Deconstruction in America

Author(s): Michael Sprinker

Source: MLN, Vol. 101, No. 5, Comparative Literature (Dec., 1986), pp. 1226-1242
Published by: The Johns Hopkins University Press
Stable URL: http://www.jstor.org/stable/2905717
Accessed: 01-06-2018 00:14 UTC

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 support@jstor.org.

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

The Johns Hopkins University Press is collaborating with JSTOR to digitize, preserve and
extend access to MLN

This content downloaded from 181.231.196.143 on Fri, 01 Jun 2018 00:14:56 UTC
All use subject to http://about.jstor.org/terms
 REVIEWS

Deconstruction in America

"We are living in a material world,

And I am a material girl."
-Madonna

Faithful, or even desultory, readers of MLN will certainly smile knowingly

at my title. They have learned, in a variety of places, from a plurality of
points of view, and in at least several distinct idioms, that the topic to
which my title alludes is one of singular moment in the current conjunc-
ture of literary study in North America. It would hardly be stretching a
point to say that American, or Euro-American culture is now and has for
some years been negotiating a passage from one historical epoch to an-
other largely under the sign of "deconstruction." Certainly many of the
authors to be discussed have not hesitated to claim this much. The evi-
dence for this claim is too plentiful and too familiar to require detailed
enumeration here. I might say, "Just look around you, isn't it obvious?"
But allow me one example.
At a seminar conducted under the auspices of the National Endowment
for the Humanities in New York City during March 1986, the then acting
assistant director of the Endowment was questioned about what sorts of
projects would be likely to obtain a favorable hearing there in this period,
what ones were less likely to do so. Like any accomplished bureaucrat, he
hedged, equivocated, back-peddled, trying hard not to commit himself.
But on one point he did not hesitate: to wit, that what he called "decon-
structionist" projects would have a difficult row to hoe. This was the case,
not because they were out of favor with the Endowment and its staff, but
because, in his words: "As we all know, 'deconstructionism' is a very con-
troversial subject these days. It is likely to provoke strong feelings on both
sides, and therefore it will fail to gain the consensus among peer reviewers
that is in most cases necessary for the Endowment to give its imprimatur to
a project." When the issue of the viability or utility of a research paradigm
has reached the middle levels of the federal bureaucracy, one can be fairly
certain that a polemical battle for the intellectual high ground in this so-
ciety has been joined on a broad front.
To think of deconstruction as a research program may seem at first an
odd notion. But as the texts accumulate, the readings of individual au-

This content downloaded from 181.231.196.143 on Fri, 01 Jun 2018 00:14:56 UTC
All use subject to http://about.jstor.org/terms
 M L N 1227

thors proliferate, and the purview of deconstruction as a distinctive mode

of reading and interpretation continually expands, it becomes ever more
apparent that the term popularized by Jacques Derrida during the 1960s
has acquired both a manifest institutional presence and persistence (in
journals like Diacritics, Glyph, October, and boundary 2, as well as in course
syllabi in literary criticism and theory), and a certain coherence and cohe-
siveness among its adherents-never mind its opponents, who see decon-
struction as the embodiment of all that is misguided, irresponsible, and
ultimately evil in intellectual life today. That deconstruction constitutes a
distinctive, if heterogeneous and uneven, method of research in philos-
ophy and literature is the fundamental assumption enabling Rodolphe
Gasch6's The Tain of the Mirror, which gives an authoritative exposition of
Derrida's work in the historical context of philosophical critiques of re-
flexivity. Gasch6 situates Derrida's project within this tradition of argu-
ment, while identifying the distinctive twist Derrida has given to the
problem that characterizes the particularity of deconstruction as a
method. If, as Gasch6 puts it, "hardly any deconstructionist critic could lay
claim to" the concept of deconstruction thus advanced, this does not eo ipso
suggest the futility of organizing the disparate practices of literary critics
and philosophers under the single rubric of deconstruction.I Similarly
centrifugal tendencies are identifiable in post-Wittgensteinian analytic
philosophy without, for all that, the overall coherence of the project being
significantly endangered.2
Nonetheless, Gasch6's insistence upon a strict separation between de-
constructive criticism and the philosophical operation of deconstruction
can provide a way of provisionally differentiating the texts under review
here, at the same time that it gives a clue to the geopolitical distinction
implied in our title. Gasch6 himself has repeatedly affirmed what he suc-
cinctly asserts at the outset of his own text: "Deconstructionist criticism
must be understood as originating in New Criticism; it is a continuation of
this American-bred literary scholarship" (TM, p. 3). Paul de Man, of
course, conceded the point on a number of occasions, recognizing his own
complicity with and training in Anglo-American New Criticism (see, for
example, "The Return to Philology," in RT, pp. 21-26), while speculating
on the extent to which the distinction between a philologically oriented
criticism and the operation of philosophical concepts could be readily dis-
tinguished in Derrida's own work. But de Man elsewhere observed that
the point de depart for each (his own pedagogical and didactic intent with
respect to texts versus Derrida's customary responsiveness to the "pres-
sure of generally philosophical issues" [RT, p. 117]) enforced a certain
difference in the style and possibly the effect of their writings. One might
begin at this point to draw a (to be sure, somewhat fluid) boundary be-
tween deconstructive criticism in its peculiarly American inflection and
the strict philosophical operation of a European-originated deconstruc-
tion.
On this account, the pedagogical and philological emphases of Hillis

This content downloaded from 181.231.196.143 on Fri, 01 Jun 2018 00:14:56 UTC
All use subject to http://about.jstor.org/terms
 1228 REVIEWS

Miller and Neil Hertz are perfectly patent. Hertz's collection contains, inter
alia, a highly entertaining essay entitled "Two Extravagant Teachings" in
which the institutional context of pedagogy is laid bare with considerable
energy and wit. Moreover, the book is suffused with a feeling for the dia-
logic encounter with texts that is the hallmark of the seminar situation.
Miller's readings of the several texts whose authors are enumerated in his
sub-title invariably focus on specific passages and engage in that patient,
careful work of exposition which in France goes under the name of expli-
cation de texte and in the U.S. is known as "close reading." On the other side
of the fence, as we observed already, stands Rodolphe Gasch6, for whom
deconstruction is a distinctive project within the history of philosophy, or
it is nothing at all.
Continuing with the works under review here, one would have to in-
clude Derrida's Wellek Library Lectures among the philological texts of
deconstructive criticism, not only because of their close exegesis of de
Man's writings, but also because, one might say, the texts of de Man re-
main effectively undeconstructed.3 Rather, what Derrida accomplishes,
with his customary patience and precision and his unerring instinct for
locating the salient points in an author's corpus where both its greatest
power and its greatest vulnerability are manifest, is to enumerate those
topics in de Man's writings that organize and unify his project from begin-
ning to end (above all, the deflection of a certain stratum of the Heideg-
gerian text), while suggesting that, after all, there was something distinc-
tive, possibly peculiarly "American" about de Man's way of proceeding.
But the difficult work of deconstructing de Man's texts, of exposing the
irreducibly aporetic structure of their conceptual organization, of rein-
scribing their terms in another discursive or conceptual space-this Der-
rida has left for another occasion.4
Less easy to situate in relation to the separation of philology from phi-
losophy is Andrzej Warminski's Readings in Interpretation. In the first place,
the text is about evenly divided between chapters dealing with the history
of Holderlin interpretation and those that take up problems in the system
of Hegelian philosophy. Since the voice of Heidegger is interpolated with
both (along with, somewhat more mutedly, that of Blanchot), the neat
division between philological exegesis and philosophical speculation is al-
most entirely effaced. Warminski himself suggests in his introduction that
the center of the text, both chronologically (it was the first chapter written)
and thematically (it inaugurates the questions and suggests the itinerary of
the other chapters) is his interpretation of Hblderlin's Patmos. The asser-
tion that the project of Hegelian philosophy hinges on the problem of
figuration and the structure of tropes suggests the grounding of War-
minski's own discourse in the domain of literature. On the other hand,
one reads the chapters on Hegel with the feeling that there, for the first
time in this text, the genuine difficulties of the philosophy of language
and its relation to aesthetic representation are posed. In this sense, the
chapters on Hegel (in particular chapters 6 and 7) are, conceptually

This content downloaded from 181.231.196.143 on Fri, 01 Jun 2018 00:14:56 UTC
All use subject to http://about.jstor.org/terms
 M L N 1229

speaking, prior to those on H6lderlin: the latter are didactic exercises in

the method of reading grounded in the concepts of language and percep-
tion developed in the former. Warminski, however, has foreseen just this
manner of construing the structure of his text, recognizing the power and
the hubris of professional philosophy to command the heights of inquiry.
In a programmatic statement whose validity could only be judged in a
more detailed exposition of his text than space here permits, Warminski
rejects philosophy's claim to control the domain of literary language and
to maintain a rigorous demarcation between the two fields:

no matter how rigorously and how vociferously it protests, denounces, or dis-

avows the simulation and mimicry of literary reading (and writing), every inter-
pretation of a movement of thought [here Heidegger's commentary on Hegel's
Introduction to the Phenomenology of Spirit] necessarily exposes itself to that very
simulation and that very mimicry precisely to the extent that the very language,
text, terms, words, of that protest, denunciation, or disowning can always be
(re)read (and [re]written) literarily.
(RI)

To the extent that philosophical speculation remains, as de Man and Der-

rida have in their different ways maintained, contaminated by the residue
of linguistic determinations it cannot entirely master, the discrimination
between deconstructive method in philosophy and deconstructive literary
criticism would seem all but insupportable in principle-whatever value it
may have as a pragmatic description of the differences separating the
texts of Derrida from the practice carried on in his name. It is by no
means certain that the negative stigma attached by Gasch6 to "deconstruc-
tive criticism" can justly be applied to any of the texts we are considering
here. The categorical distinction between deconstructive method and de-
constructive criticism may ultimately prove just as "delusive" as that once
remarked by de Man between literature and criticism.5
Finally, we have the texts of de Man himself, and we have the difficulty
presented by his own self-description as a pedagogue with strong affinities
for the discipline of close reading associated with Anglo-American New
Criticism. Those who have followed de Man over the years will be inclined
to question the lucidity of an author's own understanding of his writings,
for the author is by definition the one least in a position to say what his
texts are or are not doing-which fact never prevented any author, de
Man included, from saying a great deal about what his texts mean and
how they are intended to be read. This aspect of de Man's writings is
foregrounded in Derrida's commentary, and so we turn now to this latter
for some further indication about how to proceed in coming to terms with
"deconstruction in America."
Our topic was that initially proposed by Derrida for his Wellek Library
Lectures delivered in 1984. He finally decided against it, choosing instead
to address the work of Paul de Man, his then recently departed colleague
and friend. But this was not to abandon the topic altogether, if at all. If

This content downloaded from 181.231.196.143 on Fri, 01 Jun 2018 00:14:56 UTC
All use subject to http://about.jstor.org/terms
 1230 REVIEWS

deconstruction has indeed made its way in America, then this process,
which Derrida rightly observes is "ongoing" and thus untotalizable (M, p.
17), was in no small measure due to the writing and teaching of Paul de
Man. To comprehend the trajectory of his career, to grasp the leading
motifs that organized the field of his speculation, to come to terms (even
in a preliminary way) with his project-this is to take a considerable stride
towards describing what is characteristic about deconstruction in America.
The task is by no means simple, and Derrida's caution in putting forward
theses about the itinerary and the conceptual armature of de Man's work
should not be taken only as a sign of piety or affection. It results, rather,
from a prudent suspicion about the ultimate commensurability among the
different motifs de Man pursued over the course of more than thirty
years. The coherence one discerns in de Man's project may prove recalci-
trant to ordinary narrative totalization. One is therefore enjoined to con-
sider circumspectly formulations like the following: "Always already,
there is deconstruction at work in the work of Paul de Man, even during
the period when he did not speak of it or during the time when he spoke
of it in order to say that there was nothing new to say about it" (M, p 124).
No one is more conscious than Derrida of the caution required in pro-
posing this concept of de Man's career.
Still, something demonstrably happens, Derrida is quick to point out,
when the term "deconstruction" emerges openly into the public discourse
of the human sciences. Even if this event is not precisely locatable with the
punctual precision upon which certain forms of history continue to rely to
stabilize their understanding of the past, there remains one's stubborn
sense that de Man's work did undergo a significant permutation after his
encounter with Derrida. The most obvious mark of the effects of this en-
counter is the word "deconstruction" itself, which only begins to appear in
this period of de Man's writing. Despite one's unease in utilizing peri-
odizing concepts when discussing de Man, the structure of his corpus is
punctuated by shifts both in terminology and in conceptual focus that
allow for a provisional schematization. Derrida glances at this question
early in his lectures in a note on the debate between Rodolphe Gasch6 and
Suzanne Gearhart (M, p. 40, n. 3), but reserves his fuller thoughts for the
final lecture, when he broaches the differences separating "early" from
"late" de Man apropos of some of the (chronologically, if not necessarily
conceptually, late) prefaces de Man appended to the books he published
during his lifetime.6
Derrida's itinerary through de Man's corpus begins with de Man's pro-
longed and insistent deflection of certain Heideggerian themes. The third
lecture reopens this topic, as Derrida establishes the difference in tone
that can be heard in de Man's parodic citation of Heidegger's Die Sprache
spricht near the end of Allegories of Reading (M, pp. 94-98). The argument
then shifts to an ostensibly non-Heideggerian aspect of de Man's mature
writings: their "unprecedented bringing into play and at the same time a
subversive reelaboration of Austinian theorems and of speech act theory,

This content downloaded from 181.231.196.143 on Fri, 01 Jun 2018 00:14:56 UTC
All use subject to http://about.jstor.org/terms
 M L N 1231

which in de Man's work at the same time progresses and enters a crisis"
(M, p. 111). The turn away from the problematic of Heideggerian philos-
ophy of language is explicit: "here [in de Man's appeal to the theory of
speech acts] the interest is in texts, in textual figures (textual allegories, for
example) and not in the originality of a Sprache before any Versprechen [as
is the case in Heidegger]; here the interest is in textualization or contex-
tualization rather than the original meaning of the name" (M, p. 111).
This turn, in turn, yields to the emergence of the term "deconstruction" in
de Man's discourse.
Despite its being "always already at work in the work of Paul de Man,"
deconstruction appears openly for the first time (in propria persona, as it
were) only with de Man's encountering it in the texts of Derrida. We have
de Man's own word for the fact, given in a text to which Derrida here
refers us, the preface to Allegories of Reading. By drawing our attention to
this document, as well as to other similarly self-reflective moments in de
Man's writings, Derrida opens the door to the deconstruction of the sys-
tematicity of de Man's text which Memoires studiously declines to under-
take, but which the de Manian text itself imposes by the terms of its self-
portrayal. The preface opens as follows: "Allegories of Reading started out
as a historical study and ended up as a theory of reading" (AR, p. ix). The
opposition between "a historical study" and "a theory of reading" remains
constitutively in force throughout Allegories, from the historical and polit-
ical characterization of the current situation in criticism that opens "Se-
miology and Rhetoric," to the rehearsal of the history of Rousseau inter-
pretation inaugurating the second half of the book. The "shift" (as de Man
terms it) from "historical study" to the "theory of reading" never quite
occurs, as each new occasion for the elaboration of a method (the theory
of reading) necessarily falls back into a fatal dependence on the historical
understanding that attaches to the corpus of major writers. This remains
true of even such late de Man texts as his last completed essay, "Aesthetic
Formalization: Kleist's Uber das Marionettentheater," and the final two of his
Messenger Lectures, "Schiller and Kant" and the anomalously titled
"Conclusions," dealing with Benjamin's "The Task of the Translator" (the
latter is included in RT, pp. 74-94).
The opposition between "historical study" and the "theory of reading"
marks de Man's mature texts, and we may wish to interrogate them most
critically at just this point. It is by no means certain whether the two terms
are irreconcilable in the way de Man here suggests. The patient discipline
of philological investigation, what in traditional usage would be an indis-
pensable means to literary historical study, remains the only reliable mode
of access to the theory of reading. Once again, we have de Man's word for
it, and again in this very preface: "What emerges [from the essays in Alle-
gories of Reading] is a process of reading in which rhetoric is a disruptive
intertwining of trope and persuasion or-which is not quite the same
thing-of cognitive and performative language. The implications of this
conclusion are not easy to unfold, nor can they be stated in summary

This content downloaded from 181.231.196.143 on Fri, 01 Jun 2018 00:14:56 UTC
All use subject to http://about.jstor.org/terms
 1232 REVIEWS

fashion, separated from the intricacies of specific readings" (AR, p. ix).

We shall return to this problem later.
Derrida's attention to the moment when de Man's work first "con-
sciously comes across" (AR, p. x) deconstruction produces one of those
occasions that occur with increasing frequency in Derrida's later writings
when the personal intersects the conceptual. Derrida was of course him-
self the occasion for de Man's "coming across" deconstruction "consciously
for the first time." This occasion, it turns out, involved more than the
publication of De la grammatologie and the original French publication of
"The Rhetoric of Blindness." There was also an exchange of letters, a
private correspondence, a portion of which Derrida here, not without hes-
itation, makes public. Derrida's scruple about revealing words written in
private and for his eyes alone is finally overcome by his sense of the pecu-
liar pertinence these letters bear to the moment of deconstruction in the
discourse of Paul de Man. These texts tell a little story, an exemplary an-
ecdote, what one with only slight misgiving would call an allegory. And
what is this story the allegory of? Why reading, of course, and an exem-
plary reading at that: the reading of Paul de Man reading Derrida reading
Rousseau. In the texts thus read, Derrida finds an illustrative tale of how
deconstruction came to America. Since the dispute that occasioned these
letters is by now familiar, and the letters themselves, after Derrida's publi-
cation, destined to be scarcely less so, we can omit summarizing their
matter and select only a short passage for quotation. The passage illumi-
nates both what Derrida terms "Paul de Man's singular trajectory" (M, p.
123) and the trajectory of "deconstruction in America":

The desire to exempt Rousseau (as you say) at all costs from blindness is there-
fore, for me, a gesture of fidelity to my own itinerary. Rousseau has led me to a
certain understanding which, due allowance being made, seems to me near to
that which you have had the force to begin. And as l'Essai sur l'origine des langues is
one of the texts upon which I have been relying for such a long time, I must have
put a certain ardor into my defense of the relative insight which I have benefited
from. This having been said, I did not wish to exempt Rousseau from blindness
but only wished to show that, on the specific question of the rhetoricity of his
writing, he was not blinded. This is what gives to his text the particular status that
we would both agree, I believe, to call "literary."

(M, p. 130)

De Man discovers in Derrida what was first made apparent to him in the
texts of Rousseau, but which Derrida seems to deny is there in those same
texts. De Man's stubborn insistence on Rousseau's lucidity about the lit-
erary status of his own writings clashes with the overt message of Derrida's
text, which must deconstruct Rousseau in order to produce the concept of
deconstruction itself. As Derrida previously observed: "This was another
way of saying: there is always already deconstruction, at work in works,
especially in literary works. Deconstruction cannot be applied, after the fact
and from the outside, as a technical instrument of modernity. Texts de-

This content downloaded from 181.231.196.143 on Fri, 01 Jun 2018 00:14:56 UTC
All use subject to http://about.jstor.org/terms
 M L N 1233

construct themselves by themselves, it is enough to recall it or to recall them

to oneself" (M, p. 123). So, Derrida and de Man agree? Not quite. Not
here. Not yet.
The privilege granted by de Man to what he surmises Derrida "would
agree to call 'literary' " is just what has stranded so-called "deconstructive
criticism" on the shores of America, separating it definitively from decon-
structive method. This was the source of Gasch6's initial criticism of de
Man,7 which is continued, although de Man is not now singled out for
blame, in the critique of deconstructive criticism in The Tain of the Mirror.
The debate between Gasch6 and Suzanne Gearhart over this point is in-
structive. For reasons of economy, we cannot consider it fully here. What
is most pertinent to our present topic, however, is the degree to which
Gasche's strictures against the appropriation of deconstruction as a meth-
odological tool for literary study apply to those studies I have provisionally
lumped together under the mantle of the philological and pedagogic. To
what extent do they continue the project of de Manian deconstruction in
America, in the classroom, in the discipline of literary criticism? Are these
forms of "deconstructive criticism" deconstructions at all?
On the face of it, no work fits more closely Gasch6's characterization of
deconstructive criticism as a continuation of the practice of close reading
than Hillis Miller's The Ethics of Reading. After an opening chapter that
surveys the theoretical terrain of reading in relation to ethics, Miller
plunges directly into a series of textual explications of exemplary passages
where reading and understanding are either directly thematized (as in
Henry James's meditation on revision in the preface to The Golden Bowl) or
are implicit in the topic of the discourse (as in Kant's reflections on prom-
ising in the Critique of Practical Reason). In each case, save one, the concep-
tual complications inherent in the language of the passage give rise to a
certain incoherence or semantic interference that Miller will call "unread-
ability." By patiently unfolding the semantic potential immanent in the
texts, Miller shows how contradictory constructions are necessarily gener-
ated by the text itself, and how reading cannot but recognize this fact. This
constitutes the properly ethical moment in reading, the submission to the
law that the text imposes upon the reader independent of the reader's will
or wish. The key formulation of this principle comes from de Man and is
cited twice in Miller's lectures. De Man writes: "Reading is an argument
... because it has to go against the grain of what one would want to
happen in the name of what has to happen . .". Miller correctly takes this
formulation to indicate the contingency inherent in the act of reading,
what de Man calls the "empirical and literal event" that distinguishes
reading from, say, demonstrative proofs in the empirical sciences. As
Miller observes apropos of Henry James's understanding of what it meant
for him to revise his earlier fictions for publication in the New York Edi-
tion: "The act of re-reading is free, unbound, in the specific and again
somewhat surprising sense that it is not chained by the shackles of theory.
I suppose what James means by that is that what happens when he re-

This content downloaded from 181.231.196.143 on Fri, 01 Jun 2018 00:14:56 UTC
All use subject to http://about.jstor.org/terms
 1234 REVIEWS

reads is not determined by any theoretical presuppositions about what is

going to happen when he re-reads. What happens happens, when we
really read, as opposed to imposing on the text assumptions about what we
are going to find there" (ER).
On this point, Miller's (and de Man's) account of reading corresponds to
Gasch6's (and Derrida's) claim about the imperatives of deconstructive
methodology: "Deconstruction is never the effect of a subjective act of
desire or will or wishing. What provokes a deconstruction is rather of an
'objective' nature. It is a 'must', so to speak" (TM, p. 123). Miller's readings
of texts would fall within what Gasch6 terms "infrastructural accounting"
(TM, 142), the necessary submission to the internal law governing the
structure of a given text. It is this unavoidably empirical moment that
deconstruction (or, in Miller's and de Man's terms, "reading") fore-
grounds, and that differentiates it from other modes of theoretical de-
scription: "Deconstruction, as a methodical principle, cannot be mistaken
for anything resembling scientific procedural rules . . . deconstruction is
also the deconstruction of the concept of method (both scientific and phil-
osophical) and has to be determined accordingly" (TM, p. 123).
Up to this point, then, Miller's exposition of texts coincides with the
general project of deconstruction. It could even be said that in the course
of The Ethics of Reading Miller has taken the next step, moving from iden-
tification of the conceptual semes in certain texts that organize the prob-
lematic of ethics and reading (which could, by itself, merely reproduce the
banalities of thematic criticism), to a reversal of the hierarchy implied in
such terms as "promise" and "lie" in Kant, or "re-writing" and "revision"
in Henry James. It is less probable, however, that the distinctive mark of
deconstructive method in Derrida, what Gasch6 terms "inscription," ever
quite takes place in Miller's discourse. His continuing fidelity to the letter
of the text, to the terms and concepts which the text imposes upon any
reader, situates Miller's work squarely within that tradition of close textual
explication Gasch6 calls "deconstructive criticism." Miller's readings dis-
cover what de Man termed the "literary" features in the texts of Rousseau,
the lucidity texts exhibit (whether their authors are aware of this or not)
about their own rhetorical mode. This is also what makes Miller's project,
in its own way, scientific and methodical in the sense Gasch6 denies de-
construction can ever be. The Ethics of Reading ends where it began, Miller
avows, "before the law of the ethics of reading, subject to it, compelled by
it, persuaded of its existence and sovereignty . . ." (ER). And yet the terms
of this law cannot be extrapolated from the empirical moment of reading
itself, which returns perpetually to confront the reader with more texts to
be read, further opportunities to confirm or disconfirm the existence of
the implicit law that has governed Miller's procedure in this book: "I re-
main forced to postpone once more the direct confrontation of the law of
the ethics of reading, unless that necessity of deferring is itself the law to
which I am subject. In order to test that possibility it will be necessary to

This content downloaded from 181.231.196.143 on Fri, 01 Jun 2018 00:14:56 UTC
All use subject to http://about.jstor.org/terms
 M L N 1235

read novels and tales by Eliot, Trollope, James, and others, perhaps Kleist,
Hawthorne, Melville, Hardy, or Blanchot, for example. To that task I now
turn, but in another time and place, within the covers, under the cover, of
another book" (ER). This is simply the ordinary condition of any empirical
science in relation to its hypotheses, which remain permanently open to
revision. Miller's project is thus in principle one that will contribute to our
further knowledge about texts and their modes of organization. It may in
time yield a set of principles or procedures by which reading is guided.
Should this prove the case, deconstructive criticism will have moved
beyond the limits of deconstruction.
If one mark of deconstructive criticism is its submission to the force of
the empirical and literal event of a given text, then none of the books we
are considering is more in tune with this spirit than Neil Hertz's The End of
the Line. The very heterogeneity of the examples it brings together (Lon-
ginus on the sublime, Flaubert and Sartre, the regulations of Cornell Uni-
versity concerning plagiarism, some paintings by Courbet and the proper
shape of the Phrygian cap of the French Revolution) enforces a scholarly
and exegetical discipline that militates against general theoretical specula-
tion. How, one is led to ask, could such a disparate set of phenomena be
subsumed under any single description or theoretical principle.
Hertz's reply is that they all exhibit, when one attempts to read or un-
derstand them, certain "points of opacity which, borrowing [Kenneth]
Burke's expression, I call 'the end of the line'" (EL, pp. 219-20). This
structural property of all the texts examined necessitates what Hertz terms
"double readings" (EL, p. 220), which, it happens, coincide with the act of
reading itself. At this point, Hertz quite properly invokes de Man, for the
textual structures Hertz's interpretations have discovered in a variety of
locales (including, prominently, Freud, a place de Man himself hardly
ever visited, at least in print)-these structures resemble nothing so much
as that moment of "unreadability" de Man insisted marked the rhetorical
or literary dimension of texts. The generalized formulation is as follows:
"To call attention to end-of-the-line structures then, is no more than one
way-one somewhat excited way-of talking about what happens when
one reads; the particular opacity I've attributed to these regions in texts is
what de Man would call the mark of their 'unreadability'" (EL, pp.
222-23). This raises the difficult problem, which is another version of the
possible future we observed in relation to Miller's project of recurrent
empirical encounters with individual texts, of whether "unreadability": 1)
designates that moment in the elaboration of a theory when its limits be-
come apparent, leading to the theory's revision and the surpassing of
those limits; or 2) signifies the unsurpassable limit of theory per se.
Without being utterly dogmatic on this point, de Man tended towards the
latter conclusion, calling attention to the fact that rhetorical readings were
necessarily "theory and not theory at the same time, the universal theory
of the impossibility of theory," and that therefore the "resistance to

This content downloaded from 181.231.196.143 on Fri, 01 Jun 2018 00:14:56 UTC
All use subject to http://about.jstor.org/terms
 1236 REVIEWS

theory" (Hertz's "end-of-the-line structures") was immanent within theory

itself (RT, p. 19).
Hertz's studious avoidance of such lapidary formulations (the ones hos-
tile critics of deconstruction have a tendency to cite ad nauseam) is of a
piece with the hesitations he observes in texts at their moment of greatest
intricacy. Having confessed the recurrent fascination in his own readings
with opaque regions in texts, he is quick to point out that this is not, cus-
tomarily, the end of the matter: "I have said little so far about what be-
comes of these moments, of what comes after the end of the line. For there
is always an 'after': the structures we have been examining never coincide
with the conclusions of the particular texts in which they figure" (EL, p.
223). What succeeds them, often enough, is a moment of violence, in par-
ticular the restabilization of the momentarily fragmented self at the ex-
pense of some conveniently available other. Hertz speculates briefly over
the theoretical implications of this movement, drawing again upon de
Man and the latter's observation of "the violence associated with 'the auto-
biographical moment'" (EL, p. 223). But the resistance to theorizing is
itself too powerful in Hertz's critical operation to permit more than a
glance in this direction. He plunges directly back into some further "lit-
erary examples" in order to "look more closely at their economies" (EL, p.
223).
One could speculate on the motive for this unwillingness to theorize the
empirical encounters with texts that Hertz, de Man, and Miller all call
"reading." More productively, one recognizes the identical gesture in
these authors' own texts to that which they have located in the texts of
others. Hertz ends his book with two such exemplary moments, one from
The Prelude (and its echoing of a famous moment in The Tempest), the other
from Daniel Deronda, and observes: "Both texts exhibit the involution and
subsequent exfoliation that marks the turn at the end of the line" (EL, p.
239). Was it merely fortuitous that the word "line" here was printed, not at
the end of a line of type but on a separate line of its own? It's too neat a
conclusion not to have been the result of someone's intention, perhaps not
Hertz. Would it matter if this occurred merely by chance? On Hertz's view
of the organization and movement of texts, probably not, since the turn or
trope that resists narrative closure is inscribed in a variety of ways not
necessarily material.
Reading, for Hertz, arises from linguistic or poetic structures, at the
same time that it registers an inter-subjective relationship. What happens
in reading must occur regardless of the text's manifest material or phe-
nomenal features, for when one reads one is per definition exogenous to
the text and thus already turned away from the line of its development. In
varying degrees, reading can achieve a rapprochement between subjects
(see, e.g., the account of Milton's relation to Adam at that point in Paradise
Lost when the archangel Michael is telling the story of the Flood that will
destroy most of life on earth; EL, p. 34). But the inter-subjectivity consti-

This content downloaded from 181.231.196.143 on Fri, 01 Jun 2018 00:14:56 UTC
All use subject to http://about.jstor.org/terms
 M L N 1237

tutive of reading enforces a distance between text and reader that pre-
vents closure of the hermeneutic field, even in the apparently limiting case
(as in the various autobiographical texts Hertz considers) when author and
reader are the same figure: "because the frame separating the author/
reader from the work is permeable-and has to be, if the work is to be
read at all-a play develops between author/reader and text, regardless of
the degree of autobiographical likeness with which a particular surrogate
is represented . . ." (EL, p. 220). One could say that reading is rhetorical
because it expresses an inter-subjective relation of non-self-identity be-
tween text and reader, or that the inability of subjects to communicate
with or understand each other results from rhetorical properties of lan-
guage that prevent the meaning of any utterance from being strictly de-
limited. Either way one turns, the narrative line of reading and texts
cannot be terminated once and for all.
This moment, which recurs with disquieting frequency in the texts of
deconstructive criticism, would seem to correspond fairly precisely to what
Derrida calls difference. By showing how diffrrance manifests itself in partic-
ular texts, deconstructive criticism accomplishes that "infrastructural ac-
counting" of texts and the reversal of the hierarchy of their opposed terms
that Gasch6 has argued is the first (and indispensable) movement of de-
construction proper. This movement is perhaps most clearly and force-
fully in evidence in the work of Andrzej Warminski, for whom the re-
versal has a determinate name and a shape: the figure of chiasmus.
Gasch6 himself draws attenton to this feature in Warminski's book, enti-
tling his (Gasch6's) introduction to Readings in Interpretation, "Reading
Chiasms." Warminski's text is therefore situated squarely within the tradi-
tion of rhetorical readings which, as de Man maintained, when "techni-
cally correct ... may be boring, monotonous, predictable and unpleasant,
but they are irrefutable. They are also totalizing (and potentially totali-
tarian) for since the structures and functions they expose do not lead to
the knowledge of an entity (such as language) but are an unreliable pro-
cess of knowledge production that prevents all entities, including linguistic
entities, from coming into discourse as such, they are indeed universals,
consistently defective models of language's impossibility to be a model lan-
guage. They are, always in theory, the most elastic theoretical and dialec-
tical model to end all models and they can rightly claim to contain within
their own defective selves all the other defective models of reading-avoid-
ance, referential, semiological, grammatical, performative, logical, or
whatever" (RT, p. 19). It may therefore be apt to equate the activity of
rhetorical reading with deconstructive criticism as such and thus recognize
that the apparent recursiveness of the rhetorical structures it discovers is
subject neither for praise nor blame but merely points to a plausible fea-
ture of language use: semantic indeterminacy. The consequences one
infers from this fact of language may be variable, but it is useless, in the
absence of specific refutations of individual readings (pace de Man, any

This content downloaded from 181.231.196.143 on Fri, 01 Jun 2018 00:14:56 UTC
All use subject to http://about.jstor.org/terms
 1238 REVIEWS

reading, even a "technically correct" one, is in principle subject to refuta-

tion), to attempt to legislate the validity or invalidity of the insight. Like
any theoretical claim about an empirical phenomenon, this one must be
judged according to how closely the theory can be made to fit the evi-
dence.
But as we have previously observed, Gasch6 characterizes this decon-
structive gesture (infrastructural accounting and reversal of hierarchy) as
but the first movement in deconstruction proper. In Derrida's texts, this
preliminary stage is succeeded by the act of inscription, of rewriting the
terms of the original opposition in a new syntax that reorganizes the orig-
inal semantic field. Doubtless the most infamous of such reinscriptions is
the substitution of the "a" in differance. This gesture, while its necessity is
universal or generalizable (therefore part of a general theoretical project),
is nonetheless unpredictable in its particular form. One returns, at this
moment in the movement of deconstruction, to the irreducibly empirical
and literal event of reading. The protocols governing the reading of a
particular text cannot be determined in advance. This is the point at which
deconstruction would seem to intersect with marxism, for the randomness
and materiality that remain part of any theoretical investigation have a
name: history. De Man was never reluctant to say this much, despite his
life-long suspicion of teleological historicist schemes apparent in marxism
and conventional literary history alike. His suspicion is repeated in various
ways by Hillis Miller, Andrzej Warminski, and Derrida himself in the texts
we are discussing. In particular, Derrida takes to task "certain stereotyped
formalizations of 'late Marxism'" (M, p. 43, n. 8) for prematurely
equating the operation of deconstruction with a punctual moment in the
history of certain social formations hastily characterized by the term "late
capitalism." He has in mind the following formulation: "Deconstruction
... mirrors the effacement of ideology under the mantle of technical ra-
tionality which is the principal feature of ideology under late capitalism
... Deconstruction is the specular image of the society of the spectacle"
(cited in M, p. 43, n. 8).
I confess to having winced when I read these words, for they were my
own, quoted out of context and without my explicit permission from a
paper I once delivered orally and, for reasons that should be apparent, I
never attempted to publish. Derrida cites them against me (and not only
me), and I can scarcely blame him. I can't now take them back, much as I'd
like to, and so am condemned to watch them live on with a resilience that
is likely to embarrass me further in the future. Perhaps I may be allowed,
however, to modify them slightly. I shall in any event take heart from the
closing sentence in Derrida's note: "Fortunately all marxisms are not re-
duced to this." On the contemporary horizon, no marxism is less reducible
in this manner than that associated with the name of Althusser.
The relations between deconstruction and Althusserianism are, one
might say, complexly overdetermined. I have tried in another place to

This content downloaded from 181.231.196.143 on Fri, 01 Jun 2018 00:14:56 UTC
All use subject to http://about.jstor.org/terms
 M L N 1239

chart the parallel paths through the text of Rousseau taken by de Man and
Althusser, and to draw some of the implications from the ultimate diver-
gence of their projects.9 Derrida's case is somewhat different. It would
have been difficult for him to side-step Althusser entirely at the moment
and in the place where their mature work began to emerge, but Derrida's
encounters with Althusser (and through him with Marx) have tended to
be fleeting and generally guarded.'0 Althusser has returned the compli-
ment, citing Derrida with favor but avoiding any direct, sustained engage-
ment with the latter's project." However prudent this shared tact may
have been in the context of the highly charged and polemical atmosphere
surrounding Parisian intellectual life, in the aftermath of May '68, the gen-
eral silence reigning over these two figures with respect to the work of the
other has enabled, possibly even enforced, a critical incomprehension in
what has evolved (for historical and ideological reasons too complicated
for useful summary here) into two opposed theoretical camps.'2 I won't
pretend to be able in a review essay to bring what is in my view a needless
conflict to an abrupt end. These battle lines will surely twist and turn
many more times in the future. I can, however, suggest where the battle
might be more fruitfully joined than has been the case in the past. As
usual, the territory has been marked out in advance, if only in the most
provisional manner, by Paul de Man.
De Man's scathing criticisms of Althusser will be familiar to attentive
readers of Allegories of Reading. It will be recalled that de Man there takes
Althusser to task for misreading the text of Rousseau, above all for
missing its manifest rhetoricity. I won't take up here what I have at-
tempted to show in detail elsewhere: to wit, that de Man's fidelity to his
own project of demonstrating the irreducibly figural structure of Rous-
seau's texts has effectively blinded him to the extent of the convergence
between his account of the literary qualities in Rousseau and Althusser's
meditation on the circle of ideology that encloses Rousseau's conception of
politics and the state. De Man did, in any event, say rather different things
near the end of his life about Althusser. In the first place, Althusser was
included in a list of programmatically "aesthetic thinkers" (a list that in-
cluded Derrida and Marx), thinkers whose work "precludes, for example,
any valorization of aesthetic categories at the expense of intellectual rigor
or political action, or any claim for the autonomy of aesthetic experience
as a self-enclosed, self-reflexive totality."' 3 Further, in the interview he
granted just after delivering the Messenger Lectures in early 1983, de
Man discussed the itinerary of his own work on ideology, how it had taken
him back to Kierkegaard and Marx and to the problem of how to read
their texts: "they have to be read from the perspective of critical-linguistic
analysis to which those texts have not been submitted. There has been
very little on Kierkegaard along those lines and there has been even less
on Marx, except, of course, for elements in Althusser that, I think, go in
that direction" (RT, p. 121). I take the "I think" here to indicate that de

This content downloaded from 181.231.196.143 on Fri, 01 Jun 2018 00:14:56 UTC
All use subject to http://about.jstor.org/terms
 1240 REVIEWS

Man had not yet looked carefully enough into Reading Capital and For
Marx to know if the requisite protocols for reading Marx are produced
there. Nor is it certain that even after this investigation he would have
been satisfied with concepts like "structural causality" and "overdetermi-
nation" as acceptable instruments for unlocking the linguistic structures of
Marx's texts. While it is thus idle to speculate over what de Man would
have found in Althusser and what he would have said about the Althus-
serian intervention in the history of interpretation of Marx, I'll neverthe-
less hazard here an opinion about what he ought to have recognized: that
the lines of convergence between the double movement of deconstruction
and Althusserian theory would necessarily pass through the point of Alt-
husser's still largely neglected insight about the material existence of ide-
ology. In one of his later essays ("Hypogram and Inscription," in RT, pp.
27-53), de Man called this stubborn materiality of the sign that resists
ideological appropriation, inscription. What seems likely then, is that via
his own itinerary, to which he remained insistently faithful, an itinerary
which we have followed Gasch6 in calling deconstructive criticism, de Man
had arrived on the threshold of the second part of the movement of de-
construction. And that final gesture necessarily returns to the basic Alt-
husserian question of how the knowledge of history can be constituted on
a scientific footing, independent of its ideological appropriations. What
both deconstruction and Althusserianism lead us inexorably towards is a
science of the materiality of signs.
It would be premature tojudge whether this science is possible. Nothing
but further research into the theory and the historical manifestations of
ideology will provide even a provisional answer. Theoretical work, as de
Man's own career amply illustrates, is necessarily for the long haul. Nor is
it possible to carry on this work in isolation, either personal or geographic.
To the extent that there is "deconstruction in America," it is also, always
already, in Europe-and indeed across the globe. The enterprise of
theory is international, or it is nothing.

State University of New York at Stony Brook MICHAEL SPRINKER

NOTES

1 Rodolphe Gasche, The Tain of the Mirror: Derrida and the Philosophy of Reflection
(Cambridge, Mass.: Harvard University Press, 1986), p. 3; hereafter cited par-
enthetically as TM. The other texts under review here are cited parenthetically
as follows: M-Jacques Derrida, Memoires (for Paul de Man), The Wellek Library
Lectures, trans. Cecile Lindsay, Jonathan Culler, and Eduardo Cadava (New
York: Columbia University Press, 1986); EL-Neil Hertz, The End of the Line:
Essays on Psychoanalysis and the Sublime (New York: Columbia University Press,
1985); ER-J. Hillis Miller, The Ethics of Reading: Kant, de Man, Eliot, Trollope,
James and Benjamin, The Wellek Library Lectures (New York: Columbia Univer-
sity Press, 1986); RI-Andrzej Warminski, Readings in Interpretation: Hblderlin,
Hegel, Heidegger (forthcoming from University of Minnesota Press); RT-Paul de

This content downloaded from 181.231.196.143 on Fri, 01 Jun 2018 00:14:56 UTC
All use subject to http://about.jstor.org/terms
 M L N 1241

Man, The Resistance to Theory (Minneapolis: University of Minnesota Press,

1986). In the case of Warminski's and Miller's texts, a typescript and uncor-
rected galley proofs only could be obtained at the time of writing; page refer-
ences were therefore unavailable. References to Paul de Man's forthcoming col-
lection from the University of Minnesota Press, Aesthetic Ideology, will be to the
already published versions of the essays that will comprise it. I have had the
benefit of consulting two other typescripts to be included in this collection:
"Kant's Materialism," and "Schiller and Kant."
Work on this essay would have been impossible without the generous assis-
tance of Rodolphe Gasch&, Brenda Jokisalo, Adrienne Macaulay, J. Hillis
Miller, and Andrzej Warminski, all of whom provided me with materials in
advance of publication. I have also had the opportunity to read the typescripts
of two superior expositions of de Man's work: Rodolphe Gasche's "In-differ-
ence to Philosophy: De Man on Kant, Hegel, and Nietzsche" (scheduled to ap-
pear in the collection, Reading Paul de Man Reading, to be published by the
University of Minnesota Press); and Werner Hamacher's "De Man's Impera-
tive" (the English translation I consulted was done by Susan Bernstein; the text
will appear in German as the introduction to the German translation of de
Man's Allegories of Reading, to be published Suhrkamp Verlag).
2 The point has been argued at great length by Richard Rorty; see his Philosophy
and the Mirror of Nature (Princeton: Princeton University Press, 1979), and Con-
sequences of Pragmatism (Minneapolis: University of Minnesota Press, 1982).
3 I say this despite Derrida's warnings about the term "deconstructive criticism"
(see M, p. 84). If deconstruction does not, can never coincide with the work of
'criticism," then all the more reason to call Derrida's patient exposition of de
Man just that: it falls short of deconstruction to the degree that it remains crit-
ical and expository. No one ever said deconstruction was possible.
4 This is not to say that Derrida's text simply reproduces the movement of the-
matic criticism here. But the play in Derrida's discourse about de Man involves
less the terms and the syntax of the de Manian text than the difficulty (which
was both theoretical and personal for Derrida) presented by the occasion itself:
the task to speak in memory of Paul de Man. "Memory," while it is not an
irrelevant concept to the de Manian text (above all in a late essay like "Sign and
Symbol in Hegel's Aesthetics), is hardly the place where one would go to unlock
its structure.
5 Paul de Man, Allegories of Reading: Figural Language in Rousseau, Nietzsche, Rilke,
and Proust (New Haven: Yale University Press, 1979), p. 19; hereafter cited par-
enthetically as AR.
6 A possibly inapposite observation: Derrida's exposition of de Man is over-
whelmingly focused on the later texts. The first lecture, after a glance at de
Man's essay on Holderlin and Heidegger from the 1950s, deals primarily with
two texts from de Man's final years ("Anthropomorphism and Trope in the
Lyric" and "Autobiography as Defacement"). The second deals extensively with
texts from Blindness and Insight (including "The Rhetoric of Temporality,"
which was incorporated into the augmented edition published in 1983), but in
the context of de Man's mature formulations on memory and interiorization in
"Sign and Symbol in Hegel's Aesthetics," first published in 1982. (An unavoidable
interruption: the truly instructive break that can be charted with relative ease in
de Man's work occurs around the figure of Hegel. If one compares the deploy-
ment of Hegel in the still unpublished chapter of de Man's Ph.D. thesis on
Mallarme with de Man's two published essays on Hegelian aesthetic and lin-
guistic theory, the shift, from a vocabulary of consciousness and a concept of
time as intentional to the non-temporal dimensions of language as trope and
the materiality of signification, is patent. The residue of the more archaic ver-

This content downloaded from 181.231.196.143 on Fri, 01 Jun 2018 00:14:56 UTC
All use subject to http://about.jstor.org/terms
 1242 REVIEWS

sion of Hegel is visible in many of the essays in the original Blindness and Insight,
as well as in "Intentional Structure of the Romantic Image" and "The Rhetoric
of Temporality." No one, to my knowledge, has explored this stratum of the de
Manian text.) The third lecture takes up de Man's prefaces (all written after the
decisive moment of "The Rhetoric of Blindness") and the most mature sections
of Allegories of Reading. The temptation to see in this procedure the recurrence
of an all too familiar historical gesture is unavoidable. Has Derrida imposed on
de Man's career a coherence and trajectory that make the growth of the critic's
mind appear continuous and without significant interruption? Derrida's own
puzzlement over this question is revealing: "In the case of Paul de Man, as much
as in that of 'deconstruction in America,' the 'always already' which tends to
erase the singularity of the event is erased in its turn before the signature of this
word [deconstruction]. As precarious as this signature is, it asserts itself as his-
tory insofar as the origin of its 'taking-place' is unlocatable. I do not have a
formalizable answer to this question. But it is posed to us by the history of
deconstruction and by history as deconstruction" (M, p. 125). This unformaliz-
able answer to the question of historicity marks the limits of deconstruction as a
method. We shall return to this topic subsequently.
7 Rodolphe Gasch6, "Deconstruction as Criticism," Glyph 6 (1979): 206-9.
8 Paul de Man, Foreword to Carol Jacobs, The Dissimulating Harmony (Baltimore:
Johns Hopkins University Press, 1978), p. xi.
9 Michael Sprinker, Imaginary Relations: Aesthetics and Ideology in the Theory of His-
torical Materialism (forthcoming from Verso Editions), chapter 9.
10 See Jacques Derrida, Positions, trans. Alan Bass (Chicago: University of Chicago
Press, 1981), pp. 57-58, 63; and idem, Margins of Philosophy, trans. Alan Bass
(Chicago: University of Chicago Press, 1982), p. 214, n. 10.
11 See Louis Althusser, Politics and History, trans. Ben Brewster (London: NLB,
1972), p. 184.
12 This is evident, for example, in Michael Ryan's general dismissal of Althusser-
ianism in Marxism and Deconstruction (Baltimore: Johns Hopkins University
Press, 1982), as it is in Perry Anderson's precipitous treatment of Derrida in In
the Tracks of Historical Materialism, The Wellek Library Lectures (London: Verso,
1983).
13 Paul de Man, "Hegel on the Sublime," in Displacement: Derrida and After, ed.
Mark Krupnik (Bloomington: Indiana University Press, 1983), p. 141.

Hans Blumenberg, Work on Myth. Trans. Robert M. Wallace.

Cambridge, Mass.: MIT Press, 1985. xl + 685 pages.

"Nothing surprised the promoters of the Enlightenment more, and left

them standing more incredulously before the failure of what they thought
were their ultimate exertions, than the survival of the contemptible old
stories-the continuation of work on myth" (274).
In the 6th century B.C., Thales taught that everything was full of gods.
A professor of philosophy at the University of Mtinster, Hans Blumen-
berg writes that everything is now full of theories (167). Between these
views lies the work on myth, a perpetual labor.

What after all does the disposition toward mythical ways of looking at things
consist in and why is it not only able to compete with theoretical, dogmatic, and

This content downloaded from 181.231.196.143 on Fri, 01 Jun 2018 00:14:56 UTC
All use subject to http://about.jstor.org/terms