INTRODUCTION—
RESISTANT STRUCTURES
In describing his experience in teaching Humanities 6 at Harvard, Paul de Man spoke of the power of "mere reading" prior to any theory. He spoke of "the bafflement" that "singular turns of tone, phrase, and figure were bound to produce in readers attentive enough to notice their non-understanding behind the screen of received ideas."[1] This book is a defense of the possibility and the desirability—though not the inevitability—of such bafflement. I take seriously the idea that "reading" prior to any theory is, as de Man suggests, a strong and distinct experience. The "resistant structures" in the title of this book are, in the first instance, the structures of and in particular texts that produce "bafflement," that surprise or puzzle the reader on a large or small scale, and that in some sense resist assimilation to totalizing interpretive strategies or methods. I agree with de Man's view that such structures are, in a perfectly intelligible and defensible sense, in the texts, that texts are "bound to produce" them in readers, as Dr. Johnson would say, "uncorrupted with literary prejudices." Where I disagree with de Man is that I do not believe that the resistance of particulars to theories can itself be theorized; I do not believe that this resistance always points to the same scandal or deep truth that "it is the more or less secret aim of literary teaching to keep hidden."
[1] Paul de Man, "The Return to Philology," The Resistance to Theory , with a foreword by Wlad Godzich (Minneapolis: University of Minnesota Press, 1986), 23. The quotations from de Man in this paragraph all occur on pp. 23–24.
I do not believe, in other words, that we can know in advance what sort of bafflements texts will produce, or even that texts will produce them. What this book is against is any sort of approach to texts that knows in advance what they will or must be doing or saying, or, on the other hand, what they cannot possibly be doing or saying. De Man's knowledge of the nature of the bafflement that texts will inevitably produce is, in my view, one of the a priori approaches that dictate to particulars how they must behave, that tell critics what they must find when they look closely at texts. I want to argue for the desirability of approaching individual texts with as few presuppositions—theoretical and historical—as possible.[2] The more that the critic knows in advance what a text must or cannot do, the less reading, in the strong sense, will occur. I have taken Wittgenstein's "Don't think, but look" as the guiding motto of this book.[3] I want to be as skeptical as possible toward general claims about literature, authors, or periods, while being, on the other hand, extremely nonskeptical, as I have already suggested, about the "objective" existence of facts and structures prior to theories. "Don't think, but look" makes no sense as an exhortation if "looking" can never reveal anything that "thinking" does not.
It is sometimes said that there are no facts independent of theories. Something like this may—with enough of an account of "theories"—be true, and it is certainly true that there is no "world" without language. But this does not mean that there are no facts independent of particular theories, or that there are no facts that many different "theories" all presuppose. Literary approaches, with the exception of deconstruction, are often not "theories" in any very strong sense, and certainly particular interpretations are not "theories." There are many features of literary
[2] It has been pointed out to me that this sentence could be read as a plea for ignorance, for, as one colleague put it, "know nothingism." I understand this possibility, but I mean to be arguing for a methodological principle of something like "docta ignorantia " ("learned ignorance"), a procedure that suspends many things that one knows (or thinks one knows) in favor of an initial experience of "mere reading." Only after this has been done, I would argue, can knowledge be brought fruitfully to bear on the text (and also, I would add, be tested by it). I return to this point of procedure toward the end of Essay 1. For an application of this procedure to two historical "documents" (and a defense of this application), see Richard Strier, "From Diagnosis to Operation: The Roots and Branches Petition and the Grand Remonstrance," in David L. Smith, Richard Strier, and David Bevington, eds., The Theatrical City: London's Culture, Theatre and Literature, 1576–1649 (Cambridge: Cambridge University Press, 1995).
[3] Ludwig Wittgenstein, Philosophical Investigations, ed. and trans. G. E. M. Anscombe (New York: Macmillan, 1953), 31e. The German is "denk nicht, sondern schau! " (31).
texts that all "theories" would agree on. Agreeing to these does not mean subscribing to a theory but merely being a competent user of the language in which the texts are being discussed. These features may be thought to be trivial or obvious, but that is neither here nor there. Or rather, I would argue, it is very much here. Claims about texts that get these features wrong, that misquote, miscount, etc., are never taken to be viable. Differing interpretations of a text generally share a large number of particular agreements before they part company. And when they part company, they are still responsible to the features—I would call them facts—that they share. Interpretive conclusions, even widely held ones, do not become facts. That Hamlet delays in killing Claudius is a fact; that Hamlet is neurotic (or whatever) in doing so is not. This book means to defend the importance of the obvious, the surface, and the literal as well as the particular. It means to look very hard at moments when a critic or scholar dismisses or downgrades (or misrepresents) the obvious, the "surface," or the literal. Since I believe that these features of texts have, and ought to have, a privileged place in interpretation, and that these features have a reality independent of (particular) theories, the moments when they are denigrated are often, I will argue, moments when an a priori scheme or conception can be seen at work.[4]
[4] I have quite consciously written this paragraph (until now) without footnotes. I want to take full responsibility for the position that I have enunciated, and I want the positions that I have attacked and defended to stand or fall on their own, without being tied to competing authorities. Too often, I think, in discussions of this sort, names of holders of positions stand in for the substance of positions. The reader who is familiar with these discussions will certainly have recognized the relations of the assertions in this paragraph to the following (and the reader who is unfamiliar with these discussions may wish to consult the following). For an application of (something like) the skeptical position that I am attacking to matters of textual interpretation, see the essays in Stanley Fish, Doing What Comes Naturally: Change, Rhetoric, and the Practice of Theory in Literary and Legal Studies (Durham: Duke University Press, 1989), and for the philosophical basis of the position, see the essays in Richard Rorty, Consequences of Pragmatism (Minneapolis: University of Minnesota Press, 1982); for (something like) the position that I am defending, see the essays in Donald Davidson, Inquiries into Truth and Interpretation (Oxford: Clarendon Press, 1984). It should perhaps be noted that Davidson has been cited on both sides of this controversy. For reflections on this apparent paradox, see Christopher Norris, "Reading Donald Davidson: Truth, Meaning, and Right Interpretation," in Deconstruction and the Interests of Theory (Norman: University of Oklahoma Press, 1989), 59–83. Davidson has responded (negatively) to Rorty's use of Davidson's work in "A Coherence Theory of Truth and Knowledge," in Ernest LePore, ed., Truth and Interpretation: Perspectives on the Philosophy of Donald Davidson (Oxford: Blackwell, 1986), 307–19. For some other essays suggesting the relevance of Davidson's work to literary criticism, see Reed Way Dasenbrock, ed., Literary Theory after Davidson (University Park: Pennsylvania State Press, 1993). Also in the background of my thinking is an essay by William C. Wimsatt on "Robustness, Reliability, and Overdetermination," in M. Brewer and B. Collins, eds., Scientific Inquiry and the Social Sciences (San Francisco: Jossey-Bass, 1981), 124–63.
I do not mean to suggest that we should (or could) read without hypotheses or that hypotheses cannot reveal hitherto unnoticed features or aspects of texts. I am interested in moments when texts resist even very brilliant, illuminating, and well-founded hypotheses. I believe that these moments are often marked in criticism by a rhetoric of discounting: "This seems to be doing or saying X, but it is really doing or saying Y," or "It doesn't matter that the text seems to be doing X because," etc. My aim is not to discourage critics and scholars from forming hypotheses and from finding confirmatory evidence for them. My aim is to encourage a certain modesty in the scope that is claimed for critical or historical insights. A valid insight, an insight that explains or illuminates a great deal in a text or a period, does not have to explain everything and be everywhere valid. I do not wish to encourage critics to work less hard or to know less but only to resist the final turn of the screw, the moment when resistance in the text is overcome rather than acknowledged. The thrust of this book is to be monitory rather than negative. Tempting as I find the position, I do not mean to assert with Groucho Marx that "Whatever it is, I'm against it." I am wary of the totalizing impulse that tempts interpretive and explanatory hypotheses. By resisting our totalizing impulses and acknowledging where texts offer resistance to us, we gain the possibility of surprise and, most of all, the experience of variety. That all texts, or all texts from a particular period or of a particular kind, do not always mean or do the same thing has got to prove, ultimately, a more deeply satisfying and humanizing pleasure than that which derives from finding the same thing—the thing we expected—everywhere. That the recognition of genuine variety is ethically (and politically) more admirable than the denial of difference does not, of course, make the position that produces the admirable recognition true, but it is a happy consequence.
As I have already suggested, a priori views of texts can be divided, roughly speaking, into two groups: those which derive from general schemes and those which derive from particular assertions. This division corresponds to the two sections of this book. The first section, comprising Essays 1 through 4, deals with general schemes that mandate what a text (or a valuable text) from a particular period must or must not
do or mean. The section moves from examining a preemptive appeal to scholarship to settle issues ("Tradition") to two essays that examine closely related general schemes for how texts (or valuable texts) work ("self-consumption" and deconstruction) to an essay on an approach or praxis, "New Historicism," that falls into an untenable behaviorism.
The second section, "Against Received Ideas," deals with specific ideas that work against seeing or reading particular texts or specific themes in particular texts of the early modern period. The first essay in the section ("Devout Humanism") deals primarily with figures who have been obscured with an aura of sanctity, so that their texts have not, in any strong sense, really been read. This is especially true of St. François de Sales and the Introduction to the Devout Life (1609), but it has, until very recently, also been true of George Herbert's "The Church-porch." Both of these texts (with some by Donne) and the movement with which they have been associated ("devout humanism") are shown to be quite disturbing when read along a sociological rather than a diffused religio-moral axis. My claim is not to have discovered anything about these texts but simply to be reading them for their social attitudes, reading them literally and on the surface. What is interesting is not only what is there in these texts but why this obvious feature of their content has not been widely recognized. An appendix to this essay takes the argument in a thematically, though not methodologically, opposite direction. It asserts, in Herbert's case, the possibility of transcending the ideology of "devout humanism." This appendix continues the discussion of New Historicism, seeing it as a view that would deny the possibility of such transcendence. As a whole, therefore, Essay 5 argues that it is as inappropriate to deny as it is to assert the possibility of saintliness a priori.
Essays 6–8 form a unit. As their titles indicate, they argue for what I am calling "impossible radicalism" in the sixteenth and seventeenth centuries. They set themselves against a significant body of scholarly opinion—to which both old and new historicism have contributed—that sees conceptions like freedom of conscience, justified individual disobedience, and justified popular rebellion as "unthinkable" in the Renaissance or "early modern" period.[5] This notion of "unthinkability" seems
[5] These terms are interestingly at variance, and each has its advantages and disadvantages. "Renaissance" has the advantage of capturing the claim of many intellectuals in the (let's call it) period that something that had (supposedly) been lost was being consciously revived: good Latin; good letters; the art of perspective; the true church. "Early modern" has the advantage of neutrality with regard to privileging the arts and "high" culture, so that an account of "early modern" culture does not necessarily flatter or idealize the culture in question and does not restrict its scope or focus to cultural production in the Humanities sense of the term (invented by "the Renaissance"). The term "early modern" does, however, flatter and perhaps idealize us—the earlier period is seen to take its orientation from how it leads to "modern" culture. This too seems to be a historiographic myth, So one is left with a choice of myths—theirs or ours.
to me a very dangerous one. It is another sort of a priori. It makes it necessary for a critic or scholar to explain away—or simply not see—moments in texts where the "unthinkable" is actually thought. The view in which "unthinkability" prominently figures tends, I argue, to present periods or "discursive formations" as too homogeneous, dominant discourses as too successfully dominant (and too homogeneous in themselves), and to overemphasize breaks or ruptures between periods or discursive formations. I discuss this problem in Essay 6 (on Donne's third Satire), but the problem of acknowledging the existence and representation of genuinely radical or oppositional discourse dominates all three of the essays with which this book closes.
But what about the problem of anachronism and of "Whig" historiography? Obviously the avoidance of anachronism is important (and a historical achievement), but I would urge great circumspection in the application of the notion. Often conceptions are said to be "anachronistic" merely because they are nonhegemonic or unusual in a period. The cry of anachronism, I suggest, almost always serves the interests of a conservative picture of the past. A recent essay on King Lear , for example, devotes a great deal of useful and important scholarship to "proving" that when the mutilated and abandoned Gloucester prays that "distribution should undo excess" so that "each man have enough," Gloucester is merely saying something normal and completely familiar. To think anything else would be—and this is the normal tone and stance of such claims—"totally ahistorical."[6] Even in imagining extreme situations, Shakespeare cannot be seen as imagining radical solutions. Surely there was no such book in the sixteenth century as More's Utopia .[7]
As for "Whig" historiography, when it is false and distorting it is certainly a problem, but the revolt against the "Whig" historiography of seventeenth-century England has produced a view of the culture that makes the most dramatic historical event of the century virtually unin-
[6] Judy Kronenfeld, "'So Distribution Should Undo Excess, and Each Man Have Enough': Shakespeare's King Lear —Anabaptist Egalitarianism, Anglican Charity, Both, Neither?" ELH 59 (1992): 755–84. The quotation is from p. 757. This article is discussed in note 90 to Essay 8 below.
[7] For the "surely there [was] no such" formulation, see Essay 1, p. 25 below.
telligible. The culture somehow inadvertently or accidentally gave rise to (or allowed) the first modern revolution, an event in which a king was formally brought to trial and executed for offenses against "the people" and "the laws."[8] It may be misleading to see England in the early seventeenth century as a "prerevolutionary" society in any but a technical sense, but it certainly seems important to acknowledge the possibilities in that culture for thinking as well as enacting resistance. The texts that the last three essays treat are "resistant" in the political as well as the epistemological sense.
The ghosts of two past critics haunt these pages. The first is a spirit I have quite consciously conjured, and who stands as the guiding figure for both parts of this book, namely, William Empson. Empson's controversy with Rosemond Tuve is the subject of the first essay of the book—in which I argue that Empson is a better model for historical criticism than Tuve—and the essay on Donne and freedom of conscience (Essay 6) is explicit in its commitment to carrying on Empson's project of seeing Donne (and others in the period) as capable of genuinely radical thought. Empson is a model for me by virtue of his verbal and philosophical alertness, his nonprogrammatic curiosity and bafflement, and his complete lack of theoretically imposed inhibitions in approaching texts. The other figure who haunts these pages comes largely unbidden. At moments in this introduction and elsewhere, I found myself echoing R. S. Crane on the "high priori road" and on hypotheses in historical criticisms.[9] I discovered to my astonishment that, without ever having been trained as such, I was a member of the "Chicago school" in a stronger sense than merely teaching in the department in which Crane taught. I felt like the protagonist of one of those remarkable Islamic narratives in which choice is revealed to be predestined fate.[10] Despite Crane's lucidity and perspicacity, I am less happy with Crane
[8] For a discussion of "revisionist" (anti-"Whig") historiography of seventeenth-century England and the reaction against such "revisionism," see the "Introduction: After Revisionism" and the essays in Richard Cust and Ann Hughes, eds., Conflict in Early Stuart England: Studies in Religion and Politics, 1603–1642 (London: Longman, 1989). See also Derek Hirst, "Revisionism Revised: Early Stuart Paliamentary History—The Place of Principle," Past and Present 92 (1981): 79–99.
[9] R. S. Crane, "Criticism as Inquiry; or, The Perils of the 'High Priori Road,'" and "On Hypotheses in 'Historical Criticism': Apropos of Certain Contemporary Medievalists," in The Idea of the humanities and Other Essays (Chicago: University of Chicago Press, 1967), 2: 25–44, 236–60.
[10] See the discussion in Rudolf Otto, The Idea of the Holy , trans. John W. Harvey (London: Oxford University Press, 1950), 88–90.
than with Empson as a guiding spirit here, however, because Crane (like de Man) seems to me to have had a dogmatic as well as a critical side. I do not believe that nature and Aristotle are the same. And I share Empson's puzzlement at why, for instance, "because Macbeth is 'imitative,' it can't be 'didactic' as well."[11]
The question obviously arises as to how I can avoid the tu quoque . Do not I have my own presuppositions and schemes? In one sense, I cannot answer this. It is not for me to say. I can say that I have tried never to brush aside the obvious and the "surface." When I have said "Yes, but," I have tried to give weight to the affirmative as well as to the adversative. I can also say that I have tried to keep my own claims methodologically modest (other forms of modesty I do not aspire to). When arguing for the presence of a theme in a work, I have not argued that the theme in question is the only one in the work or even the "central" one. Similarly, in arguing for the existence of genuine radicalism in the early modern period, I have not argued that such radicalism is the spirit or the essential spirit or the true spirit of the age. I have only argued that the features to which I point are "there," in a quite strong sense, and that they matter, not that they are the only things that are or do. I have tried to make my orientations explicit, and though I do a good deal of polemicizing against various critical and historical schemes, I hope not to have put forth any dogmas.
But is "not having dogmas" itself a dogma? I cannot say that I find the claim that it is such interesting or powerful—or even fully intelligible. The attempt to present "pluralism as dogmatism" involves treating all founding principles as dogmas and, in its desire to present a "benign" picture of dogmatism, involves ignoring historical evidence that the function of the promulgation of dogma was always to repress heresy, to shut down or narrow rather than to expand or encourage public discussion.[12] It is certainly, however, worthwhile to be reminded that "pluralism" can be a cover for arrogance and for setting oneself "above the fray" (there is indeed an element of this in R. S. Crane). It is also useful to be reminded that, in the world of international politics, pluralism can be used as a cover for and justification of aggression and suppression,
[11] "Still the Strange Necessity," in William Empson, Argufying: Essays on Literature and Culture , ed. John Haffenden (London: Hogarth, 1988), 122.
[12] See W. J. T. Mitchell, "Pluralism as Dogmatism," Critical Inquiry 12 (1986): 494–502. For the "benign" picture, see p. 496; for an assertion of the point that dogmas were first formulated against heretics, see the paragraph from the Encyclopedia of Religion and Religions that Mitchell quotes (495).
though it should be noted that even the author of this attack on pluralism (W. J. T. Mitchell) distinguishes between "real tolerance" and "the use of tolerance as a code word for repression."[13] To this distinction, of course, I can only say "Amen." Mitchell's need for it implies that his presentation of "pluralism as dogmatism" is not truly in good faith. "Real tolerance" (to Mitchell's credit) escapes unscathed. As to location with regard to "the fray," I hope to have located myselfin rather than "above" it. Finally, I want to note that it is no longer as clear as it may have seemed in 1986 (the date of Mitchell's essay) that "pluralism is the reigning ideology of American politics on both the Right and Left." If this has changed, it may be especially important now to argue for pluralism—in the fray, and inside and outside the academic world.
[13] Mitchell, "Pluralism as Dogmatism," 502.