«THE TANNER LECTURES ON HUMAN VALUES Delivered at Princeton University March 8 and 9, 1995 ANTONIN SCALIA is Associate Justice of the United States ...»
and what he had offered to give in exchange for the cocaine was an unloaded firearm, which he showed to the drug-seller. T h e Court held, I regret to say, that the defendant was subject to the increased penalty, because he had “used a firearm during and in relation to a drug trafficking crime.” The case was not even close (6–3). I dissented. Now I cannot say whether my colleagues in the majority voted the way they did because they are strict-construction textualists, or because they are not textualists at all. But a proper textualist, which is to say my kind of textualist, would surely have voted with me. The phrase “uses a gun” fairly connoted use of a gun for what guns are normally used for, that is, as a weapon.
When you ask someone “Do you use a cane?” you are not inquiring whether he has hung his grandfather’s antique cane as a decoration in the hallway.
Smith v. United States, 508 U.S. 223 (1993).
Common-Law Courts in a Civil-Law System 99 [SCALIA] But while the good textualist is not a literalist, neither is he a nihilist. Words do have a limited range of meaning, and no interpretation that goes beyond that range is permissible. My favorite example of a departure from text -and surely the departure that has enabled judges to do more freewheeling lawmaking than any other -pertains to the Due Process Clause found in the Fifth and Fourteenth Amendments of the United States Constitution. It says that no person shall “be deprived of life, liberty, or property without due process of law.” It has been interpreted to prevent the government from taking away certain liberties beyond those, such as freedom of speech and of religion, that are specifically named in the Constitution. (The first Supreme Court case to make that extension, by the way, was Dred Scott20 — not a desirable parentage.) Well, it may or may not be a good thing to guarantee additional liberties, but the Due Process Clause quite obviously does not bear that interpretation. By its inescapable terms, it guarantees only process. Property can be taken by the state; liberty can be taken;
even life can be taken; but not without the process that our traditions require -notably, a validly enacted law and a fair trial. To say otherwise is to abandon textualism, and to render democratically adopted texts mere springboards for judicial lawmaking.
Besides being accused of being simple-minded, textualism is often accused of being “formalistic.” The answer to that is, of course it’s formalistic! The rule of law is about form. If, for example, a citizen performs an act -let us say the sale of certain technology to a foreign country -which is prohibited by a widely publicized bill proposed by the administration and passed by both Houses of Congress, but not yet signed by the President, that sale is lawful. It is of no consequence that everyone knows both Houses of Congress and the President wish to prevent that sale. Before the wish becomes a binding law, it must be embodied in a bill that passes both Houses and is signed by the President. Is that not formalism? A murderer has been caught with blood on his hands, Dred Scott v. Sandford, 60 U.S. (19 How.) 393, 450 (1857).
The Tanner Lectures on Human Values bending over the body of his victim; a neighbor with a home-video movie camera happens to have filmed the crime; and the murderer has confessed in writing and on videotape. W e nonetheless insist that, before the state can punish this miscreant, it must conduct a full-dress criminal trial that results in a verdict of guilty. Is that not formalism? Long live formalism. It is what makes a government a government of laws and not of men.
II I described yesterday the common-law system of judicial lawmaking that has acquired such a firm grip upon the American legal mind and discussed its unfortunate extension into the field of statutory interpretation, which has been accomplished principally by replacing a search for the meaning of the text with a supposed search for the unexpressed intent of the legislator. I described briefly what I consider to be the proper approach to statutory interpretation, which I am content to call textualism, and distinguished that from strict constructionism. Today I intend to discuss some of the techniques of statutory interpretation, good and bad, and to raise some special considerations applicable to the construction of constitutional texts.
Textualism is often associated with rules of interpretation called the canons of construction -which have generally been criticized, indeed even mocked, by the legal commentators. Many of the canons were originally in Latin, and I suppose that alone is enough to render them contemptible. One, for example, is expressio unius est exclusio alterius. Expression of the one is exclusion of the other. What it means is this: If you see a sign that says children under 12 may enter free, you should have no need to ask the proprietor whether your 13-year-old can come in free. The inclusion of the one class is an implicit exclusion of the other. Another frequently used canon is noscitur a sociis, which means, literally, “it is known by its companions.” It stands for the principle that a word is given meaning by those around it. If you tell me “I took the Common-Law Courts in a Civil-Law System [SCALIA] 101 boat out on the bay” I understand “bay” to mean one thing; if you tell me “I put the saddle on the bay” I understand it to mean something else. Another canon -perhaps representing only a more specific application of the last one -is ejusdem generis, which means “of the same sort.” It stands for the proposition that when a text lists a series of items, a general term included in the list should be understood to be limited to items of the same sort. For instance, if someone speaks of using “tacks, staples, screws, nails, rivets, and other things” the general term “other things” surely refers to other fasteners.
All of this is so commensensical that, but for the fact it is Latin, you would find it hard to believe anyone could criticize it.
But in fact, the canons have been attacked as a sham. As Karl Llewellyn put it in a derisive piece in the 1950 Vanderbilt Law Review that is much cited: “[T]here are two opposing canons on almost every point. An arranged selection is appended. Every lawyer must be familiar with them all: they are still needed tools of argument.” 21 Llewellyn appends a list of canons in two columns, the left-hand column headed “Thrust,” and the right-hand column “Parry.” But if one examines the list, it becomes apparent that there really are not two opposite canons on “almost every point”-unless one enshrines as a canon whatever vapid statement has ever been made by a willful, law-bending judge. For example, the first canon he lists under “Thrust,” supported by a citation of Sutherland, is “A statute cannot go beyond its text.” Hooray for that. H e shows as a “Parry,” with no citation of either Sutherland or Black (his principal authorities throughout), the following: “To effect its purpose a statute may be implemented beyond its text.” That is not a generally accepted canon, though I am sure some willful judges have used it, the judges in Church of the Holy Trinity, for example. And even if it were used more Karl N. Llewellyn, “Remarks on the Theory of Appellate Decision and the Rules or Canons about How Statutes Are to Be Construed,” Vanderbilt Law Review 3 (1950) : 395, 401.
The Tanner Lectures on Human Values than rarely, why not bring to the canons the same discernment that Llewellyn brought to the study of common-law decisions? Throw out the bad ones and retain the good. There are a number of other faux canons in Llewellyn’s list, particularly in the “Parry” column.
For example, Parry No. 8: “Courts have the power to inquire into real -as distinct from ostensible -purpose.” Never heard of it.
Mostly, however, Llewellyn’s “Parries” do not contradict the corresponding canon, but rather merely show that it is not absolute. For example, Thrust No. 13: “Words and phrases which have received judicial construction before enactment are to be understood according to that construction.” Parry: “Not if the statute clearly requires them to have a different meaning.” Well of course. Every canon is simply one indication of meaning; and if there are more contrary indications (perhaps supported by other canons) it must yield. But that does not render the entire enterprise a fraud -not, at least, unless the judge wishes to make it so.
Another aspect of textual interpretation that merits some discussion is the use of certain presumptions and rules of construction that load the dice for or against a particular result. For example, when courts construe criminal statutes, they apply -or should apply, or say they apply- what is known as the “rule of lenity,” which says that any ambiguity in a criminal statute must be resolved in favor of the defendant. There is a rule which says that ambiguities in treaties and statutes dealing with Indian rights are to be resolved in favor of the Indians. And a rule, used to devastating effect in the conservative courts of the 1920s and 1930s, that statutes in derogation of the common law are to be narrowly construed.
And another rule, used to equally devastating effect in the liberal courts of more recent years, that “remedial statutes” are to be liberally construed to achieve what is called their “intended purposes.” There is a rule that waivers of sovereign immunity are to be narrowly construed. And a rule that it requires an “unmistakably clear” statement for a federal statute to eliminate state sovereign immunity.
Common-Law Courts in a Civil-Law System [SCALIA] 103 T o the honest textualist, all of these rules and presumptions are a lot of trouble. It is hard enough to provide a uniform, objective answer to the question whether a statute, on balance, more reasonably means one thing rather than another. But it is virtually impossible to expect uniformity and objectivity when there is added, on one side or another of the balance, a thumb of indeterminate weight. How “narrow” is the narrow construction that certain types of statute are to be accorded; how clear does a broader intent have to be in order to escape i t ? Every statute that comes into litigation is to some degree “ambiguous” ; how ambiguous does ambiguity have to be before the rule of lenity or the rule in favor of Indians applies? How implausible an implausibility can be justified by the “liberal construction” that is supposed to be accorded remedial statutes? And how clear is an “unmistakably clear” statement? There are of course no answers to these questions, which is why these artificial rules increase the unpredictability, if not the arbitrariness, of judicial decisions. Perhaps for some of the rules that price is worth it. There are worse things than unpredictability and occasional arbitrariness. Perhaps they are a fair price to pay for preservation of the principle that one should not be held criminally liable for an act that is not clearly proscribed; or the principle that federal interference with state sovereign immunity is an extraordinary intrusion.
But whether these dice-loading rules are bad or good, there is also the question of where the courts get the authority to impose them. Can we really just decree that we will interpret the laws that Congress passes to mean less or more than what they fairly say?
I doubt it. The rule of lenity is almost as old as the common law itself, so I suppose that is validated by sheer antiquity. The others I am more doubtful about. The rule that statutes in derogation of the common law will be narrowly construed seems like a sheer judicial power-grab. Some of the rules, I suppose, can be considered merely an exaggerated statement of what normal, nothumb-on-the-scales interpretation would produce anyway. For The Tanner Lectures on Human Values example, since federal elimination of state sovereign immunity is such an extraordinary act, one would not normally find it to have been implied -so something like an “unmistakably clear” statement rule is merely normal interpretation. And the same, perhaps, with waiver of sovereignty immunity.
I want to say a few words -the time available will not allow me as much as I would like- about the use of legislative history in interpreting statutes. My view that the objective indication of the words, rather than the intent of the legislature, is what constitutes the law leads me, of course, to the conclusion that legislative history should not be used as an authoritative indication of the meaning of a statute. This was of course the traditional English,
and the traditional American, practice. Chief Justice Taney wrote:
In expounding this law, the judgment of the court cannot, in any degree, be influenced by the construction placed upon it by individual members of Congress in the debate which took place on its passage, nor by the motives or reasons assigned by them for supporting or opposing amendments that were offered. The law as it passed is the will of the majority of both houses, and the only mode in which that will is spoken is in the act itself; and we must gather their intention from the language there used, comparing it, when any ambiguity exists, with the laws upon the same subject, and looking, if necessary, to the public history of the times in which it was passed.22 That uncompromising view generally prevailed in this country until the present century. The movement to change it gained momentum in the late 1920s and 1930s, driven, believe it or not, by frustration with common-law judges’ use of “legislative intent” and phonied-up maxims to impose their own views -in those days views opposed to progressive social legislation. I quoted yesterday from an article by Dean Landis inveighing against such judicial Aldridge v. Williams, 44 U.S. ( 3 How.) 9, 24 (1845) (emphasis added).
Common-Law Courts in a Civil-Law System [SCALIA] 105 usurpation. The solution he proposed was not the banishment of legislative intent as an interpretive criterion, but rather the use of legislative history to place that intent beyond manipulation.
Extensive use of legislative history in this country dates only from about the 1940s. It was still being criticized by such respected
Justices as Frankfurter and Jackson as recently as the 1950s. Jackson, for example, wrote in one concurrence: