On King v. Burwell: What Is a Natural Reading?

06/29/2015 03:56 pm ET | Updated Jun 29, 2016

In King v. Burwell, decided last Thursday, the Supreme Court has once again (no doubt inadvertently) given us a lesson in the philosophy of language. Let me explain by way of an example.

Some years ago as I was driving my father back to his apartment in Florida we approached an intersection where the light had turned red. He said, "Go through the light." What did he mean? He could have meant, "Don't stop even if the light is red"; or he could have meant, "When you get to the intersection go straight, don't turn either left or right." Which was it, and how could I figure it out?

Well, it would depend on what I assumed his intention was. Did he doubt that his son, a college professor and therefore someone ill-equipped to negotiate the everyday world, was capable of remembering the way to the apartment, although he had been there many times? Or had he suddenly recalled an appointment he was late for, saw there was no one around, and decided that a little minor law-breaking would do no harm and save him a minute or two? Or was he experiencing some physical discomfort and needed to get home as soon as possible?

In order to settle on any of these meanings (and there are more that could have been listed), I might have had to ask him some questions. Or perhaps I wouldn't have had to ask questions because I knew him so well that I could pretty much be certain what he might be saying in a situation like this one. In either case -- if I were relying on my deep knowledge of him or if I were puzzled by his imperative and sought clarity -- what I would be after is his intention, a sense of what he had in mind, and it is in the light of that intention that the meaning of his words would stabilize. And if I neither asked questions nor were confident in my understanding of his character -- of why he, being who he was, might say such a thing -- his utterance would continue to be unstable with respect to its meaning; it would continue to be a series of recognizable, even ordinary, English monosyllables whose significance was multiple and unclear.

The moral (or at least the one I draw) is that words, no matter how simple or carefully crafted, do not declare their own meaning, and they are readable -- that is, construed as having a determinate sense -- only when the intention informing them is either presupposed or established by inquiring into it.

The additional moral is that the traditional distinction between what is said and what is meant doesn't hold up. That distinction assumes that there is a level of language, sometimes called literal, that is self-sufficient in its capacity to convey a message and doesn't require any external information about the speaker's intention (or about anything else) to piece it out. In this view of the matter, intention is either built into the language or is something in a speaker's head that might or might not match up with what is spoken. (Thus the possibility of the disjunction between what was said and what was meant.) But what my example shows is that there is no such literal level -- there is no literal meaning of "go through the light," only the various meanings that seem obvious and inescapable (and, yes, literal) once an intention is specified -- and that without either the assumption or the positing of intention, a string of words doesn't have a meaning at all, but is, rather, a collection of semantic possibilities waiting for something to narrow it down. While in that condition -- existing in an abstract space and tied to nothing -- words are irremediably ambiguous -- all of them.

I put it that way because of something Justice Antonin Scalia says in his dissent. The dispute in the case is over the meaning of the phrase "exchange established by the state." Chief Justice John Roberts, writing for the majority, argues that the phrase can and should be read to include an exchange established by the federal government. He explains that "exchange established by the state" is ambiguous because when read in context (as he proceeds to do) it means something different than it does when read in isolation. Scalia retorts that by the logic of such a reading, "everything is ambiguous."

That's both right and not right: everything is ambiguous, if by everything is meant words considered apart from any intention informing them; but if an intention is either taken for granted or put in place, the potential ambiguities -- produced when language is imagined as being in an unanchored state -- will have pretty much disappeared. Once I have determined to my own satisfaction what purpose prompted my father to say "go through the light," his words are no longer ambiguous, but clear as a bell. Of course I might turn out to be wrong, in which case my initial determination would be succeeded by another, and his words would still be clear as a bell, but the clarity would be different.

So it is just as cogent to say "nothing is ambiguous" (unless of course an ambiguity is intended, in which case the ambiguity is straightforward) as it is to say that "everything is ambiguous." What you can't say is that either ambiguity or clarity follows from the language of an utterance considered in and of itself; they both follow from the fact of intention or its absence. Words always mean what their speakers intend, and they fail or multiply in their meaning if no intention is attached to them. (There is no such thing as either clear or ambiguous language, although there are certainly clear or ambiguous utterances.)

Chief Justice Roberts is no less confused about this than Scalia. He thinks that in order to make his case about the interpretation of "exchange established by the state" he has to depart from what both he and Scalia call the "natural" reading or sense of the phrase: "The context and structure of the Act compel us to depart from what would otherwise be the most natural reading of the pertinent statutory phrase." By "natural reading" or "natural sense," Roberts could mean one of two things:

(1): a sense that belongs to the words by right; the words have an essential, not a merely conventional, relationship to the things they name; the words capture the nature of the things;

(2) a sense that corresponds to standard usage at the time of the text's production; the statistically, not naturally, normative sense.

Nobody (with the possible exception of George Orwell in his very bad and silly essay "Politics and the English Language") believes in (1), for that would be to believe that all language was, at least ideally, onomatopoeic -- every word would work as "bow-wow" and "meow" do.

Therefore (2) is obviously what the two justices subscribe to. But that amounts to saying no more than that, all things being equal, words and phrases should be interpreted in accord with the meaning most usually associated with them; that is, in accord with the intention most speakers have in mind when they utter them.

But when all things are not equal, when there is good reason to think that another, not statistically predominant, intentional context is presiding over the scene of communication, you affirm the reading that context generates, which then becomes the "natural" reading. The natural reading is not a reading arrived at apart from context, but the reading that seems most natural to you given the context you are reading or hearing within. Roberts is not departing from anything; he is just going with the reading that now seems inescapable to him in light of his determination of what the authors of the affordable care act had in mind -- of what they were up to.

He says as much, quoting an earlier case: "...the words of a statute must be read in their context and with a view to their place in the overall regulatory scheme." And, again: "We cannot interpret federal statutes to negate their own stated purposes." In this case, the stated purpose, he tells us, was to "improve health insurance markets and not to destroy them," and therefore "we must interpret the Act in a way that is consistent with the former and avoids the latter."

This is not to subvert the natural reading, but to insist on it. The unnatural reading is the one that regards the phrase "exchange established by the state" as if it floated in space untethered to the constraint of any actual communicative project and bore a timeless meaning that was somehow its property. The unnatural reading is Scalia's. As Justice Elena Kagan said in the oral arguments, "We don't look at four [or five] words. We look at the whole, the particular context, the more general context, try to make everything harmonious with everything else." And when we do this, she might have added, we are not engaging in a special or aberrant mode of interpretation; we're just doing interpretation in the way that it is always, and necessarily, done.

Roberts does not know how strong his position is, and he is kept from that knowledge by his signing on to the existence of a "natural" reading in the stronger (and impossible) sense of a reading the words naturally have. This mistake opens the door for Scalia's chest-thumping and table-pounding fulminations and for his accusation that the majority is rewriting, not interpreting, the Act: "'Exchange established by the state' means what it looks like it means." "Words no longer have meaning if an Exchange that is not established by the state is 'established by the state.'"

Had Roberts deconstructed (God forbid!) the idea of a naturally "natural" reading, he would not have been so vulnerable to Scalia's scathing rhetoric on this point (although he would have then been vulnerable on other grounds).

But perhaps this is too harsh a judgment. As the justices themselves often say, they are lawyers, not philosophers, and they are bound not by theses in the philosophy of language but by the practices and vocabulary of their profession. Splitting utterances into two kinds -- the ambiguous and the unambiguous -- is, I think, all wrong, but the accepted canons of legal interpretation require it (this is the famous Chevron rule), and a justice must fall in with that requirement even if it commits him to positing entities (like an objectively natural reading or a-contextual meanings) that do not exist, and commits him also to the task of finding his way out of the dilemmas that believing in those entities creates.

But, after all, Roberts manages to do it.


Stanley Fish is the Davidson-Kahn Distinguished University Professor and Professor of Law at Florida International University, and the Floersheimer Distinguished Visiting Professor of Law at Cardozo Law School.