In response to:

The Word Turned Upside Down from the October 27, 1983 issue

To the Editors:

John R. Searle’s review of Jonathan Culler’s On Deconstruction [NYR, October 27] looks like a critical but unbiased examination of deconstructive theory. One need only read it carefully to discover that appearances are, in this case at least, almost perfectly deceiving.

Early in his review (p. 74) Searle wonders if Derrida and Culler believe there is a “vulgar concept of woman” corresponding to the “vulgar concept of writing.” That Derrida thinks so is evident from, among others, a text like Spurs. And if one will read the first chapter of Culler’s book, she will (not “reasonably assume” but) out and out know that Culler thinks there is.

In a footnote (p. 75) Searle suggests that perhaps Derrida and Culler are not aware that the ancient Greeks read aloud. It is unthinkable (by anyone not out to get them) that either Derrida or Culler is bereft of this commonplace. But in his footnotes (it might be fun to deconstruct them) Searle casts off even the minimal constraints he accepts in the body of his essay. For example, another note (p. 77) cites two statements about truth and finds them inconsistent. One of them (“truth is a kind of fiction”) Culler does not make. What he says (Culler, p. 181, my emphasis) is “if truths are fictions whose fictionality has been forgotten,” a clause which Searle quotes in the body of his review without the “if.” In the passage from which this citation is taken Culler is developing the implications for literary criticism of deconstructive philosophy. The other statement (“Truth is what can be demonstrated…and what is simply the case,” Culler, p. 154) is meant to characterize the traditional logocentric view of truth. Read in context, the two statements do not contradict each other. Neither is asserted by the author. (It might be of interest to Searle that the apparent paradox, “truths are fictions whose fictionality has been forgotten” was stated and persuasively argued by Nietzsche in his essay, “On Truth and Lie in an Extra-Moral Sense.”)

In the same vein Searle quotes (p. 77) this passage from Culler’s book:

what we think of as the innermost spaces and places of the body—vagina, stomach, intestine—are in fact pockets of externality folded in [Culler, p. 198]

and remarks that “anatomists will no doubt be interested to learn” this. But Culler is not confronting anatomists with a revelation. Like Derrida, he is using this familiar fact to illustrate the impossibility of “framing” a text or a genre of texts; i.e., of demarcating its inside clearly and unambiguously from its outside. (Cf. Derrida, “The Law of Genre,” Glyph 7.) Culler’s observation bears not on anatomy but on the sort of formalism that thinks it can isolate a literary work or a literary kind, seal it off from its environment so as to prevent bleeding in or out, and treat it as a self-contained entity.

According to Searle (p. 77) Culler asserts that presence is a certain type of absence. On the page to which Searle refers (Culler, p. 106) Culler is discussing Derrida’s discussion of Rousseau’s Confessions. What he says is that “Maman’s ‘presence’ is a certain type of absence.” The quotes around “presence” are Culler’s own. Searle says that Culler says that the marginal is in fact central. What he actually says is that “what had previously been thought marginal is in fact central” (Culler, p. 140, my emphasis). This sort of thing goes on and on. Searle’s Culler says that the literal is metaphorical. Culler himself says, “The literal is the opposite of the figurative, but the literal expression is also a metaphor whose figurality has been forgotten” (Culler, p. 148). Searle’s Culler says that reading is a form of misreading. Culler speaks of “the claim that all readings are misreadings” (Culler, p. 176, my emphasis) and tries to show how the claim might be justified. The same goes for “understanding is a special case of misunderstanding” (Searle, p. 77, Culler, p. 176). “Sanity is a kind of neurosis” (p. 77). But Culler, on the page cited (160), is expounding Freud, and what he writes is that “it has become something of a commonplace that ‘sanity’ is only a particular determination of neurosis.” The quotes around “sanity” are, again, Culler’s. That “man is a form of woman” occurs—not in those words, naturally—in a passage (Culler, p. 171) in which Culler is exploring some implications of Freud’s theories of sexuality. What Culler thinks “perhaps one should say, in keeping with the Derridean model,” is “that man and woman are both variants of archi-woman.” In all these cases the appearance of absurdity is Searle’s own invention, an illusion he throws up in order to ridicule and defame the work he is supposed to be reviewing. It is unfortunate that Searle was chosen to review a book like On Deconstruction. Culler deserved at least a fair shake.


But Searle’s inability (or unwillingness) to read fairly is only one of his bad habits. For example, he accounts for Derrida’s “eccentric” reading of the history of Western philosophy by explaining (p. 76) that Derrida discusses only three major philosophers in any detail: Plato, Rousseau, and Husserl. Without going back to the texts I can recall a whole string of other philosophers discussed in some detail by Derrida: Nietzsche, Heidegger, Kant, Hegel, Condillac, Bataille, Levinas, J.L. Austin, and (yes) John R. Searle. Searle’s “distressing penchant for saying things that are obviously false” (Glyph 1, p. 203) may indicate that he is “simply misinformed” (p. 75). But perhaps Searle thinks that not all the people on my list are “major figures.”

Derrida’s proof that writing is prior to speech is, according to Searle, “based on a redefinition” (p. 76) of writing. But in “Signature Event Context,” an essay to which Searle responded in the pages of Glyph, Derrida begins with the ordinary concept of writing and shows that all the properties by which this concept distinguishes writing from speech—particularly those that mark writing as secondary, derivative, and debased—are also properties of speech as ordinarily understood. According to Searle, Derrida’s concept of writing is not based on “actual empirical study.” He makes nothing of the fact that speech is spoken and writing written (what could be made of that?), and he does not attend to the persistence of written texts through time (p. 77). It is true that Derrida’s concern is structural, not empirical. And yet he is (of course) familiar with the facts Searle mentions—some of them are discussed in his reply to Searle (Glyph 2)—and shows how they can be explained (or discounted) by his concept of the structural coincidence of speech and writing. Derrida’s view of writing and its relation to speech is not a generalization from experience, but it does account for the facts. Better (in fact) than does the traditional view.

In defense of his critique of Derrida, Searle advances two claims which, taken together, actually support deconstruction. Here is the first:

The distinction between speech and writing is simply not very important to Plato, Aristotle, Aquinas, Descartes, Kant, Spinoza, Leibniz, Hume, etc. [p. 75]

The second claim is that, from medieval developments of Aristotle’s logic down to modern symbolic logic,

philosophers have tended to emphasize written language as the more perspicuous vehicle of logical relations. [p. 75]

Now, if these claims are true (I shall not try to decide that), they strengthen rather than weaken the case for deconstruction. The first claim (like my mother’s refusal to regard libido as seriously different from affection) would provide evidence for what Derrida calls the repression of writing in the West. The second would testify to the philosophers’ continued dependence (nevertheless) on the writing whose difference (or differance) they had tried to repress.

More than once, what Searle offers as arguments against deconstruction play right into (and out of) Derrida’s hand. His contention (p. 78) that deconstruction itself is subject to deconstructive analysis is not exactly original with him. Derrida knows that deconstruction depends on (inhabits) the structures it deconstructs, and he has insisted from the first that every deconstructive reading needs deconstructing ad infinitum. But this is not the first time Searle has used Derrida to whip Derrida: cf. their “confrontation” in Glyph 1 and 2. Neither is it inconsistent for a deconstructionist to allege that he is misunderstood, as Searle thinks on the strength of a confidence from Michel Foucault (p. 77). Deconstructive reading is not meant to replace conventional reading, but to supplement it—or to exhibit its own supplementarity. Deconstructive reading presupposes conventional reading. How can one show that a discourse “undermines the philosophy it asserts” (p. 74, quoted from Culler, p. 86) unless one has first understood the philosophy it asserts? This again is a point which Derrida has made, e.g., in Of Grammatology.

Sometimes it appears that Searle has misunderstood Derrida. Derrida, thinking of Saussure, says that “the play of differences” in language forbids that “a simple element be present in and of itself, referring only to itself” (Positions, 26, quoted on pp. 75-76). Searle counters:

I understand the differences between the two sentences “the cat is on the mat” and “the dog is on the mat” in precisely the way I do because the word “cat” is present in the first while absent from the second, and the word “dog” is present in the second, while absent from the first. [p. 76]

But surely the “presence” of “elements” like “cat” and “dog” in Searle’s sentences is toto caelo different from their presence in and of themselves—their exclusive self-reference—which is what Derrida is talking about. Searle has missed the point because he has mistaken the target.


Searle intones the ritual complaint that deconstructionists are not “seeking the truth” (p. 77). What this complaint fails (or refuses) to recognize is that the deconstructionist is almost obsessively occupied with truth. By deconstructing the traditional notion of truth he hopes to achieve a deeper and sharper understanding of the meaning of the search for truth, the conditions of its possibility, and what would count as success in such an endeavor. Searle (p. 77) chides Culler for this:

The effect of deconstructive analyses, as numerous readers can attest, is knowledge and feelings of mastery. [Culler, p. 225]

But he does not quote the sentence just preceding this one:

Demonstrations of complicities between language and metalanguage, observed and observer, question the possibility of attaining a principled mastery of a domain but do not suggest that deconstruction has either achieved a mastery of its own or can ignore the whole problem of mastery from a secure position of externality. [Culler, p. 225]

Nor does he quote the sentence that follows a few lines further on and concludes the chapter:

And if the formulations produced by these analyses are themselves open to question because of their involvement with the forces and ruses they claim to understand, this acknowledgement of inadequacy is also an opening to criticism, analysis, and displacement. [Culler, p. 225]

One wonders just who is seeking the truth.

One final (for now) point. When Searle suggests (pp. 77-78) that deconstruction is part of the larger twentieth-century attack on “foundationalism,” he is not altogether wrong. There are similarities. But it is misleading (a) to domesticate Derrida within this peculiarly Anglo-American problematic (he has a distinct, if related, ax to grind), even more misleading (b) to imply that the issue of “foundationalism” has been decided (it hasn’t), and wildly misleading (c) to suppose that if we simply relax our criteria the difficulties will be disarmed. As if we could take the sting out of adultery by relativizing our concept of fidelity! Perhaps those “rigid” literary critics Searle deplores as “positivists” have learned from Derrida that problems do not go away when they are deconstructed. They return with an agonizing vengeance and will not be pacified by the mere assertion (a nervous gesture of “power” on the part of the anti-foundationalist) that they aren’t really there. Opposing the need for foundations to the groundlessness of all “foundations,” deconstruction does not dismiss nor trivialize but rather intensifies and exacerbates that “love of wisdom” by which Western philosophy continues—not the least in thinkers like Derrida—to define itself.

I do not mean to suggest that Derrida and Culler are beyond criticism. I only intend the obvious: that any responsible critique of their work must proceed from a careful and (initially at least) sympathetic reading of their texts. There are many more infelicities, confusions, and outrages to be found in Searle’s review. I would need a leave of absence and many more pages to expose them all. But perhaps the few instances I have adduced will warn your readers not to take Searle’s estimate of Culler and Derrida at face value, but to go to the texts and judge for themselves.

Louis H. Mackey

The University of Texas

Austin, Texas

John Searle replies:

In my review, I made several objections to Culler’s book and to at least certain aspects of Derrida’s philosophy, from which Culler’s views are derived. If I am right, these objections are devastating. In his defense of deconstruction Professor Mackey does not, for the most part, attempt to answer my objections; rather, he changes the subject; he distracts attention from the manifest weaknesses of deconstruction by presenting a series of claims to the effect that I have misrepresented and misunderstood Culler and Derrida. I fully anticipated that this objection would be made, and so I gave copious verbatim quotes from both authors; and where I summarized their views, I gave extensive page references.

Mackey points out correctly that the summaries are not verbatim quotes, but he claims incorrectly that both they and my choice of quotations are misrepresentations. His strongest example is this: I quote Culler as holding the view that “Truths are fictions whose fictionality has been forgotten.” Mackey claims that Culler is not asserting this view, since the clause occurs following an “if.” But consider the whole context:

Deconstruction’s demonstration [my italics] that these hierarchies are undone by the workings of the texts that propose them alters the standing of literary language. If serious language is a special case of the nonserious, if truths are fictions whose fictionality has been forgotten, then literature is not a deviant, parasitical instance of language. On the contrary, other discourses can be seen as cases of a generalized literature, or archi-literature. [p. 81]

This passage is part of Culler’s attempt to develop the “consequences for literary theory and criticism of the deconstructive practice we have been expounding” (my italics, p. 180). I believe it is quite obvious from the context that Culler is asserting (among other things) that the deconstructive practice he has been expounding has “demonstrated” that “truths are fictions whose fictionality has been forgotten.”

Sometimes Mackey’s arguments seem to border on sophistry. For example, I summarize Culler as holding the view that all readings are misreadings. Mackey counters, “Culler speaks of the claim that all readings are misreadings.” But here is the actual sentence in Culler’s book: “The claim that all readings are misreadings can also be justified by the most familiar aspects of critical and interpretive practice” (Culler, p. 176, my italics). And in the very same paragraph, Culler goes on: “every reading can be shown to be partial…. The history of readings is a history of misreadings….” Indeed this thesis occurs in several places in Culler’s book, e.g., “The best a reader can achieve is a strong misreading” (pp. 79-80).

Mackey says that my attributing to Culler the view that understanding is a special case of misunderstanding is similarly a misinterpretation. Once again let us look at the actual sentence in Culler:

We can thus say, in a formulation more valid than its converse, that understanding is a special case of misunderstanding, a particular deviation or determination of misunderstanding…. [p. 176]

I claim that Culler holds the view that the literal is metaphorical. Mackey cites the actual sentence:

The literal is the opposite of the figurative, but the literal expression is also a metaphor whose figurality has been forgotten [my italics].

Mackey thinks that “the appearance of absurdity” in Culler’s views is an “invention” resulting from my summaries. In fact the verbatim quotations are no less and no more absurd than the summaries I gave. If Mackey thinks it makes a difference, then he can simply substitute the verbatim sentences for the summaries. For example, where I had “sanity is a kind of neurosis” put in “it has even become something of a commonplace that ‘sanity’ is only a particular determination of neurosis…” (Culler, p. 160). The first was my summary; the second was Culler’s actual wording. Mackey finds the first absurd. What does he think of the second?

I believe that a careful reading of Culler’s book will show that he is committed, implicitly or explicitly, to every view I attribute to him. But suppose we try to accept for the sake of argument Mackey’s implication that Culler himself doesn’t really believe the implausible views I attributed to him. He doesn’t really think that “speech is a form of writing; presence is a certain type of absence, the marginal is in fact central, the literal is metaphorical, truth is a kind of fiction, reading is a form of misreading, understanding is a form of misunderstanding, and man is a form of woman” (p. 77). Then, the book would lose much, if not all, of its point, since the whole point of the book, as Culler states repeatedly, is to expound deconstructive strategies and show their value for literary criticism. Now the first and most important strategy, according to Culler, is the deconstruction of the traditional hierarchical oppositions between truth and fiction, between the serious and the nonserious, between understanding and misunderstanding, sanity and neurosis, man and woman, etc. And in every case the first step in the deconstruction is to show that the left-hand term is really a specialized case of the right-hand term. I summarize the results of these efforts. Mackey would have us believe that Culler is not even making the efforts.

Actually, the effect of Mackey’s complaint is to remind us of a point that I should have made more strongly in my review. Deconstructive prose tends to be systematically evasive. Several of the examples that Mackey gives are typical of this evasiveness. Crucial words are put in quotation marks so as to suggest an ambivalence in the author’s stance toward them. Thus, for example, ” ‘sanity’ is only a particular determination of neurosis.” Or central theses are imbedded in subordinate clauses and not stated directly, as in “if truths are fictions whose fictionality has been forgotten….”* In this way the deconstructionist can make implausible assertions while appearing not to; and this is part of what Culler calls the deconstructionist’s “nimbleness,” his moving “in and out of philosophic seriousness” (p. 155).

Mackey’s letter also serves to remind us of some other gimmicks used by deconstructionist writers. In addition to systematic evasiveness, they also commonly use the “heads, I win; tails, you lose” form of argument. For example, I claim that the speech/writing distinction is not very important in the history of Western philosophy, and I submit as evidence the fact that Plato, Aristotle, Aquinas, Descartes, Kant, Spinoza, Leibniz, Hume, etc., have very little to say about it. I also point out that “philosophers have tended to emphasize written language as the more perspicuous vehicle of logical relations.”

Now, the first of these points is taken by Mackey as evidence that these philosophers have repressed writing; and the second as evidence for the philosophers’ “continued dependence (nevertheless) on the writing whose difference (or differance) they had tried to repress.” Notice the form of Mackey’s argument here. The lack of evidence for Derrida’s thesis is itself taken as a form of evidence since it proves that the authors in question are trying to repress the phenomenon Derrida has uncovered, their urge to repress writing. And the fact that they are manifestly not repressing writing, the fact that written language is, e.g., the basis of modern mathematical logic, is taken as showing only that their valiant efforts at repression have failed. By such methods of reasoning one can “prove” absolutely anything. And as Mackey states, Derrida uses these sorts of methods over and over.

In addition to the devices of systematic evasiveness, and the heads-I-win-tails-you-lose form of argument, there is a third strategy exemplified in Mackey’s letter, and that is the straw-man argument. In my review I pointed out that the Saussurean claim that language is a system of differences does nothing whatever to undermine the distinction between presence and absence, because the system of differences is precisely a system of presences and absences. “I understand the differences between the two sentences ‘the cat is on the mat’ and ‘the dog is on the mat’ in precisely the way I do because the word ‘cat’ is present in the first while absent from the second, and the word ‘dog’ is present in the second, while absent from the first.” Mackey counters that what Derrida is really talking about is “their presence in and of themselves—their exclusive self-reference.”

This is an example of the obscurantism that pervades deconstructive prose. What is it for words to be present “in and of themselves” except for them to be present? The explanation Mackey and Derrida give in terms of “exclusive self-reference” is worse than no help at all. If we consider the two sentences in our example as abstract sentence types, apart from any particular context of utterance, then the words don’t refer to anything. If we consider the sentences as occurring in actual token utterances, in particular real life contexts, then the expression “the dog” would normally be used to refer to a dog, and the expression “the cat” would normally be used to refer to a cat. The topic of self-reference is one that logicians and philosophers have been discussing now for over three-quarters of a century; and no one to my knowledge has ever supposed that examples of this sort, whether as tokens or as types, had “exclusive self-reference.” Once the obscurity of the prose is stripped away, Derrida’s apparently startling claim that words are never “simply present” is now converted into the triviality that they don’t normally have “exclusive self-reference,” a thesis no one has ever denied.

The single most implausible claim that Mackey makes is that “the deconstructionist is almost obsessively occupied with truth.” If he means to imply that they seek the truth, then a purely textual analysis of the works that I have cited would show that that is simply not the case. Authors who are concerned with discovering the truth are concerned with evidence and reasons, with consistency and inconsistency, with logical consequences explanatory adequacy, verification and testability. But all of this is part of the apparatus of the very “logo-centricism” that deconstruction seeks to undermine.

Mackey thinks that it is “a familiar fact” that “what we think of as the innermost spaces and places of the body—vagina, stomach, intestine, are all in fact pockets of externality folded in.” But this is not a “familiar fact” at all. Unless Mackey is referring to very early stages of fetal development, the claim is just anatomically unintelligent. And such bad anatomy offers no support one way or the other for theories about literary formalism. He also claims that it is “unthinkable” that Culler and Derrida should be “bereft of [the] commonplace” that the Greeks of antiquity only read aloud. Unthinkable or not, they certainly give no indication of knowing that the relation of reading to speaking in antiquity was quite different from that of the modern era. And in any case it is hardly a “commonplace”; indeed, the precise extent to which it is true is in dispute in the two works I cited (p. 75).

At bottom Mackey’s real objection to my discussion of deconstruction is that I am not sympathetic to it. But there are reasons for my lack of sympathy. I believe that anyone who reads deconstructive texts with an open mind is likely to be struck by the same phenomena that initially surprised me: the low level of philosophical argumentation, the deliberate obscurantism of the prose, the wildly exaggerated claims, and the constant striving to give the appearance of profundity by making claims that seem paradoxical, but under analysis often turn out to be silly or trivial. In my review, I gave examples of all these phenomena. There is an atmosphere of bluff and fakery that pervades much (not all, of course) deconstructive writing. What becomes even more surprising is that the authors seem to think it is all right to engage in these practices, because they hold a theory to the effect that pretentions to objective truth and rationality in science, philosophy, and common sense can be deconstructed as logocentric subterfuges. To put it crudely, they think that since everything is phony anyway, the phoniness of deconstruction is somehow acceptable, indeed commendable, since it lies right on the surface ready for further deconstruction. Thus, the general weaknesses of the deconstructive enterprise become self-justifying. With such an approach I am indeed not sympathetic.

This Issue

February 2, 1984