1. See Steven Knapp and Walter Benn Michaels, "Against Theory," Critical Inquiry 8 (1982): 723-42; "A Reply to Our Critics," Critical Inquiry 9 (1983): 790-800; "A Reply to Richard Rorty: What Is Pragmatism?" Critical Inquiry 11 (1985): 466-73; "Against Theory 2: Hermeneutics and Deconstruction," Critical Inquiry 14 (1987): 49-68. The first three items are reprinted in Against Theory: Literary Studies and the New Pragmatism, ed. W. J. T Mitchell (Chicago: University of Chicago Press, 1985), 11-30, 95-105, and 139-46, respectively.

2. Hans-Georg Gadamer, Truth and Method, trans. and ed. Garrett Barden and John Cumming (New York: Continuum, 1975), 291; hereafter cited in text as TM.

3. Sanford Levinson, "Law as Literature," in Interpreting Law and Literature: A Hermeneutic Reader, ed. Sanford Levinson and Steven Mailloux (Evanston, Ill.: Northwestern University Press, 1988), 158-59.

4. Paul Ricoeur, Interpretation Theory: Discourse and the Surplus of Meaning (Fort Worth: Texas Christian University Press, 1976), 76.

5. Ronald Dworkin, Law's Empire (Cambridge: Harvard University Press, 1986), 53; hereafter cited in text as LE. Dworkin's account of interpretation has much in common with Gadamer's, as Dworkin himself notes; see 55, 62, 42On.

6. David Hoy, "Intentions and the Law: Defending Hermeneutics," in this volume; all further references to Hoy will be to this essay.

7. Or if that seems too extreme, perhaps Hoy would want to say that no set of marks produced by chance can be a text until it is mistaken for a text.

8. At one point Hoy offers the case of ambiguity as a counterexample to our account of meaning: "An ambiguous sentence," he writes, "could be described as one having two different 'meanings,' whether both of them were intended or not." Adopting for hermeneutics the distinction, familiar in speech-act theory, between sentence meaning and speaker's meaning, he asserts, "Explaining ambiguity requires making a distinction between the two sentence meanings before raising the question about which of these was intended, or whether both were." But sentence meaning—the meaning assigned to a set of marks by the semantic and syntactic rules of some particular language—is not an alternative to intended meaning; it is just the meaning that would be produced by someone who intended to produce a meaningful text or utterance by following the rules of the language in question. Hence the choice between two sentence meanings (or between a sentence meaning and some other meaning) is simply a choice between two possible intended meanings.

9. We present this example in hypothetical terms not because we are unaware of or indifferent to the complex history of the Fourteenth Amendment and of its interpretation but because the details of that history are beside the point of the present discussion.

10. Ronald Dworkin, "The Bork Nomination," New York Review of Books, 13 Aug. 1987, 8.

11. Ibid., 6.

12. In the same passage Hoy raises the issue of "ulterior purposes" and urges us to say something about "the difference between linguistic intentions ... and nonlinguistic intentions." It is sometimes proposed as an objection to intentionalism that legislators can, for ulterior purposes such as getting reelected, enact a law in which they do not themselves believe. But such purposes are not part of the meaning of the law; they are motives for enacting a law with that meaning. The ordinary case of lying is an obvious parallel; the fact that one can produce an utterance with an intention to deceive does not alter the meaning of the utterance itself. There are simply two intentions, one to produce an utterance with a certain meaning and one to use that utterance to deceive. Hoy himself appears to be confused about this distinction when he conflates the intention to mean something with the intention to make that meaning "valid or sound." The intention to say something true is no more a part of the meaning of what one says than is the intention to say something false.

13. Dworkin cites H. Jefferson Powell, "The Original Understanding of Original Intent," Harvard Law Review 98 (1985): 885-948, as providing "persuasive historical evidence that the framers intended that their own interpretations of the abstract language they wrote should not be regarded as decisive in court" ("Bork Nomination," 3). For Dworkin this evidence suggests that intentionalism is "self-defeating," but if all Powell shows is that the framers thought the public or "plain" meanings of their words were better evidence of their intentions than were the records of their private debates, then what he has provided is not an argument against intentionalism but an account of where the framers thought the best evidence of their intentions could be found. If, however, the claim is that the framers meant that their intentions themselves should be disregarded, then it's hard to see how the framers could be understood to have intended that interpretations of the Constitution be interpretations of the text they wrote, since the only thing the Constitution and the text they wrote would have in common would be the marks on the page.

14. Hoy is consequently right to say that in a text the meaning of a sentence "is constrained by its inherence in a context formed by the other sentences." But this is true only to the extent that those other sentences count as evidence of what the author intended in the sentence in question.

15. Recovering authorial intentions may, however, be difficult. And the general difficulty of recovering the intentions of authors who wrote, for example, in the distant past seems to many to be exacerbated in the case of a document such as the Constitution, which is the product of collective authorship. For if meaning is determined by intention, how does one interpret a text whose various authors may have had different intentions? Here the difficulty goes beyond the problem of figuring out what the authors intended; indeed, it emerges only when, having figured out what the authors intended, one discovers that by the same marks they intended different things. Of course, it frequently happens that two authors produce the same marks with different intentions, but there is no difficulty here; the two identical sets of marks simply constitute two different texts. The difficulty arises when two or more authors intend to produce a single text with a single agreed-upon meaning but fail to do so because, as it turns out, they each meant something different by the marks they collectively produced. It would be a mistake, in our view, to imagine that they have thus produced a single text with more than one intention since, as we have shown, the identity of a text depends on an identity of intention (identity of marks is not enough). Instead, they are in the position of two different authors meaning different things by the same set of marks except that they intended to produce a single text but failed. Their intention to produce a single text is not, however, part of the meaning of any of the texts they have in fact produced, nor does its failure in any way stop the meanings of those texts from being determined by their intentions. A judge who discovered that the framers of the Fourteenth Amendment meant different things by "equal" would not be in a position of deciding whether to disregard authorial intention (no interpreter can ever do that while continuing to interpret). But she would have to decide which of the texts produced by the framers counted as the Fourteenth Amendment.