Philosophy without Principles

Author(s): Richard Rorty

Source: Critical Inquiry , Mar., 1985, Vol. 11, No. 3 (Mar., 1985), pp. 459-465
Published by: The University of Chicago Press

Stable URL: https://www.jstor.org/stable/1343367

REFERENCES
Linked references are available on JSTOR for this article:
https://www.jstor.org/stable/1343367?seq=1&cid=pdf-
reference#references_tab_contents
You may need to log in to JSTOR to access the linked references.

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].

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

The University of Chicago Press is collaborating with JSTOR to digitize, preserve and extend
access to Critical Inquiry

This content downloaded from

129.67.111.33 on Mon, 20 Feb 2023 00:43:16 UTC
All use subject to https://about.jstor.org/terms
Pragmatism and Literary Theory
II

Philosophy without Principles

Richard Rorty

My colleague E. D. Hirsch has skillfully developed the consequences for

literary interpretation of a "realistic" epistemological position which he
formulates as follows: "If we could not distinguish a content of consciousness
from its contexts, we could not know any object at all in the world." Given
that premise, it is easy for Hirsch to infer that "without the stable de-
terminacy of meaning there can be no knowledge in interpretation."' A
lot of people disagree with Hirsch on the latter point, and they look to
philosophy for replies to the premise from which it was inferred. But it
is not clear where in philosophy they should look: To epistemology?
Ethics?2 Philosophy of language? WhatJacques Derrida calls "a new logic,
... a graphematics of iterability"?3 Where do we find first principles from
which to deduce an anti-Hirsch argument?
I want to argue that there is no clear or straight answer to this
question and that there need be none. I shall begin by criticizing the
strategy used against Hirsch and others by my fellow pragmatists Steven
Knapp and Walter Benn Michaels. They think that one can start with
philosophy of language and straighten things out by adopting a correct
account of meaning. I share their desire to refute Hirsch, their admiration
for Stanley Fish, and their view that "theory"-when defined as "an
attempt to govern interpretations of particular texts by appealing to an
account of interpretation in general"-has got to go (p. 723, and see p.
742). But they want to defend this position by exposing a mistake which
they think common to all theory so defined: an error about the relation
between meaning and intention. They assert that "what is intended and

Critical Inquiry 11 (March 1985)

? 1985 by The University of Chicago. 0093-1896/85/1103-0007$01.00. All rights reserved.

459

This content downloaded from

129.67.111.33 on Mon, 20 Feb 2023 00:43:16 UTC
All use subject to https://about.jstor.org/terms
460 Pragmatism and Literary Theory Richard Rorty

what is meant are identical" and that one will look for an "account of
interpretation in general" only if one fails to recognize this identity (pp.
729, 723). Such failure leads to an attempt to connect meaning and
intention (as in Hirsch) or to disconnect them (as in Paul de Man). But
such attempts must fail, for they presuppose a break "between language
and speech acts" which does not exist (p. 733).
Knapp and Michaels defend this latter claim by saying that marks
which are shaped like a sentence of English do not count as language
unless the marks are backed up by an intention-unless they are inscribed
by somebody who meant something by them. If one grants this point,
Knapp and Michaels argue, one will not, with speech-act theorists such
as H. P. Grice, distinguish between what sentences mean and what a
given utterer means by them on a given occasion. Grice would say of a
pattern of marks created on a beach by random wave motion that it
means whatever the sentence it has been construed to token means, even
though nobody ever meant anything by it. Knapp and Michaels would
deny this (see p. 733, esp. n.13).4 For Grice's distinction opens up the
logical space they want to close: the space in which one asks the traditional
interpretive question "Granted that the sentence means such and such,
did its author use it to mean that on this particular occasion?"
Since I regard this as a useful question, I should like to keep the
space open and, thus, to side with Grice. So I would urge that anything-
a wave pattern, an arrangement of stars, the spots on a rock-can be
treated not only as language but as any given sentence of English if one
can find some way to map its features onto the semantic and syntactic
features of that sentence (and other actual and possible patterns or ar-
rangements or spots on the other sentences of English). "Linguistical-
ity" is, on this view, cheap. You can impute it to anything simply by
working out a translation scheme, just as you can impute goodness to
anything by imagining a desirable end to which it can be a means. The
question of whether the thing is really a sentence or whether we are
simply pretending it is, is just as bad as the question of whether goodness
is objective or subjective. Both questions should be eschewed by us prag-
matists, since both presuppose Hirsch's Husserlian distinction between
content and context, between essential and accidental properties (as op-
posed to the harmless distinction between normal or familiar properties
and abnormal or unfamiliar ones). So I think Knapp and Michaels should
not go out on a metaphysical limb by saying that the absence of an

Richard Rorty is Kenan Professor of Humanities at the University

of Virginia. He is the author of Consequences of Pragmatism (Essays: 1972-
1980), among other works, and is currently writing a book on Martin
Heidegger. His previous contribution to Critical Inquiry, "Deconstruction
and Circumvention," appeared in the September 1984 issue.

This content downloaded from

129.67.111.33 on Mon, 20 Feb 2023 00:43:16 UTC
All use subject to https://about.jstor.org/terms
Critical Inquiry March 1985 461

intending inscriber means that the marks in the sands ar

For this leaves them open wider than necessary to Ad
question "Now how do Knapp and Michaels know this?" and
that "their premises beg their question" (Rosmarin, p. 78
Continuing this more-pragmatic-than-thou line, I wou
Knapp and Michaels not try to undergird Fish by constructing
of language which will make it illicit to form a general th
pretation. Rather, we should follow W. V. Quine, Donald
Jeffrey Stout in saying that the question "What is the mea
is as useless as the question "What is the nature of the good
are supposed to treat everything as a matter of a choice
nothing as a matter of intrinsic properties. They dissolv
functions, essences into momentary foci of attention, and
success at reweaving a web of beliefs and desires into mo
elegant folds. So I think that Knapp and Michaels' distrus
epistemological project" (defined by them as "bas[ing] int
a direct encounter with its object, an object undistorted b
of the interpreter's particular beliefs") would be better e
direct attack on the image of "direct encounter with objec
attack on Grice's handy distinction between more and less fam
in which to place words (p. 737). Knapp and Michaels' claim
is identical with intention suggests that we put the te
context we find useful and then call the result a discovery
intention.6 But why call it anything in particular? Why n
context, describe the advantages of having done so, and for
of whether one has got at either its "meaning" or "the auth
But Rosmarin's metaphilosophical question arises at th
up. How do I, how does any pragmatist, know that there
encounters with objects? Don't my premises beg all the in
tions? They do indeed. There is a large circle of concepts-
knowledge, truth, object, science, reference, meaning, inte
so on-such that a realist or a pragmatist analysis of any
premises from which to deduce a parallel analysis of any
There is no natural order of priority among such concep
you how to start at the very top and work your way dow
any such order among the various areas of philosophy
with metaphysics and move down to epistemology, or wi
and work down to metaphysics, or with epistemology and
ethics and down from there to metaphysics. It may seem,
that all of us who debate these matters start out with either Hirsch-like
or Fish-like intuitions and then go round in circles, defending them in
one guise by appealing to them in another.
If one thinks of philosophy as entirely a matter of deductive argument,
then this game of mirrors will, indeed, be one's only recourse. But one
can also think of philosophy in other ways-in particular, as a matter

This content downloaded from

129.67.111.33 on Mon, 20 Feb 2023 00:43:16 UTC
All use subject to https://about.jstor.org/terms
462 Pragmatism and Literary Theory Richard Rorty

of telling stories: stories about why we talk as we do and how we might

avoid continuing to talk that way. When you find yourself at an argu-
mentative impasse, baffled by your opponent's refusal to stop asking
questions which you think you really should not have to answer, you can
always shift the ground by raising questions about the vocabulary he or
she is using. You can point out that the issue is biased in one's opponent's
favor by the unfortunate jargon which has developed, a jargon which
gives one's opponent an unfair advantage. You can use historical narratives
to show why the issue previously discussed is moot and why it needs to
be reformulated in terms which are, alas, not yet available.
This strategy of using narrative where argument fails is what makes
Heideggerian and Derridean attempts to "problematize" the vocabulary
used by contemporary philosophers so attractive. Inconclusive debates
between reformers of the Right and the Left make revolutionaries look
good. So it is tempting to think that the pragmatist should stop offering
analyses of knowledge and truth and instead fall back on a quasi-Hei-
deggerian account of how we got into our present dead end. Derrida
tells us that unless we go back and deconstruct what Plato built, we shall
always be haunted by his ghost-by the idea that there is some natural
starting point and resting-place for thought, something like Hirsch's
context-free contents of consciousness. Accordingly, one might think that
only by overcoming the metaphysics of presence-ceasing to use not
only "meaning" and "intention" but all the bad old Platonic oppositions
which make these notions seem inevitable-could pragmatists like Knapp
and Michaels end "theory."
Pragmatists and Derrideans are, indeed, natural allies. Their strategies
supplement each other admirably. But there is no natural priority of one
strategy over the other. It is not the case that we shall have rational
grounds for rejecting realism only if we can overcome the metaphysics
of presence. The notion of "rational grounds" is not in place once one
adopts a narrative strategy. (That is why Derrida looks bad whenever he
attempts argument on his opponents' turf; those are the passages in
which he becomes a patsy for John Searle.) For if we ever did get rid of
all the jargon of the tradition, we should not even be able to state the
realist's position, much less argue against it. The enemy would have been
forgotten rather than refuted. If Derrida ever got his "new logic," he
would not be able to use it to outargue his opponents. Whatever a "gra-
phematics of iterability" might be good for, it would be of no use in
polemic. The metaphysics of presence was designed precisely to facilitate
argument, to make questions like "How do you know?" seem natural,
and to make a search for first principles and natural resting-places seem
obligatory. It assumes that all of us can tell such a resting-place when
we see it and that at least some of our thoughts are already there. You
can't argue against that assumption by using the vocabulary of the tradition,
but neither can you argue that the tradition is wrong in its choice of
vocabulary. You can argue only against a proposition, not against a vo-

This content downloaded from

129.67.111.33 on Mon, 20 Feb 2023 00:43:16 UTC
All use subject to https://about.jstor.org/terms
Critical Inquiry March 1985 463

cabulary. Vocabularies get discarded after looking bad in

with other vocabularies, not as a result of an appeal to o
metavocabularies in which criteria for vocabulary choice can be f
This means that narrative philosophy should not be expe
gaps left vacant by argumentative philosophy. Rather, the
of narrative philosophy is that persuasion is as frequently
getting people to drop a vocabulary (and the questions they p
it) as of deductive argument. So, though I think Rosmarin
suggesting that the linguistic reform which Knapp and Mich
is the wrong way (because it is a needlessly paradoxical w
their antitheoretical point, I do not mean that linguistic
generally bad-or ineffective-strategy, nor that Rosmar
"How do you know?" should be pressed. What is wrong with
is that, as asked by Socrates and the Platonic tradition, it a
we know what knowledge is like and can tell when we have
this notion of knowledge as an introspectable state is ju
Platonic myth. The right way to construe this question is "
find what you just said persuasive?" That is a question which
traditional distinctions between reasons and causes, psycholo
rhetoric and demonstration. It is a practical question, a poli
the question "What am I going to have to do to convince yo
To return to the question of theory, one can be against w
and Michaels define as the attempt to get outside practice a
it, and agree nonetheless with another of their critics, Stev
that theory should "continue doing what all discursive practices d
to persuade its readers to adopt its point of view, its way of
and the world" (Mailloux, p. 766).7 For, in its unobjectio
"theory" just means "philosophy." One can still have phil
after one stops arguing deductively and ceases to ask wh
principles are coming from, ceases to think of there being a s
of the world-or the library-where they are found. In parti
"literary theory," as the term is currently used in America, t
of philosophy, an attempt to weave together some texts tra
labeled "philosophical" with other texts not so labeled. It
practice of splicing together your favorite critics, novelists
such, and your favorite philosophers. This is not exactly wh
calls "metapractice (practice about practice)," for that term
vertical relationship, in which some practices are at higher
others (Mailloux, p. 766). Rather, it is just more practice
sort, using a slightly different set of raw materials. Thinkin
way helps one get rid of the idea that philosophy is somehow
level. It lets one think of "philosophical" and "literary" texts
the same mill.
To conclude on a blatantly practical note, I would offer as one reason
in favor of my version of pragmatism and against Knapp and Michaels'
that they are driven to the conclusion that we should "eliminate the

This content downloaded from

129.67.111.33 on Mon, 20 Feb 2023 00:43:16 UTC
All use subject to https://about.jstor.org/terms
464 Pragmatism and Literary Theory Richard Rorty

'career option' of writing and teaching theory" (p. 800). In my view, this
career option consists in an opportunity to discuss philosophy books-
as well as novels, poems, critical essays, and so forth-with literature
students. Knapp and Michaels, however, construe it as the attempt to
supply foundations for literary interpretation. I would hope that the
latter rhetoric could be discarded while the career option remains. The
recent emergence of this option seems to me one of the healthier features
of American academic life. For, as Paul Alpers has remarked, courses in
"literary theory" have become "ports of entry" for a tradition of European
philosophical thought which had been neglected in America. There is
no particular reason why this tradition should be taught in literature
departments rather than in philosophy departments, but there is also no
particular reason why it should not be. It should certainly be taught
somewhere in our universities, and it seems to me greatly to the credit
of our literature departments that they have given it a home.

1. E. D. Hirsch, Jr., The Aims of Interpretation (Chicago, 1976), pp. 3, 1.

2. See ibid., where Hirsch offers a "fundamental ethical maxim for interpretation"
which, he says, "claims no privileged sanction from metaphysics or analysis" (p. 90). Here
and elsewhere Hirsch suggests that it may be ethics rather than epistemology which provides
the principles that govern interpretation. There remain other passages, however, in which
he retains the view, conspicuous in his earlier writings, that an analysis of the idea of
knowledge is the ultimate justification for his approach.
3. Jacques Derrida, "Limited Inc abc ... ," Glyph 2 (1977): 219.
4. It seems to me that Knapp and Michaels are wrong in thinking of John Searle and
H. P. Grice as "arriv[ing] at determinate meanings by adding intentions [to language]" (p.
733). They are, rather, distinguishing between two sets of intentions-the ones normally
had by users of a sentence and some special ones had, or possibly had, by an individual
user. More generally, I cannot think of anybody who would deny that one has language
only where there is a system of community intentions, of conventions (in the sense analyzed
by David Lewis). So I am not sure that there are any "anti-intentionalist accounts of
meaning" (a phrase used by Knapp and Michaels; see p. 727). Paul de Man (cited as an
example of an anti-intentionalist; see Knapp and Michaels, "A Reply to Our Critics," p.
795 n.4) does not seem to me to hold such a view. Nor, I think, does Michel Foucault. I
would take Foucault to be saying that one can tell useful historical stories if one takes
language rather than human beings as one's subject, bracketing questions about why human
beings changed their linguistic habits. But that does not commit him to the claim that
language can exist without human beings establishing conventions, any more than atheistic
idealism is committed to the view that minds existed before rocks did. Questions about
what comes first in causal sequence are irrelevant in both cases.
5. See Jeffrey Stout, "What Is the Meaning of a Text?," New Literary History 14 (Autumn
1982): 1-12.
6. See Knapp and Michaels: "Any interpreter of any utterance or text, within the
institution of professional literary criticism or not, is, if we are right, attempting to understand
the author's intention" (p. 800). It seems to me that Knapp and Michaels do not satisfactorily
answer Hirsch's claim that their arguments show only that "a text's meaning ... must
always be what an author intends it to mean" and not that it "must always be what its author
intends it to mean" (Hirsch, p. 745).

This content downloaded from

129.67.111.33 on Mon, 20 Feb 2023 00:43:16 UTC
All use subject to https://about.jstor.org/terms
Critical Inquiry March 1985 465
7. Cf.: "What Knapp and Michaels treat as inherent and erroneous dissociations might
be more subtly and profitably discussed as analytic strategies" (Rosmarin, p. 780). In their
"Reply," Knapp and Michaels say that Mailloux and Rosmarin are both "negative theorists"
(p. 795). (As Knapp and Michaels define it, negative theory tries to preserve "the purity
of language from the distortion of speech acts" [p. 733].) But I should think that the
"hostility to method" which Knapp and Michaels say characterizes negative theory was
independent of any such motive (p. 733; and see n. 4 above).

This content downloaded from

129.67.111.33 on Mon, 20 Feb 2023 00:43:16 UTC
All use subject to https://about.jstor.org/terms