I’ve got a long-running debate with colleague and landlord Gene Healy about constitutional interpretation, which we replayed in a bit more detail over drinks the other night. The dispute arises over a tension that characterizes a lot of libertarian thinking about the law. On the one hand, we want to interpret “commerce” (as in “Congress shall have the Power… To regulate commerce… among the several states”). That is, “commerce” means what it would’ve meant to the ordinary citizen at the time of ratification: traffic in goods across state boundaries. It doesn’t mean any kind of manufacture or production for commercial purposes, and it sure as hell doesn’t mean carrying a gun near a school. On the other hand, we want to read phrases like “privileges and immunities” or “equal protecton” or “due process” quite broadly—to “drive a truck through them,” as Gene puts it. He thinks, probably correctly, that a lot of us are inclined to change our interpretive methodology to fit the conclusions we want to reach, conflating the quesiton of what the Constitution means with what we think it ought to mean.

Well, that still leaves the question of which is right. Gene tends to opt for a more directly historical standard: both “commerce” and “equal protection” mean what the public and legislators at the time of ratification would have understood them to mean. So commerce gets a narrow construction, but the fact that the 19th century gives us plenty of examples of laws with racial or gender discrimination built in precludes a reading of the Fourteenth Amendment that mandates legal color blindness or gender neutrality. Now, I feel about having legal arguments with Gene roughly the same way he says he does about debating philosophy with me, only substantially more so. I took a couple of undergraduate classes in Con law at NYU; he’s got a J.D. from the University of Chicago. Advantage Healy, to put it mildly. Nevertheless, I’m inclined to take a different tact, one that leaves the proper boundaries of both “equal protection” and “commerce” at least somewhat fuzzy and open to revision.

This is at least in part because those boundaries are already necessarily fuzzy. That is to say, is there any reason to think there must have been widespread agreement about how constitutional language should be interpreted in application even at the time of ratification? And even if there was, are we permitted to observe that the shared understanding might be irrational or inconsistent?

One way I’ve characterized our disagreement in conversation is by means of the contrast between meaning and extension. In philosophy of language (notice how I shift the debate to home territory!) it was popular for a while to regard them as the same—the slogan was “meaning is extension.” That is to say, what a term like “red” means is the extension of the term, which is to say, the concrete things to which it may be applied. The most famous counterexample is the terms “renate” and “cordate,” meaning “creature with a kidney” and “creature with a heart” respectively. It was at least believed for a long time that these terms had the same extension: creatures with kidneys always had hearts and vice versa. But the terms don’t mean the same thing at all. Now, Gene’s approach is (I don’t know if this is really a word) extensional: if we want to know how to interpret “cruel and unusual,” we look at the set of punishments that were considered “cruel” in 1789, and for novel cases (e.g. chemical castration), we try to figure out whether they’re more analogous to members of the set, or to other punishments not considered cruel. My approach would be to ask something like: what was the background principle that caused them to group these particular punishments together as “cruel?” That leaves open the possibility that the framers were themselves mistaken about the application of their own principle: things they would not have considered “cruel” may be, and the converse.

Part of my rationale here is that you can’t really avoid doing this anyway. If you’re going to have to reason by analogy to new cases, then you need some criterion for what’s going to be considered “the same” or “different,” which means a general background idea of what is relevant to sameness and difference. One recent Supreme Court case, for instance, concerned whether the use of infrared scanning technology to detect marijuana growrooms constituted a “search” within the meaning of the Fourth Amendment. Until the 20s, there was a bright-line property test for what was meant by search. If you crossed someone else’s property to go through their stuff, that was a search. If you didn’t, it wasn’t. So early in the century, wiretapping wasn’t a search. Later, it was, and the Court ditched the property test in favor of a “balancing test” that looked at whether someone had a “reasonable expectation of privacy” in a given context. It’s pretty obviously meaningless to ask what the Founders thought about the propriety of wiretap or infrared scanning, and reasoning by analogy to other searches is just a roundabout way of asking whether these new ways of searching have the general features of intrusion that they meant to prohibit.

Now, let’s turn to “equal protection.” It’s a fair bet that the average citizen in 1870 didn’t think that “equal protection” meant the law couldn’t discriminate against gay people, or minorities, or women. The extension of the phrase was narrow. But why? What, in other words, was the principle? If it was just that there were ad hoc, arbitrary exceptions to the general idea of treating citizens equally, based in an irrational prejudice, then should we read those into the text? Ronald Dworkin uses the following example in his writing on constitutional interpretation: someone charges you to deal on his behalf in selling some property. He tells you not to do anything “unfair,” but you know that he doesn’t regard withholding relevant information about the property as “unfair.” So do you follow the general injunction as your own understanding indicates, or his? Now, the problem Gene raises is that de facto, if we go with the former option, the stabilizing function of the law is undermined: nine unelected justices decide what the law will be. But I think that’s putting it strongly. We can read the Constitution hermeneutically, with the text and original public understanding drawing the general boundaries (“commerce” and “equal protection” don’t just mean anything) but allowing for revision to purge internal inconsistency and to accord with background liberal values implicit in the public political culture.

Gene is surely right to object that this weakens the rule of law to some extent. But I think to see why this is necessary, we have to consider the source of the Constitutionâ??s normative authority. The standard story, as found in Marbury v. Madison is that “the people have an original right to establish, for their future government, such principles as, in their opinion, shall most conduce to their own happiness.” But as Lysander Spooner famously argued, that’s a pretty difficult principle to defend. Why do those who wrote and ratified the Constitution have a right to bind those who didn’t? And why do any of them have a right to bind future generations? As a matter of consent theory, I doubt they do. So what do we have to fall back on? For one thing, the rule of law argument: the general desirability of a stable framework to govern our lives together. But equally important, I think, is that the Constitution codifies good liberal principles that ought to structure a society.

Now G.E. Moore coined the term “naturalistic fallacy” to describe the erroneous belief that a reductive, descriptive account of normative terms could be given. Meaning, in other words, is not extension—especially the meaning of normative terms. “Just” and “right” do not mean the same, even in my own mouth, as “the things I currently take to be just and right.” To the extent that the Constitution makes use of intrinsically moral terms—”equal protection” or “due process” or “cruelty”—those terms must be at least partly open. If the understanding of “equal protection” a hundred years ago contained exceptions motivated by nothing more elevated than arbitrary bigotry, then we should not feel obligated to replicate those exceptions or that narrow reading in our own interpretation. We should, in short, act as though the authors and ratifiers of Constitutional provisions are present in current debates over interpretations. If we cannot at least imagine them giving some kind of rational argument for treating “equal protection” as not extending to certain minority groups, then why should we incorporate their prejudices into our own construction of those clauses?