The Fourth Dimension of a Poem

by M H Abrams

  6. Roland Barthes, “An Introduction to the Structural Analysis of Narrative,” New Literary History 6 (1975): 271; and “The Death of the Author,” in Image/Music/Text, p. 146. (See also Barthes’ essay “From Work to Text,” in Textual Strategies: Perspectives in Post-Structuralist Theory, edited by Josué V. Harari [Ithaca, N.Y., 1979], pp. 73–81.) Jacques Derrida, Positions, p. 66; “Signature, Event, Context,” in Limited Inc., pp. 3, 19, 20.

  7. Jonathan Culler, Roland Barthes (New York, 1983), pp. 81–82.

  8. Barbara Johnson, The Critical Difference (Baltimore and London, 1980), pp. 143–44.

  9. John Searle, “Reiterating the Differences: A Reply to Derrida,” Glyph 1 (1977): 201–2.

  10. Derrida, “The Supplement of Origin,” in Speech and Phenomena, translated by David B. Allison (Evanston, Ill., 1973), pp. 93, 96. “Signature, Event, Context,” in Limited Inc., pp. 3, 8, 11, 20.

  11. Derrida, Of Grammatology, pp. 158–59; see also p. 163. For his explanation that in these passages he intended to “recast the concept of text by generalizing it almost without meaning,” see his essay “But, beyond . . . ,” Critical Inquiry 13 (1986): 167–68.

  12. Derrida, “Afterword,” Limited Inc., p. 137; Of Grammatology, p. 159.

  13. Barbara Johnson, A World of Difference (Baltimore, 1987), p. 45.

  14. Derrida, “Letter to a Japanese Friend,” in Derrida and “Différance,” edited by David Wood and Robert Bernasconi (Evanston, Ill., 1988), pp. 3–4. See also Hillis Miller: “Deconstruction is not a dismantling of the text but a demonstration that the text has already dismantled itself” (“Stevens’ Rock and Criticism as Cure,” The Georgia Review 30 [1976]: 335–36).

  15. Roland Barthes, “Texte, théorie du,” in the Encyclopaedia universalis as cited by Culler, Roland Barthes, p. 118; “The Death of the Author,” p. 143.

  16. Michel Foucault, Power/Knowledge, edited by Colin Gordon (New York, 1980), pp. 98–119.

  17. Stephen Greenblatt, “Introduction,” Representing the English Renaissance (Berkeley, Calif., 1988), p. viii.

  18. Foucault, “The Discourse on Language,” in The Archeology of Knowledge (New York, 1972), p. 229, and The Foucault Reader, edited by Paul Rabinow (New York, 1984), pp. 56–57.

  19. Derrida, Of Grammatology, p. 112; and “Aphorism Countertime,” in Acts of Literature, edited by Derek Attridge (New York and London, 1992), pp. 416–33. Paul de Man, Allegories of Reading, p. 296.

  20. Derrida, Of Grammatology, pp. 157–58. J. Hillis Miller, Fiction and Repetition: Seven English Novels (Cambridge, Mass., 1982), p. 53.

  21. de Man, Allegories of Reading, pp. ix, 131. Johnson, The Critical Difference, p. 5.

  22. Barthes, “The Death of the Author,” pp. 142–48. Michel Foucault, “What Is an Author?” (1969), in The Foucault Reader, pp. 101–20.

  23. Foucault, “What Is an Author?” pp. 107–11, 118–19.

  24. Barthes, “The Death of the Author,” pp. 142–43, 146–47.

  25. Barthes, “The Death of the Author,” pp. 142–43. Foucault, “What Is an Author?,” pp. 101, 108–9, 119.

  26. Foucault, “What Is an Author?” p. 108.

  27. In his Epistle I.xx.1–2, Horace whimsically charges his book with an unseemly eagerness to hurry to the business district of the booksellers, in order to be exposed to sale.

  28. See Frederick G. Kenyon, Books and Readers in Ancient Greece and Rome, 2nd ed. (Oxford, 1950), especially pp. 81–84.

  29. For a discussion of the shifting emphasis, within the historical frame of critical discussions, on the world, the audience, and the work itself, see M. H. Abrams, The Mirror and the Lamp (New York, 1953), chapter 1, “Orientation of Critical Theories.”

  30. René Descartes, Meditations on the First Philosophy, Meditations 1 and 2.

  31. David Hume, A Treatise of Human Nature, edited by L. A. Selby-Bigge (Oxford, 1928), pp. 264, 269–71.

  32. Ludwig Wittgenstein, Philosophical Investigations, translated by G. E. M. Anscombe (Oxford, 1953), nos. 217, 325–26; and p. 226. On Certainty, edited by G. E. M. Anscombe and G. H. von Wright (New York and Evanston, Ill., 1969), pp. 162, 204, 358.

  33. Wittgenstein, On Certainty, pp. 105, 151, 162.

  34. John Keats, The Fall of Hyperion, canto 1, line 304.

  35. For example, David Hume, when writing as a literary critic and not a philosophical skeptic, takes a stance within the humanistic frame of reference. See his essay “Of Tragedy” (1757), which discusses tragedy as “an imitation of human actions,” and asserts that “the whole art of the poet is employed in rousing and supporting the compassion and indignation, the anxiety and resentment, of his audience.”

  36. Derrida, “Force and Signification,” in Writing and Difference, translated by Alan Bass (Chicago, 1978), pp. 3–30.

  37. Derrida, “White Mythology: Metaphor in the Text of Philosophy,” New Literary History 6 (1974): 5–74.

  38. For example, “White Mythology,” p. 13: “The task is . . . to dismantle the metaphysical and rhetorical structures which are at work . . . not in order to reject or discard them, but to reconstitute them in another way.” And “Afterword,” Limited Inc., p. 146: “The value of truth (and of all those values associated with it) is never contested or destroyed in my writings, but only reinscribed in more powerful, larger, more stratified contexts.”

  39. Johnson, A World of Difference, pp. xvii–xviii.

  40. Richard Rorty, “Is Derrida a Transcendental Philosopher?” Yale Journal of Criticism 2 (1989): 207, 215. See also Rorty, “Philosophy as a Kind of Writing: An Essay on Derrida,” in Consequences of Pragmatism (Minneapolis, 1982).

  41. Derrida, Limited Inc., pp. 61–62; “Afterword,” p. 129.

  42. Wittgenstein, On Certainty, p. 92. On some notable philosophical conversion experiences to a radically different initiating position, see M. H. Abrams, “Coleridge and the Romantic Vision of the World,” The Correspondent Breeze (New York and London, 1984), pp. 199–206.

  43. Goethe, Dichtung und Wahrheit, Part III, Book xi.

  44. James Boswell, The Life of Samuel Johnson, entry for 1763.

  45. Experimental and postmodernist types of “metafiction” that undertake to dispense with recognizable characters, setting, and sequential narrative are second-order literary productions, in that they achieve their specific effects by presupposing, in order to frustrate, readers’ expectations based on traditional modes of fiction.

  46. Byron, letter to Douglas Kinnaird, October 26, 1819.

  47. Barthes, “An Introduction to the Structural Analysis of Narrative,” New Literary History 6 (1975): 271. Clara Claiborne Park, Rejoining the Common Reader (Evanston, Ill., 1991), p. 227.

  48. Derrida, “This Strange Institution Called Literature” (1989), in Acts of Literature, edited by Derek Attridge (New York and London, 1992), pp. 39–40. Later in the interview Derrida said, “One of the main reasons for my interest in literature” is that it “teaches us more, and even the ‘essential,’ about writing in general” (pp. 71–72).

  49. de Man, “Autobiography as Defacement,” in The Rhetoric of Romanticism, p. 81.

  50. de Man, Allegories of Reading, p. 199. In a related discussion of the representation of guilt by Proust, de Man asserts that “no one can decide whether Proust invented metaphors because he felt guilty or whether he had to declare himself guilty in order to find a use for metaphor”; he adds, “The second hypothesis is in fact less unlikely than the first.”

  51. J. H. Miller, “Stevens’ Rock and Criticism as Cure, II,” Georgia Review 30 (1976): 341.

  52. de Man, Blindness and Insight, p. 111. Also de Man’s Allegories of Reading, p. 277: An author, “just as any other reader . . . is bound to misread his text. . . . Language itself dissociates the cognition from the act.”

  53. Derrida, “After
word,” Limited Inc., p. 157. Derrida is responding to Habermas’ attack in The Philosophical Discourse of Modernity (1987).

  54. Derrida, “Afterword,” Limited Inc., pp. 143–44, 146, 151.

  55. J. H. Miller, “The Triumph of Theory,” PMLA 102 (1987): 284.

  56. Hume, A Treatise of Human Nature, pp. 269–70.

  57. Derrida, Limited Inc., pp. 21, 152.

  58. Derrida, “Structure, Sign, and Play,” in The Languages of Criticism, pp. 264–65.

  The Language and

  Methods of Humanism*

  I FIND ENLIGHTENING AND useful Frederick Olafson’s sketch of the conceptual scheme—the frame of often implicit assumptions—that has been a distinctive feature within the various disciplines traditionally grouped as “the humanities.” This scheme establishes coordinates that enable us, in a preliminary way, to map out what is humanistic, what is nonhumanistic, and what is antihumanistic; it also enables us to detect recent tendencies within the humanities themselves that threaten to subvert, or else to abandon silently, the very premises on which the Western humanist, through the centuries, has undertaken to understand and assess man, his actions, his history, and his intellectual and imaginative productions.

  Olafson finds that the traditional humanistic concern is with the world of the distinctive person, for whom nonhuman nature is the theater for his activities, and who confronts in the world other persons similarly endowed and engaged. These persons are thinking and feeling agents who manifest intentions and purposes and have some measure of control over, and thus responsibility for, their own destinies. If there is purpose, there is also a choice between alternative actions, and the choice and its consequences can be judged by criteria of better or worse, right or wrong, good or evil. Human life and history are viewed as a narrative sequence, or drama, in which there are conflicts within a person, and between persons, and between a person and the conditions of his milieu. This drama, as it evolves, displays love and hate, mutual achievements and mutual destruction, individual successes and failures, comedy and tragedy, the sublime and the ridiculous. In his intellectual and imaginative products, man expresses his human concern—his “form of life,” his informing vision, his assumptions and structure of values—and the humanist scholar who undertakes to understand and place and assess these products views them, ultimately, through his own perspective of concern.

  What I want to do is bring these traditional humanistic concepts to bear on two topics that have emerged at this conference. The first topic is the risk of skepticism and relativism to the modern humanistic enterprises; the second is the question of whether effective teaching in a liberal education is primarily a teaching of “arts” and methods in the humanistic disciplines, or whether it is primarily a teaching by example, by instancing in the person of the individual humanist a stance and procedure that is representatively humanistic.

  I

  The humanist typically addresses himself to texts that are not written in the highly refined and specialized languages of the logician or the scientist, but in the ordinary language that has been developed over many centuries to express and to deal with the complexities, the ambiguities, the nuances, and the contradictions of the human predicament—the predicament of purposive, fallible, perplexed, and feeling persons, who, for better or worse, act and interact and manifest what Keats called “the fierce dispute / Between damnation and impassioned clay.” Also, the traditional language of humanistic critics and scholars, despite some technical elements specific to a particular discipline, remains the ordinary language of the persons and documents with which they deal—and it has to do so if humanists are to carry out their traditional functions.

  From this fact follows a conclusion that some of us find hard to accept; namely, that in many of the central and most distinctive judgments of the humanistic scholar and critic we can never achieve certainty. In fact, one way to identify the humanities is to say that they are those disciplines whose concern is with the areas of human action and production where valid knowledge is the aim, where a rational procedure is essential, but where certainty is impossible.

  I agree with Gertrude Himmelfarb’s concern about the dangers to the humanities of threats from within—the threats of radical skepticism and relativism, and of giving up the old search for truth. I recognize—with equal dismay—the same tendencies in literary studies that she finds in historical studies: a surrender to irrationality, a stress on multiple “interps” instead of meaning, the drift to a profound skepticism, even to nihilism, with respect to values. It would be a bad mistake, however, to combat such tendencies by the counterclaim that humanistic studies, in all their central enterprises, can yield certainty and a single and ultimate truth. Some parts—some basic parts—of these studies are indeed factual, and therefore subject to the criteria that govern the sciences: the criteria of valid empirical reasoning and the established ways to support or falsify hypotheses. But when you get down to matters of explanation, interpretation, and evaluation—whether of Hamlet or of Aristotle’s Poetics or of the French Revolution or of the nature of justice—then you are out of the realm where certainty is possible and where universal, or even very widespread, agreement is to be found.

  Even when I, as a literary critic, feel quite certain that I am asserting the truth about a complex work, I have long since given up being surprised to find that other commentators, who are indubitably expert, sensitive, and rational critics, sharply disagree. If we should be so misguided as to claim to our students that in this realm our conclusions are certain and at least approximate the single and universal truth of the matter, then we shall be quickly found out and discredited, with the risk of discrediting as well the whole humanistic enterprise in which we are engaged.

  But I do not believe that the denial of certainty entails radical skepticism and relativism. To say that the humanist operates in a realm where, in large part, certainty and the single truth is impossible is not to say that it is a realm where uncertainty reigns, where no truths are achievable and therefore anything goes. What produces confusion in the use of the terms “certainty” and “truth” is that, in professional discourse about rational procedures, they are closely tied to highly specialized verbal and symbolic models. In the distant past, when divine authority yielded the ultimate certainty, there was also a symbolic enterprise—formal logic—that was the model of the certainty that could be achieved by human endeavor. When what we think of as modern science developed, apologists had a difficult time justifying the validity of their new systematic procedures, and they tried to mitigate the differences from the deductive logical model by bridge-concepts such as the law of sufficient reason. But after the exact sciences had triumphantly established their own validity and authority, the codified rules of scientific language and scientific reasoning achieved a status equivalent to that of logic in guaranteeing certainty, in the sense that all qualified practitioners would consent to its conclusions.

  The criterion of certainty, as applied to the humanities, is usually tied to one or both of these alien and highly specialized models for achieving formal certainty or practical certainty, deductive truths or empirical truths. The language of the traditional humanist, however, is very different from the specially developed, sharply defined, and strictly rule-bound ideal languages of logic or the natural sciences. For the humanist’s language, although responsible to the formal rules of logic and in its own fashion empirical, must perform its central functions beyond the point where these simplified calculi of logic and the exact sciences come to a stop. To achieve its traditional aims, the language of the humanist is necessarily flexible, loose, uncodified, nuanced, and lacking sharp definitional boundaries. Ultimately, it is also what Olafson calls the language of concern—that is, it engages with its subject matter in an area in which it is subject only to such soft-focus criteria as good sense, tact, insight, aesthetic sensibility, and a sound sense of moral values. Such discourse is
rarely capable of rigid codification, and therefore rarely capable of achieving strictly conclusive arguments; but it can and should be responsible and rational—with the kind of rationality that alone is adapted to fulfill its own humane purposes and to achieve sound knowledge in its elected area of understanding. And if (when judged by the alien criteria of simplified calculi) it is not certain, neither is it, strictly speaking, uncertain. The reasoning and conclusions of humanistic discourse are subject to criteria, but these are criteria appropriate to its own intellectual enterprise, such as coherent or incoherent, inclusive or omissive, sensible or outré, disinterested or partisan, central or overly ingenious, clear or obfuscative, sound or unsound.

  “The kind of certainty,” Wittgenstein has said, “is the kind of language-game.”1 Now we can, if we choose, apply the word “certain” to a humanistic conclusion that satisfies the positive criteria I have listed. But if we are to avoid confusion and error, we must keep in mind that this is often a very different kind of certainty from the formal certainty of logic or the practical certainty of an exact science. An even more serious error is to try to make the humanities capable of the scientific kind of certainty by “objectifying” (to use Olafson’s term) humanistic inquiry; that is, by substituting for its ordinary, loose, and flexible language of interpretation, evaluation, and concern the specialized and codified language of science and, above all, by translating the language of purpose and responsibility into a calculus of causality. Either this translation is no more than a merely lexical substitution, which results in scientism—a simulacrum of scientific procedure and conclusions—or else it is inherently incapable of accomplishing the central humanistic functions. For the language of science has achieved its precision, the codified rigor of its reasoning, and its kind of certainty by systematically eliminating from consideration all those aspects of human experience and judgment that, to the humanist and to all of us with our human interests and concerns, matter most. And when the humanist genuinely commits himself to a language modeled on that of the sciences, he finds that the specifically human aspects of his subject matter, such as individual personality, purpose, passion, drama, and value, ineluctably elude his linguistic grasp.