Stanley Fish—A Generation Behind on Textualism

• July 19, 2012 • 4:28 pm

In a recent article for the New York Times, Stanley Fish critiques the defense of textualism offered by Justice Scalia and Bryan Garner in their important new book, Reading Law. Despite his qualms about its central position, Fish has high praise for the book, which he calls entertaining and “wonderful.” Along the way, he also dispatches Living Constitutionalism as “a form of political gerrymandering rather than as a form of interpretation.”

In this post, I’d like to respond to Fish’s critique. In brief, Fish articulates two lines of attack: 1) Statutory meaning is not “‘in the text’ in the sense Scalia and Garner insist on,” and 2) All interpretation “begins and ends” with “the assumption or specification of an intention without which there would be no text . . . .” Both arguments draw on Fish’s influential body of scholarship dating back to the 1970s; both arguments show their age. The first argument has teeth only against an archaic version of statutory formalism that has been roundly rejected by modern textualists; the second has little relevance to the “second generation” of textualism that has emerged over the last decade.

Fish’s first argument rightly emphasizes the importance of context in the interpretation of texts. This line of attack harkens back to a snarky little debate he had with Ronald Dworkin in the 1980s over the “objectivity” of interpretation. The centerpiece of the argument is Fish’s discussion of the ambiguity of the word “draft.” As he notes, the sentence “let’s avoid the draft” is ambiguous between several meanings, such as “let’s get out of military service” or (spoken by a general manager of a professional sports team) “let’s trust in free agency.” Moreover, Fish insists that “the dream of adding enough words to a text so that its meaning is clear and indisputable cannot be realized,” noting that the expanded sentence “let’s avoid the draft and go to Canada” is, though more prolix, still susceptible to both meanings. Fish concludes that meaning just isn’t “in the text” in any deep sense.

Of course, Fish has rigged the example here. “Let’s avoid the draft and go to Canada” may not resolve the ambiguity, but how about “let’s avoid the military draft.” Indeed, it’s hard to state the argument without falsifying it — since to get the point through, Fish has to communicate to his readers both alternative meanings that the ambiguous sentence might take, and he has to communicate these meanings by means of a text! Nonetheless, Fish needn’t have cheapened his argument by relying on a jiggered example. Ambiguity (multiple meanings of the word “draft”) will often be resolved by a sentence’s syntax or the meaning of other words in the sentence, but not always — and occasionally, context will create ambiguity where none otherwise appears. Moreover, there are plenty of other ways in which resorting to context is indispensable to textual interpretation, and textualists are happy to concede the point. (None other than Justice Scalia once wrote that “In textual interpretation, context is everything.”)

Consider two examples: semantically underdetermined words, and implication. As an instance of the former, consider possessives, such as “Stanley’s”. The phrase “Stanley’s horse” could mean the horse Stanley owns, the horse Stanley is considering buying, the horse Stanley has wagered on, the horse standing next to Stanley, and so on. The semantic content of “Stanley’s horse” merely establishes a range of possible meanings; we need context to supplement or “enrich” the phrase’s semantics and tell us which meaning is the right one.

A second example of the importance of context is the phenomenon of implication. Imagine that my daughter asks me if she may have a glass of milk and a cookie and I respond, “You can have a glass of milk.” A fair implication of my response is that my daughter had better not help herself to the cookie; but this implication—a very important part of what I’ve communicated—is not determined by the meaning of the words I uttered. It’s determined only by consideration of the context, including knowledge of my daughter’s question, the authority structure of a parent-child relationship, and the strategies speakers use to communicate information as efficiently as possible.

Modern textualists are happy to accept the importance of context: Scalia and Garner note that their method of interpretation readily includes the “fair implication” of a statutory text, and Scalia’s famous dissent in Smith v. United States is a pretty good example of an attempt to grapple with the semantically underdetermined word “use.” Indeed, textualists concede that a statute’s context includes some evidence of “purpose.”

To be sure, Scalia and Garner in some sense invited Fish’s line of attack by attempting to cabin the permissible evidence of purpose to that “derived exclusively from [the] text.” Scalia and Garner are to be lauded for attempting to delimit the permissible scope of the textualist inquiry into purpose (this is perhaps the most challenging issue in textualist theory), but there is an obvious circularity in their insistence that evidence of purpose—which is to form part of the context which gives meaning to the text—be derived exclusively from inside the text.  And Fish is quick to exploit this weakness. But elsewhere in their book, Scalia and Garner emphasize that “[o]f course, words are given meaning by their context” and that “[a]ny meaning derived from signs involves interpretation, even if the interpreter finds the task straightforward.” Fish’s criticism of the attempt to find meaning “in the text” has bite against modern textualism’s estranged “plain meaning” ancestors from the 19th century. Yet against the modern version of textualism defended by Scalia and Garner, it cuts little ice.

Fish’s second line of attack centers on his insistence that all interpretation necessarily relies on the “construction of an intention” behind the text. This argument (best articulated in a famous article by literary theorists Steven Knapp and Walter Benn Michaels, and imported into legal theory by scholars including Larry Alexander and Sai Prakash), runs something like this: suppose you’re walking along a beach and observe the tide recede, leaving random patterns of markings in the sand that coincidentally resemble a word or phrase. Can we really conceive of “interpreting” these random marks, knowing that they weren’t produced by an intentional agent? It seems not. While we can imagine a social practice of assigning meaning to randomly produced marks, it bears little similarity to our concept of “interpretation.” But it’s important to be careful about what this argument does and doesn’t prove.  It certainly seems to prove that in order to conceive of something as a text that is susceptible to interpretation, we have to assume that the text was intentionally produced. But the argument doesn’t tell us much at all about how to interpret the text (indeed, Fish himself has reached a similar conclusion, albeit from a different, less plausible direction). Our concept of interpretation has limits, but those limits are fairly expansive.  How we should go about interpreting a text within those limits depends on a rich array of factors such as the nature of the author, the text’s genre, and the reasons we have for interpreting it in the first place.

Modern textualist theory has developed largely along these lines. Although early textualists relied on Arrovian social choice theory to undermine the plausibility of intentionalism, more recent textualist work has generally conceded that statutes are produced by intentional agents.  “Second generation textualism” has moved past intent-skeptical arguments in favor of arguments based on constitutional structure, the nature of legislative compromise, and the reasons we have for regarding statutes as authoritative. Again, on this issue Scalia and Garner in some sense lead with their chins. In good lawyerly fashion, they are reluctant to abandon arguments that have proven serviceable. Accordingly, their defense of textualism includes intent-skeptical arguments, such as the implausibility of “group intent.” Moreover, they scold interpreters for even using the word “intent,” fearing that it “invites fuzzy-mindedness.” But these intent-skeptical arguments constitute merely one strand of Scalia and Garner’s work, which also relies on justifications for textualism that are far more plausible. These arguments go without any response from Fish.

In a way, then, Fish can only be faulted for going after the low-hanging fruit. Scalia and Garner articulate a broad variety of arguments for textualism, some of which modern textualist theory has moved away from for reasons like those Fish articulates. But Reading Law also includes arguments—sounding in constitutional structure and the nature of legislative compromise—that have far more currency among modern scholars of statutory interpretation. By ignoring these arguments in favor of scoring easy points by attacking a version of textualism that textualists have largely abandoned, Fish does little to advance the debate.


  • allouchsit says:

    “Fish insists that “the dream of adding enough words to a text so that its meaning is clear and indisputable cannot be realized . . ..”

    This is true only because lawyers are so ingenious at misunderstanding the written word. Witness Justice Roberts’s opinion that a “penalty” is not a tax, except when it is.