not always wrong. He distin- progressively solved.
guishes between the blind acceptance of authority that 16 The danger that critics who disagree about funda-
"obedience to a command" typically entails and the mental principles will not hear one another's arguments
"recognition of superior knowledge" that he describes is precisely what gives urgency to Booth's warning
as a freely made decision and a reasonable act of defer- against dismissing too hastily the views of others. His
ence (248). On the other side, however, Michel Fou- description of the responsibilitiesof "a good citizen in
cault and Edward Said argue that the power of author- the republic of criticism" is an important pragmatic
ity to control how we think, speak, and write is more guide for the successful functioning of critical ex-
pervasive and insidious than we ordinarily believe. Said change-the ongoing debate that provides one of the
is certainly right when he contends, in explaining Fou- bases for our field's claim to the status of a "rational
cault, that "a text is a place among other places (in- enterprise" (Critical Understanding 1-34, 351-52).
cluding the body) where the strategies of control in 17 I am grateful to Austin E.
Quigley for his criti-
society are conducted" (704). But it is also true that cisms and suggestions during the formative stages of
he and Foucault enjoy the privileges of the sort of my thinking on this topic and to Evelyne Keitel for her
claim to authority that Gadamer defends. helpful advice about an early version of the essay. I
15 Two further
problems with Hirsch's defense of the would also like to thank Wolfgang Iser and Murray
author as the norm for correct interpretation should Krieger for their many useful suggestions. I wrote the
also be mentioned, even though they are tangential to first draft with the support of a summer grant from the
the point I make in discussing this quotation. First, as Committee on Research at the University of Virginia
Booth rightly argues, "author" is an essentially con- and prepared the final version under the sponsorship
tested concept (Critical Understanding 8). It cannot be of the Alexander von Humboldt Foundation.

Works Cited
Barthes, Roland. "The Death of the Author." In his Selected Essays 1917-1932. New York: Harcourt,
Image/Music/Text. Trans. Stephen Heath. New 1932, 241-50.
York: Hill and Wang, 1977, 142-48. Ellis, John M. The Theory of Literary Criticism: A
Bloom, Harold, et al. Deconstruction and Criticism. Logical Analysis. Berkeley: Univ. of California
New York: Continuum, 1979. Press, 1974.
Booth, Wayne C. Critical Understanding: The Powers Fish, Stanley E. Is There a Text in This Class? Cam-
and Limits of Pluralism. Chicago: Univ. of Chi- bridge: Harvard Univ. Press, 1980.
cago Press, 1979. . "Normal Circumstances, Literal Language,
. "The Limits of Pluralism." Critical Inquiry Direct Speech Acts, the Ordinary, the Everyday,
3(1977):407-23. the Obvious, What Goes without Saying, and Other
Brooks, Cleanth. The Well Wrought Urn. 1947; rpt. Special Cases." Critical Inquiry 4(1978) :625-44.
London: Dennis Dobson, 1968. . "One More Time." Critical Inquiry 6(1980):
Bultmann, Rudolf Karl. "The Problem of Herme- 749-51.
neutics." In his Essays Philosophical and Theo- . "A Reply to John Reichert; or, How to Stop
logical. Trans. James C. G. Greig. New York: Worrying and Learn to Love Interpretation."
Macmillan, 1955, 234-61. Critical Inquiry 6(1979): 173-78.
Collingwood, R. G. An Autobiography. Oxford: Ox- Foucault, Michel. "What Is an Author?" In Textual
ford Univ. Press, 1939. Strategies. Ed. Josue V. Harari. Ithaca, N.Y.:
Conrad, Joseph. Lord Jim. Ed. Thomas C. Moser. Cornell Univ. Press, 1979, 141-60.
1900; rpt. New York: Norton, 1968. Gadamer, Hans-Georg. Truth and Method. Trans.
de Man, Paul. Blindness and Insight. New York: Ox- anon. New York: Seabury, 1975.
ford Univ. Press, 1971. Graff, Gerald. Literature against Itself. Chicago: Univ.
Eliot, T. S. "The Metaphysical Poets" (1921). In his of Chicago Press, 1979.
352 The Conflict of Interpretations and the Limits of Pluralism
. Poetic Statement and Critical Dogma. Evan- Economy." In Karl Marx and Frederick Engels.
ston, Ill.: Northwestern Univ. Press, 1970. The German Ideology. Ed. C. J. Arthur. New
Heidegger, Martin. Being and Time. Trans. John Mac- York: International Publishers, 1970, 124-51.
quarrie and Edward Robinson. New York: Harper Merleau-Ponty, Maurice. Phenomenology of Percep-
and Row, 1962. tion. Trans. Colin Smith. London: Routledge and
Hirsch, E. D., Jr. The Aims of Interpretation. Chicago: Kegan Paul, 1962.
Univ. of Chicago Press, 1976. Peirce, Charles Sanders. "The Fixation of Belief."
. Validity in Interpretation. New Haven: Yale In the Collected Papers of Charles Sanders Peirce.
Univ. Press, 1967. Ed. Charles Hartshorne and Paul Weiss. 1934;
Holland, Norman N. 5 Readers Reading. New Haven: rpt. Cambridge: Harvard Univ. Press, 1965,
Yale Univ. Press, 1975. 5:223-47.
Ihde, Don. Hermeneutic Phenomenology: The Philoso- Pepper, Stephen C. World Hypotheses: A Study in
phly of Paul Ricoeur. Evanston, Ill.: Northwestern Evidence. Berkeley: Univ. of California Press,
Univ. Press, 1971. 1942.
Ingarden, Roman. The Literary Work of Art. Trans. Pratt, Mary Louise. Toward a Speech Act Theory of
George G. Grabowicz. Evanston, Ill.: Northwest- Literary Discourse. Bloomington: Indiana Univ.
ern Univ. Press, 1973. Press, 1977.
Iser, Wolfgang. The Act of Reading. Baltimore: Johns Reichert, John. "But That Was in Another Ball Park:
Hopkins Univ. Press, 1978. A Reply to Stanley Fish." Critical Inquiry 6(1979):
. "The Reading Process: A Phenomenological 164-72.
Approach." In his The Implied Reader. Baltimore: ."Making Sense of Interpretation."Critical In-
Johns Hopkins Univ. Press, 1974, 274-94. quiry 6(1980):746-48.
James, William. A Pluralistic Universe. 1909; rpt. . Making Sense of Literature. Chicago: Univ. of
Cambridge: Harvard Univ. Press, 1977. Chicago Press, 1977.
. The Will to Believe. 1897; rpt. Cambridge: Ricoeur, Paul. The Conflict of Interpretations. Ed. Don
Harvard Univ. Press, 1979. Ihde. Evanston, Ill.: Northwestern Univ. Press,
Jauss, Hans Robert. Asthetische Erfahrung und liter- 1974.
arische Hermeneutik. Miinchen: Wilhelm Fink, . Freud and Philosophy. Trans. Denis Savage.
1977. New Haven: Yale Univ. Press, 1970.
Kuhn, Thomas S. The Structure of Scientific Revolu- Said, Edward W. "The Problem of Textuality: Two
tions. Rev. ed. Chicago: Univ. of Chicago Press, Exemplary Positions." Critical Inquiry 4(1978):
1970. 673-714.
Lentricchia, Frank. After the New Criticism. Chicago: Spitzer, Leo. Linguistics and Literary History. Prince-
Univ. of Chicago Press, 1980. ton: Princeton Univ. Press, 1948.
Levi-Strauss, Claude. "The Structural Study of Myth." Toulmin, Stephen. Human Understanding: The Collec-
In his Structural Anthropology. Trans. Claire tive Use and Evolution of Concepts. Princeton:
Jacobson and Brooke Grundfest Schoepf. Garden PrincetonUniv. Press, 1972.
City, N.Y.: Anchor-Doubleday, 1967, 202-28. Wellek, Rene. "The New Criticism: Pro and Contra."
Marx, Karl. "Introduction to a Critique of Political Critical Inquiry 4(1978):611-24.