Hostname: page-component-78c5997874-ndw9j Total loading time: 0 Render date: 2024-11-07T22:28:12.373Z Has data issue: false hasContentIssue false

Beyond Legislative Intent

Published online by Cambridge University Press:  08 October 2024

Barbara Baum Levenbook*
Affiliation:
North Carolina State University, Raleigh, NC 27695, United States
Rights & Permissions [Opens in a new window]

Abstract

There is a widely held view that legislative intention determines the meaning of a statute. The focus of this article is meaning in the sense of full linguistic content, which may not be the same as legal content. Linguistic intentionalism appears to have its greatest appeal when a statute has a partly implied linguistic meaning. It seems natural to suppose that the part of the meaning that is implied by the explicit wording in the statute is determined by an intention of the legislature—roughly, what the legislature intended to imply. The purpose of this article is to present some new reasons to doubt linguistic intentionalism and to offer an alternative for a particular kind of implied statutory meaning that some philosophers of language call implicitures. Rather than legislative intention, the meaning of implicitures is metaphysically determined by intention-free pragmatic linguistic norms, the grasp of which is part of the language competence of ordinary law subjects. The use of these norms, in combination with certain facts about background social practices and other conventions, generates implicitures that may be inconsistent with actual communicative intentions. These intention-free pragmatic norms, and the implicitures they generate, are explanatorily prior to the fact—if there is such a fact—of what impliciture it is reasonable to infer that the enacting legislature (or the legislators collectively) intended. Reference to legislative linguistic intention (actual or reasonably imputed) is thus explanatorily inert.

Type
Research Article
Copyright
Copyright © The Author(s), 2024. Published by Cambridge University Press

[T]he texts that represent the legal authorities’ communications of their determinations about what ought to be done mean only what the authorities intend them to mean.Footnote 1

As my initial quotation suggests, there is a widely held view that legislative intention determines the meaning of a statute. The view represented by Alexander is taken as an article of faith by many legal scholars. It is, they think, the rationale behind the frequent contention of judges in several countries, including England, Australia, Canada, and the United States, that statutes should be interpreted in accordance with legislative intent conveyed explicitly or implicitly by the statutory text. In this article, I will call the view that the full linguistic content of a statute is determined by legislative intentions concerning meaning or determined by inference to such intentions linguistic intentionalism, hereafter referred to as intentionalism.Footnote 2 Alexander's quotation states the heart of intentionalism.

I am not sympathetic to the general approach that assigns legislative intentions the main metaphysical role in determining statutory application, legal content or, in the present discussion, linguistic meaning of statutes for a variety of reasons, some of which are published elsewhere and some are detailed later in the article. However, I admit that such an approach appears to have its greatest appeal when a statute clearly has a partly implied linguistic meaning. It seems natural to suppose that the part of the meaning that is implied by the explicit wording in the statute is determined by an intention of the legislature—roughly, what the legislature intended to imply. In ordinary conversation, if what you said implied something fuller than the explicit language, it seems reasonable to assume that what it implied is what you intended to imply. It seems a small step to conclude that this picture is writ large in the case of legislatures and statutes, especially if we add two ideas. The first is that legislatures must somehow have exercised authority in the creation of the statute. The second is that this exercise of authority involves authoring the linguistic meaning, in the sense of assigning the statute a meaning—including what is implicit.

The purpose of this article is to reject (linguistic) intentionalism and sketch an alternative for a certain class of statutory meaning that philosophers of language call implicitures. The project also applies to city ordinances, by-laws, and other legislated texts and written regulatory instruments in law. Some of the project, as we shall see, can be applied to referenda proposals and bills submitted for legislation. However, for the sake of simplicity, this article focuses on statutes. The solution will require a shift in focus, from the maker(s) of statutes to the law subjects whose compliance with or obedience to the statutes is sought.

I. First Things

This article is concerned with intentionalism as a theory of the full linguistic content of statutes and with developing an alternative, not with what sets the legal content, or the legal effect, of statutes. The two may come apart. The second issue has long been considered important in the literature, but the issue under consideration in this article is important in its own right. All reasonable theories of legal content agree that the linguistic meaning of a statute is at least a large factor in determining legal content.Footnote 3 Since implicitures are sometimes part of the linguistic meaning of statutes, what fixes implicitures can affect legal content, and is thus of interest to legal theorists.

We might approach my main thesis, which makes a metaphysical claim, from the epistemic, following Alexander's lead. In order for statutes to communicate a message to law subjects (e.g., a standard of conduct), there must be uptake by them. For uptake, there must, on intentionalism, be a correct inference to legislative meaning-intention(s) or a correct inference to the reasonably attributed meaning-intention of the legislature (or legislators collectively). Yet intentionalist theories uniformly fail to explore in depth how inference to intention works in the case of statutory implicitures. When one examines how law subjects can grasp statutory implicitures, one discovers, I maintain, a body of norms that themselves establish the existence and content of the impliciture. Specific legislative intention to imply the impliciture and reasonable attribution of such an intention then drop out of consideration, becoming explanatorily inert. It is not legislative intention, however that is understood, nor reasonable inference to it, but what lies beyond legislative intention that engenders a statutory impliciture.

This article is structured as follows. Part II discusses and clarifies the notions of statutory meaning and statutory interpretation employed in this essay. Part III explains the idea of implicitures and applies it to statutes. Part IV provides a closer look at (linguistic) intentionalism and a brief critique of one or two prominent defenses of it. Part V exposes two root problems with intentionalism. Parts VI through VII develop and defend a novel account of statutory implicitures. Part VIII replies to an objection that can be expected from proponents of a dominant theory of pragmatics, and Part IX summarizes and addresses a few related issues.

II. Clarifying Terms

Both “statutory meaning” and “statutory interpretation” are phrases that have invited confusion and conflations—even equivocations—in the literature, so it is important to be clear about the use of these phrases in this essay. What I have called the full linguistic content of a statutory provision is the kind of meaning at issue. For example, one might ask what the following statutory text means:

Whoever has sexual intercourse with a person without the consent of that person is guilty of a felony.

One mainstream, and intuitive, initial reply is that the sentence means that a person commits a felony by having sexual intercourse with another without that person's consent. This reply treats the meaning as preserved by certain transformations of the sentence (e.g., “Any person who has non-consensual sexual intercourse with another person is guilty of a felony”) and by sentences in other languages (e.g., translations into German), regardless of who utters them. (As we shall see, one form of linguistic intentionalism rejects this mainstream reply.) Here I want to note that even on this mainstream view the provision may mean something richer, and implicit, beginning with “Whoever has sexual intercourse with a living person without the consent of that person.” (The existence of this implicit content has been a bar to some U.S. state prosecutions for necrophilia.)

This article is not making claims about the analogue for a statutory provision of the extension of a predicate. Call the former the statutory application. It may be thought of as the set of all of the act-tokens satisfying the statutory provision in all of the circumstances in which it applies. One might ask, “Does the statutory provision of rights to a ‘member of a family living in a common household’ mean those rights are held by a relative in the household who is there only for care during a period of his illness?” This question is using “mean” in a different sense, and asking for (part of the) application. For several reasons, questions of statutory application (according to local law) do not dissolve into questions of straightforwardly applying the full linguistic meaning of a statute.Footnote 4

When a statutory sentence uses a vague predicate and the case arises in the borderline, the court may be asked to make a choice about a potential (and putative, according to one of the parties) application. That choice may be in repair of the law, and there may be various normative considerations that should govern the choice, which might include fidelity to a perceived intention of the legislature. This use of the idea of legislative intention, and this kind of interpretation, is not at issue in this article. (There are other reasons for this kind of interpretation, such as ambiguity, but they can be set aside here.)

But sometimes statutory interpretation involves simply clarifying—or at least assuming—the (or some relevant part of the) full linguistic content of the statutory provision in question, and applying it directly to the instant case. This article is about what makes the clarification, or the assumption, involving a word or phrase that is not explicit in the statutory text correct. (Linguistic) intentionalism represents one answer.

III. On Statutory Implicitures

Bach set out the idea that there are conversational implicitures originally in 1994.Footnote 5 Implicitures, he said, are the part of linguistic meaning implicit but not explicit in an utterance. They are, he said in another work, expansions or completions of what is explicitly stated.Footnote 6 When an utterance has an impliciture, what is conveyed by that utterance is not its literal meaning, but something else (closely related), something richer.

Conversational implicitures arise, according to Bach, because of two kinds of semantic under-determination in the conversational utterance. (The explanation that follows is geared to sentences and assertions, not to language that states norms or rules.) In one kind, the sentence must be completed in order to yield a proposition. (“Hard to say”; “Too expensive”; “Maybe next time”). In the other, the sentence yields a proposition, yet communicates not it but a richer or expanded closely related proposition.Footnote 7

There are other ways in which information is conveyed beyond what is explicitly said. There are, for example, implicatures. They do not arise from semantic incompleteness in a particular remark. They are not, in Bach's terms, implicit in an utterance, but are said to be implied by the setting of the utterance in its conversational (or textual) context, or the manner of its delivery. Implicatures in Bach's terminology are additional messages implied by the language used and its context of usage. “In implicature,” Bach claims, “one says and communicates one thing and thereby communicates something else in addition.”Footnote 8 For example, there is the conveying of a low opinion of the philosophical abilities of a student applying for an academic philosophy position in the following a letter of recommendation: “Dear Sir, Mr. X's command of English is excellent, and his attendance at tutorials has been regular. Yours, etc.”Footnote 9

I am going to adopt Bach's rough and ready distinction between implicitures and implicatures, and will not address implicatures specifically in this article. How to characterize the difference between implicitures and implicatures more precisely is a difficult question, and one that will not be addressed here. (In this connection, I am uncertain about the possibility of determining on some defensible grounds for every candidate whether it is an impliciture or an implicature.) If it should turn out that much of what I say applies to implicatures as understood by others writing about pragmatics, or as understood by Bach, so much the better for my thesis.

One of the clearer examples of impliciture expansion offered by Bach is this: “I haven't done my homework.” The normal, perhaps standard, impliciture is: I haven't done my current homework.Footnote 10

Examples of remarks, commands, and questions with implicitures abound in ordinary conversation. “What happened?” “I quit.” “Have you eaten?” “Take the car.” “Are you doing anything on your birthday?” “Where is everybody?” It should be clear that what is implicit varies with the conversational context. For example, “What happened?” when asked of the distraught babysitter by the returning parents may have the impliciture What happened while we were gone in this household to have upset you so, but when asked of his stockbroker by a shocked investor, its impliciture is something quite different.

To avoid misunderstanding, we should note that conversational implicitures are defeasible. A remark attributed in 1941 (apparently incorrectly) to the American comedian Groucho Marx illustrates the point. Marx is said to have remarked, after a social evening, “I've had a perfectly wonderful evening … but this wasn't it.”Footnote 11 Defeasibility is often taken as a property of inferences, and as such it is the inference that is defeated by the addition of premises stating the information. I will say that the defeasibility of implicitures consists in the fact that more information will destroy or cancel the generation of them. That information in the foregoing example comes from what the speaker explicitly adds. “I've had a perfectly wonderful evening,” delivered at the end of a social evening to the host or hostess prima facie implicitly includes this evening. However, the addition of “this wasn't it” cancels the generation of the impliciture.

To anticipate, I will claim that statutory implicitures, though not conversational, are defeasible. They can be defeated through additional verbiage in the statutory text. It is an interesting question, and one that cannot be explored here, what else can defeat them. Can they be defeated by the verbiage in a judicial opinion “interpreting” the statutory provision inconsistently with its prima facie impliciture? The answer depends in part on whether the “interpreting” application is based on factors that trump the full linguistic meaning of a statutory provision. But there are other issues. For example, do interpreting judges have the (linguistic) authority in a legal system to establish and regulate the existence or nonexistence and content of a statutory impliciture?

I will not be discussing implicitures precisely as Bach understands them. Bach is very clear that implicitures are something speakers mean,Footnote 12 and I shall shortly challenge the centrality of speaker meaning for statutory implicitures. Bach's general idea, however, is that implicitures are something “directly” conveyed by an utterance that is “more elaborate” than its semantic meaning but closely connected to it. I will say that I am talking about what a (portion of a) text says (or, if you prefer, states or asserts or prescribes or evaluates or directs) that is more elaborate than its semantic meaning but closely connected to it. Implicitures can be inferred, but I follow Bach in holding that not everything readily inferable is an impliciture. To take an example from law: If Rex orders his subjects to wear red on a particular day, it can be inferred that Rex wishes his subjects to wear red on that day, or has a positive attitude toward their wearing red on that day. However, these propositions are not implicitures. Bach's explanation is that these propositions are no part of what the speaker (Rex) meant by his order. Mine is that these propositions are not part of the linguistic content of Rex's order. In contrast, implicitures are part of full linguistic content.

This article began by assuming that statutes have implicitures. There are cases where statutes state one thing explicitly, or literally, but something richer and closely related is implicit. The reader may recall the rape provision in Part II. Four additional examples may now be offered.

I discussed the first, which has attracted some attention in the literature, in an earlier article.Footnote 13 A U.S. federal statuteFootnote 14 states that certain stiffer penalties must be visited upon a defendant who “during and in relation to … [a] drug trafficking crime[,] uses … a firearm.” In Smith v. United States,Footnote 15 the defendant offered to barter an (unloaded) automatic weapon for drugs, was convicted, and was given the stricter punishment under this statute. Since the semantic meaning of the text does not answer the question of the purpose for which a firearm must be used, the semantic meaning cannot answer the question of whether the statute applies to Smith's act once the empirical facts about Smith's act and its circumstances are established. Enter the impliciture: Soames, following Bach, claims that we may infer that the whole implicit phrase is: uses … a firearm as a weapon.Footnote 16

I have also discussed the second example in an earlier article.Footnote 17 A Canadian by-law specifies that all drug shops shall be “closed at 10 p.m. on each and every day of the week.” The impliciture ends with something like and shall not reopen until the morning. In Regina v. Liggets-Finlay Drug Stores Ltd, defendants closed their drug shop at 10 p.m., then opened it a few minutes later. They claimed that they had complied with the by-law. The court held that the defendants had not complied. It seems clear that the court thus paid attention to the impliciture.Footnote 18

The third example involves a statute entitled the “Massachusetts Public Ways and Works Statute.” Section 21 of the chapter, “Repair of Ways and Bridges,” provides that notice of claims for injuries from ice and snow suffered on private property must be given in writing to the owners within a certain period of time. The relevant part of the impliciture seems to be that it is notice of claims for injuries suffered from outdoor ice and snow. However, in Smith v. Hiatt, an action for injuries was brought by a nurse who looked after a baby in a private home.Footnote 19 On a July morning, the nurse slipped on ice that Mrs. Hiatt had dropped on the kitchen floor while defrosting the refrigerator. The Supreme Judicial Court of Massachusetts held that the statute applied to this case. The court did not look at chapter or statute headings, and did not treat the section as having an impliciture. The omission is precisely what seems overly literalistic about the decision.

A fourth example involves a fictional statute discussed by Pufendorf, Blackstone, and mentioned by Judge Earl in Riggs v. Palmer.Footnote 20 I have also discussed this example previously.Footnote 21 The setting is seventeenth-century Bologna, where dueling has become a serious social problem. A statute expressly prohibits “drawing blood in the streets.” A surgeon aids a passerby in an emergency and is prosecuted under the statute. The surgeon has clearly drawn blood in the streets, but it is not plausible to hold that the statute applies to his case. The reason is that the statutory text has an impliciture, the relevant part of which is something like by antagonists using a weapon. (This fictional statute and the point are also discussed in United States v. Kirby.Footnote 22)

The issue before us is how best to account for the existence and content of such implicitures. Once more, intentionalism has a deceptively simple answer: they are what the creator of the statute (whether it is the legislators aggregately or the legislature as super-agent)Footnote 23 intended, or what it is reasonable to infer that the creator intended. The next two sections are devoted to casting some new doubts on this view.

IV. Why Intentionalism?

Surprisingly, there have been few direct arguments for linguistic intentionalism. The assumption that statutes are legislative communications is not usually defended.Footnote 24 It is sometimes presented as entailed by a wider theory of communicative linguistic content.

In its subjective form, the wider communicative content theory holds that linguistic content is determined by (actual) speaker (or writer) meaning-intention.Footnote 25 The addressee or audience must infer the intention in order to grasp the meaning. A Gricean theory postulates that the addressees assume that the speaker (or writer) intends them to figure out what the speaker means, and this gets translated in intentionalism as the assumption that the legislature intends its law subjects to figure out its full meaning-intention for a statute.

The extreme subjective form of the wider communicative content theory is unacceptable to anyone (including me) who thinks that speakers can misspeak by saying what they do not intend to say (perhaps thinking they have said something else). A common example is misusing a foreign language one imperfectly understands—for example, meaning to greet an old woman in Greece with the Greek equivalent of “Good morning!” (“Kalimera”), and saying “Calamari” (squid) instead.Footnote 26 Some semantic encoding or expression meaning must be postulated to account for this misspeaking. An independent linguistic meaning must also be postulated to allow the explicit language to be a clue to the speaker's intended meaning. Finally, an independent linguistic meaning must be postulated to account for the fact that, within and without law, commands and rule formulations that fail to communicate their producers’ meaning-intentions (in a case of misspeaking or because there is no corporate meaning-intention) nonetheless can and do guide behavior. (For example, she said, “Open the door,” meaning open the window, so he went to the door and opened it.) Such guidance would be possible only if these linguistic devices have independent linguistic content.

Many are willing to acknowledge that there is semantic encoding (even sentence meaning). They agree that it is an ingredient in linguistic content, which they may identify with speaker meaning. They will, nonetheless, insist that what is implied—and this includes implicitures—is fixed entirely by speaker meaning-intention. On this more sophisticated view, it is subjective speaker meaning-intentions that account entirely for implicitures, in law and elsewhere.

Unfortunately, there seems to be an analogue of misspeaking for implicitures. Consider the following scenario: The guests are leaving the party and thanking the host. One guest, Alfred, is autistic. He hears other guests saying to the host, “I've had a wonderful evening.” Not catching the implied tonight, thinking the other guests are simply reporting past experiences, and remembering that he has enjoyed himself (on another occasion that was not social), Alfred remarks to the host, “I've had a wonderful evening, too.” In this example, Alfred intended no impliciture, but it isn't clear that there wasn't one. Certainly, “I didn't mean to imply that” doesn't entail “I didn't imply that.” In some countries, such as Australia, defamation law can be construed as recognizing this lack of entailment. It is not a defense that the speaker did not intend to imply such-and-such if a reasonable person would have taken the published remark as “conveying” such-and-such.Footnote 27

There is also an objective intentionalist theory of wider communicative content. Its best-known form goes like this: a linguistic artefact has the full linguistic meaning that a rational hearer (or reader), knowing the relevant background and context, would be warranted in taking the speaker (or writer) to have intended it to have.Footnote 28 The meaning may, of course, involve an impliciture. Applied to legislatures on the assumption that the best theory for face-to-face conversational communications is also the best theory for statutes, the full linguistic meaning of a statute (including implicitures) is what it is reasonable to believe the legislature (or the legislators collectively) intended to communicate by the statutory text.

Such a theory can say that a good deal of the meaning of statutory provisions is semantically encoded. That encoding prevents, for example, it being reasonable to infer that in saying “It is prohibited to fail to crate, cage, leash or otherwise restrain a pet in a state park,” the legislature meant that it is permitted to allow one's pet to roam freely in state parks. As with the second version of the subjectivist account, this theory will, however, claim that statutory implicitures are not encoded (or not normally or typically encoded), and that their existence and content depend upon what impliciture it is reasonable to infer the legislature meant (intended) in the statutory text.

Such a theory might be accused of confusing an epistemic question—how do law subjects know the full linguistic meaning of a statute?—with a metaphysical one—what are the determinants of full linguistic meaning of a statute?Footnote 29 However, I will not pursue this point here.

There is a superficial argument for intentionalism based on legal practice that can be easily dismissed. It was touched on in the introduction. There are widespread practices of alluding to legislative intention, and of using as alleged evidence information in addition to the text of the statute, in familiar legal systems when statutes are being officially interpreted and applied.Footnote 30 One might assume that judges and other officials know what they are doing. Hence, if they think reference to an earlier draft of the statute or to the official legislative debate on the related bill is relevant to determining the interpretation or construction of the statute, then it might be argued that there are weighty reasons to agree. That puts the burden on the philosophical theorist denying these relevancies to produce a compelling reason to the contrary.

The problem with this argument was foreshadowed in Part II. Although it is incontrovertible that such practices exist, such practices are not indisputably, let alone manifestly, practices of discovering the full linguistic meaning of statutes. There are at least three possible ends that the use of such alleged evidence, and compliance with the associated conventions, might serve: (1) discerning the full linguistic content of the statute; (2) discerning the contribution to law that the statute makes; and (3) assigning a meaning to a statute to repair its linguistic content. The first two ends are logically distinct. In certain legal systems, for example, a statute may fail to make any contribution to law because of its linguistic content (as when in the United States a statute contravenes one of the amendments of the federal Constitution). It follows that activity (1) can be carried out, and is logically prior to, activity (2). Moreover, absent a definitive theory of the nature of law, we must leave open the possibility that a statute makes a contribution to law that is inconsistent in some significant way with its linguistic meaning. The interaction of statutes with each other, with judicial decisions, with common law, and so forth may be such that the statutory content is transformed, in its legal effect, by these other factors.Footnote 31 For example, a statute unqualified in its claim that any person may bring an action for relief on a certain basis may actually only contribute the legal right to bring a lawsuit within four years of the existence of the statutory cause of action, due to a previous statute of limitations on civil suits. These inconsistencies between linguistic content and legal contribution would provide a further reason that some considerations relevant to activity (2) are not relevant to activity (1).

Therefore, we need a convincing philosophical argument that alleged evidence of legislative intent is sought as part of activity (1). Even if all participants in the practice agreed that it is (which is doubtful), their view would not be conclusive unless the possibility that they have confused, conflated or blurred the distinction between the three activities can be eliminated.Footnote 32

The introduction touched on elements of some additional arguments for linguistic intentionalism—usually underdeveloped—that can be found in the literature: the idea of legislative authority, a very stringent view that authorship entails and is entailed by a creative control of meaning, and the assumption that legislation is not relevantly different from ordinary conversation with regard to meaning.Footnote 33 There is, however, little need to develop these arguments or to look for others; and I shall resist the temptation to catalogue the weaknesses of the ones proffered. For, quite apart from familiar objections to it,Footnote 34 intentionalism faces two fundamental problems. (I will assume, for the sake of argument in the rest of this article, that it is coherent to talk of legislative meaning-intentions and that at least sometimes the legislature can intend a specific linguistic meaning for a statutory text.)

V. Two Root Problems

The first fundamental problem confronts objective intentionalism. Recall that objective intentionalism holds that the full linguistic meaning of a statutory text, including its impliciture, is fixed by what it is reasonable to infer that the enacting corporate legislature (or perhaps legislators collectively) intended it to have. If the theory is true, some inferences must be reasonable, and some must not. There must be standards determining which inferences are reasonable and which are not.Footnote 35 Herein lies the vulnerability of the objective intentionalist view. It needs an account of the content of these standards, particularly with regard to implicitures. The latter cannot, on pain of vicious circularity, refer to what it is reasonable to infer the legislature intended. These standards must refer to some independent factor(s). We are entitled to ask: Why aren't these factors the metaphysical grounds of the impliciture?

In the following sections, I will argue that if there are such standards, the factors referred to are the true metaphysical grounds of the impliciture. It is because under the circumstances a statutory text bears a particular impliciture independently that it may be plausible to think the legislature (or the legislators collectively) intended that impliciture, or reasonable to ascribe a hypothetical meaning-intention to them. The idea of inference to a legislative meaning-intention effectively drops out of a metaphysical account of statutory meaning, and in particular an account of statutory impliciture.

The second fundamental problem is that any version of the intention model makes the wrong methodological assumption. Recall that intentionalism takes the meaning-intentions (or the reasonably inferred meaning-intentions) of officials (in this case, legislators) or official agencies (the legislature) as metaphysically determinative of meaning (and especially of implicitures). This is a mistake, and not because the wrong officials (e.g., legislators instead of judges) are chosen, but because the metaphysical role is assigned to what is true of or what is reasonable to attribute to the official producers of the texts. What is true of or what is reasonable to attribute to the potential readers of the texts, and in particular what we might call their end users—the law subjects—is either configured to fit or ignored altogether.

I propose to shift the standpoint and put potential law subjects in the center of the picture. To be clear, the idea is not that the factors engendering implicitures are in the law subjects’ direct and deliberate control. Rather, these factors are intersubjectively accessible to potential law subjects, both proximate and remote to the enactment of the legislation. When we add that the factors are not shared assumptions about legislative meaning-intention or what would be reasonable to assume is legislative meaning-intention, such a move has an immediate advantage over both forms of intentionalism. From this position, there is hope of accommodating the reasonable assumption that if the legislature—or the voters in a voter-initiated statute—approve(s) without verbal changes a bill or referendum proposal, the full linguistic meaning of the bill or proposal that was before them must be identical to its linguistic meaning after approval. We can explain why, if the text had an impliciture in the proposal, the text has the same impliciture after approval. The meaning is preserved because the determinants of meaning are the same.

In contrast, the assumption of meaning-preservation cannot be accommodated by standard intentionalist theories. For there is no legislative meaning-intention at the time that the bill or referendum proposal comes up for a vote. Moreover, if it is relevant information that the legislature has yet to act (and one assumes it is), no inference to a legislative meaning-intention for the bill or proposal is reasonable.

Of course, the text in the bill or proposal is not meaningless; so perhaps someone sympathetic to intentionalism might maintain there are drafter's intentions that set its full linguistic meaning. However, since legislative meaning-intention is allegedly an independent factor, the determinants of bill and statutory text meaning are not the same according to subjective intentionalists. Something similar can be said about objective intentionalists. If the inference in question is inference to this independent factor, the determinants of bill and statutory text meaning are likewise not the same. Thus, on these intentionalist accounts, it will not be true that preservation of verbiage guarantees preservation of meaning (and, in particular, of implicitures). For there is no guarantee that the full meaning intended by drafters is identical to the one legislatively intended, or that what it is reasonable to think about the former is reasonable to think about the latter.

Out of an abundance of caution, I should add that these last remarks do not appear to apply to the version of objective intentionalism that holds that the full linguistic meaning of a statutory provision is the content of the meaning-intention it would be reasonable for the enacting legislature, or legislators collectively, to have held (see note 28). This view presupposes that there is an independent meaning, for there is nothing else that it would be reasonable for the legislature to have intended. However, if the position's proponents concede that the full linguistic meaning—including implicitures—of statutory provisions is fixed by factors independent of any sort of inference to hypothetical if not counterfactual legislative intentions, they concede much of my thesis. The only issue that might remain between us then would be the specification of these factors.

VI. A New Approach

The work so far has suggested a position on statutory implicitures that does not privilege the actual (impliciture) meaning-intention of a corporate legislature or a collectivity of legislators (or drafters), nor does it privilege a reasonable inference to it. On this view, statutory texts bear implicitures arising from another source. The view requires elaboration, to which I now turn.

In very general terms, statutory implicitures are created by facts about the statutory formulation and its written context and other (background) facts about social practices that are common knowledge and are picked out by what can be called, broadly speaking, pragmatic (language) norms and certain other language norms or conventions, semantic and syntactic. (A comparable remark can be made about implicitures in a bill or referendum proposal.) These norms are grasped by competent readers of the language in question.Footnote 36 Their use, in combination with the aforementioned facts, results in a pragmatically enriched public meaning for the statutory text.

It is worth emphasizing that the pragmatic norms in question do not make essential reference to legislative meaning-intentions. It is true that these pragmatic norms are norms for using certain information in addition to the text of the statutory provision in question. However, that information is not about the producers’ intentions or about what is likely true of (or reasonably inferred about) the producers’ intentions, collectively or corporately. I call these additional norms intention-free pragmatic norms.

The intention-free pragmatic norms are intersubjectively accessible to law subjects competent in the language in which the statute is written. With the promulgation of the (whole) statute, the determinants of meaning are accessible to law subjects. It follows that, on the proposed account, what creates statutory implicitures is a combination of something already intersubjectively accessible to law subjects, both proximate and remote in space and time to the context of enactment, and something that becomes intersubjectively accessible to them upon promulgation of the statutory text. On this approach, statutory meaning shifts from being seen as the (reasonably inferred) speaker's meaning of a legislative speech act,Footnote 37 and becomes more like an analogue of utterance meaning that might be called a statute's text meaning. This text meaning may or may not be distinct from legislative meaning, depending upon what analysis one accepts for the latter.

The foregoing picture affords an easy explanation of how statutes can convey formal, public, durable standards in writing for law subjects, both immediate and remote in time and space from the act of enactment. For this view treats the full linguistic meaning as the one law subjects can grasp, both when they have almost no information about the “author” (whoever or whatever that is taken to be) or the context of enactment (or of drafting), and also when they do have this information.

Clearly, the lynchpin of my view is the characterization of the pragmatic norms in question and the claim that they generate statutory implicitures. Why believe this? The reply begins by explaining the kind of impliciture at issue.

Statutory implicitures illustrate a phenomenon in everyday language use: the existence of a special type of impliciture. Consider this piece of apocryphal movie dialogue (mistakenly attributed to American comedian Groucho Marx and American actress Margaret Dumont). Marx has just made an inappropriate sexual pass at a particularly staid upper-class matron he has met. Drawing herself up, she exclaims:

  1. (1) “Well, I never!”

To this, Marx replies,

  1. (2) “Oh, you must have!”

Sentence (1) is semantically underdetermined and, in this conversation, has a conversational impliciture. On an intention model of conversational impliciture, the matron, by uttering (1), conversationally implies what she intended to imply, namely that she has never been so offended, or words to that effect. Part of what Marx is doing in his reply is continuing his rudeness (and a campaign to embarrass the lady?) by perversely failing to interpret sentence (1) as the matron means it, and instead interpreting it as if it implies that the matron has never engaged in sexual behavior. This is a remark she would never make to Marx under the circumstances, as all those in the scene, and in the movie audience, know full well.

The dialogue (1)–(2) looks like a case of (feigned) misunderstanding, not what the speaker meant. But it does not follow that it is a case of misunderstanding, at least not entirely, what the speaker's remark meant—that is, its full linguistic meaning. Marx's reply is funny, and it is funny only because it is true.Footnote 38 Moreover, Marx's reply is not a non sequitur. It is not linguistically inappropriate, however morally or socially inappropriate it may be. It determines a semantically underdetermined sentence, sentence (1), in a way that is in keeping with the dialogue surrounding sentence (1). Remember, Marx brought up the subject of sexual behavior immediately before sentence (1) was uttered. So sentence (2), Marx's reply, is a linguistically eligible interpretation of sentence (1). There is something that makes it eligible, a mechanism, for which, as far as I can tell, philosophy of language has developed no technical term. Grice writes about “generalized” conversational implicatures, which may be close.Footnote 39

I will say that Marx, in sentence (2), is responding to a contextual impliciture of sentence (1). What evokes a laugh in the audience is in part a surprised recognition that the matron, by uttering sentence (1), has created not one impliciture, but two.Footnote 40 One is a contextual impliciture. Contextual implicitures are not produced by specific intentions of the utterers of the sentences that have them. Indeed, as in this example, contextual implicitures may be contrary to specific communicative intentions of those utterers. To revert to an earlier point, it is not plausible, as everyone sees at once, to suppose that the matron in this example had the intention to communicate what Marx takes to be implied; nor is it rational to impute such an intention to her. There is, in short, no account on which what is contextually implied is intended by the originator of sentence (1).

The phenomenon of unintentional contextual impliciture occurs in writing, too. There are signs and notices that are funny—and also socially or morally inappropriate—because of unintended but obvious contextual implicitures. The signs and notices carry, as it were, one impliciture too many. This sort of thing is—or at any rate was—the stuff of late-night television comedy in the United States.Footnote 41 I offer two examples. The first is a sign posted at one point by the Church of the Cross, a Methodist church, which read:

  1. (3) “Don't let worries kill you. Let the church help.”Footnote 42

The second is an alleged newspaper headline:

  1. (4) “Panda mating fails; veterinarian takes over.”

The point is that these signs and notices can carry one impliciture too many only if recognizing contextual implicitures is not a mistake under the normal circumstances in which the linguistic artefacts are read.

The reader may recall the claim that implicitures are defeasible. I will now say that in these examples, the contextual impliciture is not, or not yet, defeated. As a rough approximation, that is because there is no (or not yet) explicit verbiage in the context to defeat it.

Once we search for contextual implicitures, it becomes clear that they are not uncommon. Indeed, some are standard in that they become the default implicitures; they become the implicitures hearers (or audiences or readers) presume. Consider the following examples. In each, the relevant portion of the contextual impliciture is in parentheses.

  • “Can anybody hear me?” (besides me)

  • “I've had a perfectly wonderful evening.” (this evening)

  • “I have (exactly) two children.”

  • Host to guest at restaurant: “Order whatever you like.” (on the menu)

  • Bartender to patron: “What (drink) will you have?” (from the bar)

  • Speaker, holding a door open and facing oncoming hearer: “(I will enter and release the door) After you.” (enter)

  • Sign on dishwasher: “(The contents of this dishwasher are) dirty.”

  • Sign on restaurant: “No guns allowed” (to be brought into this restaurant)

Normally, in these cases, someone who hears or reads any of the above can grasp the standard contextual impliciture. Call this person a competent addressee. If the interest is in the intention of the speaker or writer, the standard contextual impliciture forms, as it were, a baseline. In the absence of defeaters, the competent addressee may reasonably infer—at least, in a conversational setting with another person—an intention to match the standard contextual impliciture. If there are defeaters, the defeaters of the inference operate on this baseline. A remark may have more than one standard contextual impliciture. One may be intended and the other(s) may not.

Another attempt to secure the point begins with an example suggested by Goldsworthy.Footnote 43 There are ceremonial uses of language that incorporate a standard contextual impliciture, whether or not intended. The standard contextual impliciture of the “I do” uttered at the appropriate point by groom at a wedding ceremony using The Book of Common Prayer begins, “I do take this woman to be my lawful wedded wife, to have and to hold, from this time forward” and ends with mentioning parting at death. Suppose the groom is trying to make a bad joke and intends a different impliciture—for example, I do intend to part from her at death. No one makes a mistake in holding that what the groom said, asserted, and conveyed was the standard contextual impliciture instead.

Finally, contextual implicitures can be exploited for purposes of misrepresentation. I adapt and somewhat reverse an example Greenberg has provided in which, for political purposes (e.g., to gain adherents, to avert public criticism, to avert a veto by the executive), the sponsors of a piece of legislation choose a text that has a misleading impliciture diverging from (some of) their true intentions in enacting the legislation:

(5) Suppose in a healthcare bill, it is provided that “federally funded facilities [as defined elsewhere] may provide abortions if necessary to save the mother's life.” The literal meaning of this [provision] … leaves open whether federally funded facilities may provide abortions in other circumstances …Footnote 44

There is a common phenomenon called “perfecting the conditional,” in which an “if” clause standardly implies “if and only if.” An example can be found in this remark by a father to his teenage son:

(6) I will give you ten dollars if you wash my car within the hour.

In perfecting the conditional in the healthcare bill, the impliciture of the statutory provision would be that abortion may be provided if and only if necessary to save the mother's life. Such an impliciture might be expected to be read into the statute by those opposed to abortion, which may include many of the sponsors’ constituents and the executive whose signature is necessary to turn the legislation into law. One can imagine that the sponsors intend, however, to leave the provision open to the possibility of permitting abortions not necessary to save the woman's life. They foresee the conveying of, but do not intend to imply, the perfection of the conditional. (I leave it entirely open whether the intentions of the sponsors ought to count as the intentions of the legislature.)

The dialogue in (1)–(2), the sign (3), the headline (4), and some of the other examples above demonstrate that the intended impliciture, when it exists, and a contextual impliciture may diverge.

We can now say that the implicitures found in statutory texts are contextual implicitures. If we are confident that use as a weapon is implied by the statutory provision on drug trafficking using a firearm, it is because, as Scalia tries to explain in his dissent in Smith v. United States, uses a firearm has a standard contextual impliciture that comes from the normal purpose of adopting firearms (i.e., as weapons, rather than as mediums of exchange), just as uses a cane has a standard contextual impliciture that comes from the purpose of adopting canes (to aid walking) (and there are no defeaters).Footnote 45

But how can implicitures “come from” purposes or normal background social practices? The answer returns us to the idea of intention-free pragmatic norms. They are intention-free in at least two respects. Since the generators of the contextual implicitures for statutes are neither intentions nor reasonable inference to intentions, if there are norms governing these implicitures, the norms cannot make essential reference to intentions (or reasonable inference to intentions). Moreover, as in the examples from ordinary speech, such norms can generate results that are contrary to actual meaning-intentions and to reasonable inferences to them.

I have maintained that there are such norms. Such norms can be or are ordered by weight. Competent addressees know these norms, share them, and apply them appropriately in grasping implicitures in statutory provisions. The ability of competent addressees to do so helps explain the efficiency and rationality of promulgating statutory texts. This is a point to which I will return at the conclusion of this article.

A similar account of pragmatic norms might be offered to explain the ability of competent addresses to grasp implicitures for some written rules outside of the legal setting and perhaps for posted signs such as “No trespassing” and “Out to lunch.” Such an account, which cannot be developed or defended here, would explain the efficiency and rationality of promulgating the former and posting the latter on the basis of public pragmatic norms that make the full linguistic content intersubjectively accessible, finesse some temporal and spatial distances, and can operate in “at remove” settings.

On the theory I am advancing, statutory provisions that carry implicitures can convey them to law subjects because the law subjects are competent in the application of the impliciture-generating norms. Legislatures (or legislators collectively) would be able to communicate intentions to imply, if such intentions exist, by anticipating (that is, if legislatures can anticipate) the result of the use of these intention-free pragmatic norms, and applying these norms themselves (that is, if legislatures can apply such norms) in advance.Footnote 46 Law subjects are not, in this process, gathering evidence about another agent—even a collective or corporate one. Nor are law subjects making reasonable inferences about another agent's meaning-intentions. Law subjects converge on a statutory content on a different basis, one that provides an independent meaning-content for the statutory text. If it were reasonable to infer a legislative intention to mean that content, the fact of that inference, or the existence of that intention, would thus do no explanatory work.

VII. The Norms

The account needs sharpening in at least one respect: it needs an inquiry into the intention-free pragmatic norms generating statutory implicitures. I will offer four norms. They may seem familiar to readers, for they echo canons or principles of statutory interpretation used in several different legal systems.Footnote 47

It bears emphasizing that these norms are not intended to account for all standard contextual implicitures; in particular, the norms are not intended to account for standard implicitures in all non-legal settings—though an analogue of some of these norms may apply. In particular, it may apply to nonlegal textual material with uses similar to those of statutes.Footnote 48

Normality

The type of act, situation, circumstance or agent referred to in the statute is the normal or expected type unless the difference is made explicit in the statutory language. (The normality norm has a corollary as a maxim for legislative drafters: If you want people to do or refrain from doing something normally done one way or under one circumstance in a different way or under an unusual circumstance, you must make the latter explicit.)

This is the place where the previously mentioned background social practices, other than those responsible for the semantics and syntax of the language in question, play a part in generating statutory implicitures. The application of this norm explains the example with which we began, the rape statute with an impliciture beginning: Whoever has sexual intercourse with a living person without the consent of that person. It also explains other examples previously considered.

So, for example, when the U.S. federal statute provides for stricter penalties for committing a narcotics offense using a firearm, the statutory language implies that the penalties are for using a firearm in the expected manner—as a weapon. Because implicitures are defeasible, we must add: the impliciture isn't cancelled by explicit language to the contrary.

So, for example, when the Canadian by-law specifies that all drug shops shall be “closed at 10 p.m. on each and every day of the week,” this implies the normal or expected kind of business closing for the day. Normally, such closings are long enough to permit a cashing out of the till, a cleaning of the premises, some economic advantage to turning off the lights, and so on. Again, the impliciture isn't cancelled by explicit language to the contrary.

So, for example, when Pufendorf's anti-dueling statute prohibits “drawing blood in the streets,” this implies the normal way in which, at that time, blood would be drawn in the streets —that is, by an antagonist using a bladed weapon.

Unity of Subject

In a written text, a change of subject is explicitly indicated in ways appropriate to the conventions of the genre. (The corollary as a maxim to writers is: make the change of subject explicit in ways appropriate to the genre.) The absence of a change of subject marker defeasibly implies a unity of subject.

In the case of a U.S. statute, a change of subject is explicitly indicated by a change in section numbers, chapter numbers, title numbers, part numbers, or headings. So, for example, in the chapter labelled “Counterfeiting and Forgery” in a federal statute, the paragraph forbidding the making of counterfeit money and the paragraph forbidding possessing counterfeit money are on the same general subject. The first paragraph of subsection 485 (of 18 U.S.C.) explicitly restricts a prohibition on making counterfeit money to U.S. currency. The second paragraph of this same subsection prohibiting possession of counterfeit money lacks this qualifying language, but it implies it, under the circumstances.Footnote 49

Relevance

All parts of a single statute that meet the unity of subject requirement are relevant contributions, including the statutory preamble, statutory title and subtitles of sections. (The corollary maxim for legislative drafters is a special adaption of one of Grice's maxims for conversational speakers: Make each contribution to the statute relevant.)

So, for example, the relevance of the section title, “Repair of Ways and Bridges” is presupposed in the previously mentioned Massachusetts statutory provision about claims for injuries from ice and snow suffered on private property. The impliciture is about outdoor ice and snow (on an outdoor way or bridge).

Limitation to Jurisdiction

Absent explicit language to the contrary, the laws, persons, actions, circumstances, and locations referred to in a statute are limited to the jurisdiction of the enacting legislature.Footnote 50

Criminal statutes are often of the form, “Whoever … commits a crime/offence/misdemeanor” (also, “Any person who …,” “A person who …” and “No person shall …”). So, for instance, the British Parliament in 1934 enacted the Protection of Animals Act, one provision of which read:

No person shall promote, or cause or knowingly permit to take place any public performance which includes any episode consisting of or involving … riding, or attempting to ride, any horse or bull which by the use of any appliance or treatment involving cruelty is, or has been, stimulated with the intention of making it buck during the performance ….Footnote 51

The impliciture begins, “No person in the United Kingdom shall promote, or cause or knowingly permit to take place any public performance in the United Kingdom.” The statute would be grossly misread to be stating a rule concerning rodeos in the American states of Texas or Wyoming.

Similarly, if a statute coming out of the Tennessee legislature reads, “Anyone engaging in metal detecting in a state or municipal park commits a misdemeanor,” the impliciture is: “Anyone engaging in metal detecting in a state or municipal park in Tennessee commits a misdemeanor.”

So, for example, in that U.S. federal statute on counterfeiting and forging, a provision states: “Whoever, except as authorized by law, makes or utters or passes … any coins … intended for use as current money, whether in the resemblance of coins of the United States or of foreign countries, or of original design” commits a crime. What is implied is: “except as authorized by domestic law.” (A court rejected the government's claim that the phrase, “except as authorized by law” implies domestic or foreign law.Footnote 52

The discussion in this section has been exploratory and expository, rather than argumentative. It is not the thorough account that the subject deserves. The list of intention-free pragmatic norms I offer is not meant to be exhaustive. The discussion also does not include the weightings, as the reader will have noted. Weightings, I suspect, are system-specific. At the least, there will also be legal-system specific norms to add to this list, and some of the latter might outweigh some of the norms already listed. It is worth pointing out, however, that the list of norms above has the virtue of not being parochial.

VIII. An Objection: Beyond Contextual Norms is Speaker Intention

I have been hesitant about assuming that the norms for statutory contextual implicitures are identical to, or just special cases of, norms for contextual implicitures universally. But suppose we tentatively accept this assumption. It may then be suggested that the ultimate reason for beliefs about the existence and content of contextual impliciture norms is that they capture the implicitures normally intended by speakers. Alternately, it may be suggested that we borrow from a dominant theory of pragmatics for ordinary language and view the impliciture norms as having a purpose: to retrieve the implicitures normally intended by speakers.Footnote 53

Though it is a seductive thought that speaker intentions might be reintroduced as important in supporting a belief about the existence and content of the impliciture norms, the purpose of this section is to argue that whatever good sense there is in the idea does not undermine my thesis.

Recall the examples of obviously unintended contextual implicitures in (1)–(2), (3), and (4). To repeat a point made earlier, recognizing the contextual impliciture is not a linguistic mistake. Marx is correct to recognize the impliciture I never engaged in sexual behavior, just as the reader is correct that Let the church help kill you is an impliciture in (3). Since it is correct to recognize the contextual implicitures in question, the existence of contextual implicitures is rule- or norm-governed, as I have maintained.Footnote 54

Such norms are what some philosophers of law call social rules. One of the conditions governing the existence of a social rule is a regularity in behavior of adherents to it. It makes little, if any, sense to say that norms aim at retrieving one of their existence conditions. It follows that the claim that statutory impliciture norms retrieve implicitures normally intended by their speakers or other producers (such as corporate legislatures) makes little or no sense if what is normally intended by them is an existence condition of the norms in question.

Although one of the existence conditions of a social rule is a regularity in behavior of adherents to it, Hart famously made the point that social rules are not reducible to generalizations about social behavior.Footnote 55 (Such generalizations are observations about habits.) It follows that norms for contextual implicitures are not reducible to generalizations about linguistic behavior, including the behavior of speakers intending to convey an impliciture. So any claim about habits of most speakers (and writers)—including a regularity in the linguistic behavior of speakers intending an implicature—is insufficient to support a belief in the existence of contextual impliciture norms. There must be more.

A small divergence is in order here, as this discussion provides an opportunity to make good on my Part V claim that the “factors” engendering statutory implicitures are not shared assumptions about legislative meaning-intentions. Those factors are norms, I have argued. And norms can no more be reduced to shared assumptions, or any beliefs, than they can be reduced to habits.

All of this may be conceded by the critic, who nonetheless may assert that the driver and shaper of the content of the norms is what speakers generally intend as implicitures—that is, a generality about their impliciture intentions. This is to give speaker meaning-intentions alone the ultimate explanatory role for contextual implicitures. Such a move is, however, blind to the role of another existence condition for these norms: the beliefs of those who understand an impliciture in a linguistic artefact. In other words, how listeners or readers generally “fill in the blanks” of what they perceive as “missing words” helps to create these norms and determine their content. These hearers and readers need not understand an impliciture as intended by the producer in order for the norm to exist. The recipients may understand the impliciture as committed to by the producer of the remarks instead. Alternatively, the recipients may misunderstand the impliciture as semantically encoded. For the existence of the norm, it matters not why the hearers and readers fill in the blanks the way they do.

Though producer choice drives recipient understanding to some extent, we should not be blind to the interactivity involved, and in particular the truth of the converse. Recipient understanding and dispositions to understand also drive and shape producer choice—and so help to shape the contents of contextual impliciture norms.

Nonetheless, it might be protested, the regularity of linguistic behavior in everyday language whereby individual speakers or writers intending a “longhand” use a “shorthand” is relevant to the belief in the existence and content of contextual impliciture norms applying to statutes. The nature and type of that relevance is not, however, as straightforward as may be supposed. In particular, the connection is not straightforward if it is alleged to work through legislative meaning-intentions.

One tendency in the literature is to assume that legislative impliciture meaning-intentions are aggregations of individual legislators’ (or drafters’) impliciture meaning-intentions. There are difficulties with this idea, some of them elucidated in the literature on legislative intentions.Footnote 56 It is unclear which persons count, and whether any answer is defensible. When these individuals vary in their meaning-intentions, which may not be uncommon for impliciture meaning-intentions, is there no contextual impliciture in the statutory provision? Most readers would deny that when these individuals vary on other meaning-intentions (on, say, the semantics), the statutory provision has no linguistic meaning at all. The reduction of legislative meaning-intention to an aggregate of individual meaning-intentions for impliciture then appears somewhat unmotivated.

Unless there is a legislative meaning-intention independent of statutory contextual-impliciture norms, the idea of the normal legislative meaning-intention for implicitures would be useless for specifying an ultimate metaphysical generator of these norms. One import of the preceding remarks is that providing a defensible account of an independent legislative meaning-intention for implicitures is more challenging than many assume.

The main point I wish to press, however, is that even if the pragmatic norms I have been discussing for statutes are not intention-free in every respect—that is, do not exclude generalizations about speaker linguistic intentions in their existence conditions—they are free of the sort of immediate intention-dependence postulated by intentionalists.

IX. Last Things

I have argued, roughly, that statutory implicitures are contextual implicitures generated by an ordered set of norms accessible to law subjects, the grasp of which is part of their language competence. These norms make no essential reference to an intended impliciture nor to an impliciture it is reasonable to believe is intended—by the legislature or the legislators collectively. They generate implicitures even when (or if) it is not reasonable to believe this. The fact that the enacting legislature intends to imply an impliciture, if it is a fact, might have a bearing on explaining, causally, why the legislature issued a statute with particular sentences (if legislative intentions can be causes of legislative actions). However, it is irrelevant to explaining the existence and content of a statutory impliciture.

Note that the determinants of linguistic meaning for statutory texts are relatively stable. This feature, combined with the rest of the account, affords an explanation of the stability (and uniformity) of the full linguistic meaning—including implicitures—of statutes over a significant portion of space and time.

It is worthwhile in this last part to underscore two points. The first returns us to the question of where the account leaves inference to legislative meaning-intention and to a claim from Part V. The intention-free pragmatic norms, and the implicitures they generate, are explanatorily prior to the facts, if there are such facts, about what impliciture it is reasonable to infer the enacting legislature (or the legislators collectively) intended. The norms would explain the reasonableness of inferring that the legislature intended a particular statutory impliciture when this is the impliciture the text bears. Such an inference would, of course, be based on these norms. It would not be based on probabilities, as would an inference from a generalization about what legislatures normally intend.

But might there be conditions under which it would be reasonable to believe that the legislature intended a particular statutory impliciture that does not correspond to the independently-engendered impliciture? Such a question cannot be answered in the absence of a specification of what counts as evidence of a legislative intention to mean the impliciture, an adequate defense of the claim that this is evidence, and a specification of the degree of confirmation (or probability) necessary for an inference or belief about such a legislative meaning-intention to be reasonable. Without all of this, what can be said is that if it were sometimes reasonable to believe that the legislature intended an impliciture other than the one generated by intention-free pragmatic norms, this fact does no linguistic work. The legislature in such a case simply fails to produce a statutory text with that meaning.Footnote 57 The discussion of contextual implicitures shows that intention to imply is not necessary for the existence of this type of impliciture. It is not sufficient even when coupled with explicit verbiage. To create an impliciture in a statutory text, one must instead play by the (impliciture-generating) rules.

The second point returns us briefly to the matter of the full linguistic meaning of bills and referendum proposals. The account I espouse vindicates the previously-mentioned assumption that whenever the textual language remains unchanged during the legislative or referendum process, the implicitures are preserved. This feature and others in the account explain why it is possible for persons considering voting for the adoption or enactment of a bill or proposal to understand, upon reading, not only its full linguistic meaning but also, and thus in advance, what that meaning would be if the draft legislation or referendum proposal were to be adopted or enacted verbatim.

I end with a final thought, which adapts a normative argument previously advanced for my position on applications of statutes and returns us to the epistemic and the ideas of communication and uptake. A theory of statutory implicitures (indeed, of statutory linguistic content) is pro tanto less attractive if a rival theory does a better job of supporting the rationality of the practices of promulgating written statutory texts (for general epistemic guidance—and so, to law subjects) and the texts of proposals for voter-initiated statutes. As I have demonstrated, the theory of statutory implicitures advanced here does a better job on the latter than intentionalism. My theory may have a small advantage regarding the text-only promulgation of statutes over intentionalism, too—at least, in those legal systems in which there are no (system-specific) norms generating implicitures that are inaccessible to law subjects. (Note that jurisdiction implicitures are fairly common in statutes forbidding or requiring behavior among law subjects; and so, at least some implicitures can generally be counted on to be present in the set of statutes promulgated for a given jurisdiction.)

To belabor a point, on the approach advanced here when, for example, the legislature promulgates a sentence setting forth certain stiffer penalties for a defendant who “during and in relation to … [a] drug trafficking crime[,] uses … a firearm,” the semantic and syntactic norms of the language in question, together with intention-free pragmatic norms and certain facts, generate that impliciture. It is largely because of these intention-free pragmatic norms that law subjects who pay attention to the statutory text can (and normally will) understand—and converge in their understanding, though separated in space and time—that uses as a weapon is implied. Setting aside statutes that use technical language, everything else law subjects need is either a competence in language or can be assumed to be common knowledge. So promulgation of the text only is a rational enterprise if the goal is to give law subjects the opportunity to grasp its message. (If the goal is to guide them to actions required or permitted by the statute, then more needs to be said—in particular, about the relationship between grasping the full linguistic meaning and grasping its practical import—that is, the application of the statute to act-tokens in specific circumstances.Footnote 58)

In contrast, consider the position of subjective intentionalism. Promulgation of the text only (to convey the impliciture) is not rational in cases in which the legislature intends an impliciture other than the one generated by intention-free pragmatic norms. Additional information, usually about the context of enactment, is not common knowledge but will be necessary on this view for law subjects to grasp the impliciture. Moreover, there is nothing typically in promulgation to mark the cases where first impressions, based on intention-free pragmatic norms, are misleading.Footnote 59

In the light of the uncertainties surrounding relevant evidence, the situation is less clear with respect to objective intentionalism. If the reasonably-inferred-as-intended impliciture can be grasped with just the promulgated text (given language competencies and knowledge of certain social practices—the same factors I have identified as generating intention-free implicitures), then the theory espoused in this article has no advantage in justifying text-only promulgation practice over objective intentionalism. If, however, additional information is needed to draw the reasonable inference (such as about the context of enactment) and that information is not common knowledge across the temporal and spatial spectrum of affected law subjects, there is an advantage to my theory. For objective intentionalism will not justify text-only promulgation when the reasonably-inferred-as-intended impliciture differs from the one generated by intention-free pragmatic norms.

Of course, text-only promulgation is not always justifiable to convey full linguistic meaning, even on my own view. It will not be so justifiable when statutes use technical language that is not common knowledge across the temporal and spatial spectrum of affected law subjects. (This is true whether or not the statutory text has an impliciture.) That is not an alarming feature of my view, as both versions of intentionalism, for different reasons, also have this feature. The point is that my view may have a small advantage in supporting the rationality of text-only promulgation of statutes when technical language not common knowledge across the spectrum of affected law subjects is absent (but implicitures are present).

There may be other advantages of my theory, particularly in handling the case in which legislation must be signed by an executive with veto power, but the impliciture legislatively intended is not the impliciture intended by the executive doing the signing. However, the full discussion of this issue must await another forum.

Footnotes

*

I am indebted to Hrafn Asgeirsson, Michael Pendlebury, Robert Mabrito, David Auerbach, David Austin, Kevin Richardson, Stefan Sciaraffa, Luis Duarte D'Almeida, and Liam Murphy; to the participants of the October 19, 2023 NYU Colloquium in Legal, Political and Social Philosophy, the 2015 IVR World Congress Working Group 4, the March 12, 2015 seminar on legal argumentation at the Edinburgh Centre for Law and Society, and the 2012 Monash University Workshop on Law and Language. I have greatly benefitted from their comments on earlier drafts.

References

1 Alexander, Larry, All or Nothing at All? The Intentions of Authorities and the Authority of Intentions in Law And Interpretation: Essays In Legal Philosophy 381 (Andrei Marmor ed., 1995)Google Scholar.

2 I have discussed other views that I have called intentionalism in previous publications.

3 See Asgeirsson, Hrafn, Expected Applications, Contextual Enrichment, and Objective Communicative Content: The Linguistic Case for Conception Textualism, 21 Legal Theory 115, 119 (2015)CrossRefGoogle Scholar.

4 I first approached the idea of statutory application in Levenbook, Barbara Baum, How a Statute Applies, 12 Legal Theory 71 (2006)CrossRefGoogle Scholar. There, I referred to full linguistic meaning as “plain meaning.” See the brief discussion of statutory legal contribution, infra Section IV and of application to act-tokens, infra section IX and n. 52.

5 Bach, Kent, Conversational Impliciture, 9 Mind & Language 124 (1994)CrossRefGoogle Scholar.

6 Kent Bach, Impliciture vs. Explicature: What's the Difference? in Explicit Communication 126 (Belén Soria and Esther Romero eds. 2010). Bach thinks implicitures are expansions or completions of the semantic content, which he understands rather strictly. In this article, I will sidestep controversies about the limits of semantic content and, for that matter, the use of the elastic notion of what is said.

7 Some would put “Hard to say” in the second category, claiming it expresses a complete proposition but with unarticulated constituents. (I am indebted to Kevin Richardson for this point.)

8 Bach, supra note 5 126.

9 Example from Grice, H. P., Logic and Conversation in Syntax and Semantics 52 (Peter Cole and Jerry L. Morgan eds., 1975)Google Scholar.

10 Bach, Kent, Replies to My Critics 13 Croatian Journal of Philosophy 217, 222 (2013)Google Scholar.

12 Bach, supra note 6.

13 Barbara Baum Levenbook, Soames, Legislative Intent, and the Meaning of a Statute, in Pragmatism, Law, and Language (Graham Hubbs and Douglas Lind eds., 2014), pp. 40–55.

14 Title 18 U.S.C. §924(c)(1).

15 Smith v. United States 508 U.S. 223 (1993).

16 Scott Soames, Interpreting Legal Texts: What is, and What is Not, Special About the Law, 1 Philosophical Essays 414 (2009).

17 Levenbook, supra note 4.

18 R. v. Liggetts-Finlay Drug Stores Ltd [1919] 3 WLR 1025. This case has drawn much attention from philosophers and legal theorists. See, e.g., Rupert Cross, Statutory Interpretation 67 (2nd ed., 1987); Jeffrey Goldsworthy, Parliamentary Sovereignty 234–235 (2010); Hrafn Asgeirsson, On the Possibility of Non-Literal Legislative Speech in Pragmatics and Law: Theoretical and Practical Perspectives 90–91 (Alessandro Capone and Francesca Poggi eds., 2017).

19 Smith v. Hiatt, 329 Mass. 488 (1952).

20 115 N.Y. 506 (1889).

21 Levenbook, supra note 4.

22 United States v. Kirby 74 U.S. [7 Wall] 482, 487 [1868].

23 For the view that the relevant intention is the intention of the legislature as a super-agent, see Richard Ekins, The Nature of Legislative Intent (2012).

24 Exceptions include Andrei Marmor, The Language of Law 18 (2014); Mark Van Hoeke, Law as Communication (2002). There is an implied argument from the purpose of law in Paolo Sandro, The Making of Constitutional Democracy: From Creation to Application of Law 194 (2022). See also Deborah Cao, Legal Speech Acts as Intersubjective Communicative Action in Interpretation, Law and The Construction of Meaning 65 (Anne Wagner, Wouter Werner, & Deborah Cao eds., 2007) for an argument (more like an analysis) from the assumption, which is declined here, that legislation is a speech act and, in particular, from Habermas's account of speech acts. For noncommunicative models of statutory texts, see Hurd, Heidi M., Sovereignty in Silence 99 YALE L.J. 945 (1990)CrossRefGoogle Scholar and Michael Moore, Interpreting Interpretation in Law and Interpretation 1 (Andrei Marmor ed., 1995).

25 The reader will no doubt note that subjective intentionalism as a broad theory of language is at odds with the mainstream reply about linguistic content given in Part II.

26 Think also of slips of the tongue. This one is attributed to former New York Congresswoman Bella Abzug: “We need laws that can protect everyone. Men, women, straights, gay, regardless of sexual perversion …” (https://www.thoughtco.com/slip-of-the-tongue-sot-1692106). A four-year-old of my acquaintance recently said, “If you don't have juice for breakfast, you can eat oranges or mangroves.” One sees this phenomenon in the extreme in semantic paraphasia in post-stroke patients, who think they have said something else. See also Marmor, supra note 24 at 21 and Jeffrey Goldsworthy, Moderate versus Strong Intentionalism: Knapp and Michaels Revisited 42 SAN DIEGO L.R. 677 (2005).

28 Or, often, would be warranted “in taking as the communication-intention of the speaker (or writer).” See, e.g., Goldsworthy, supra note 26 at 680: “the meaning of an utterance is the meaning which evidence readily available to its intended audience suggests it was intended to mean.” See also Asgeirsson, supra note 18, at 74. (He maintains that it will rarely be the case that statutes communicate implicitures.) A variation of objective intentionalism for statutes holds that the full linguistic meaning of a statutory provision is the content of the meaning-intention it would be reasonable for the enacting legislature (or legislators collectively) to have held. David Tan discusses a similar view in terms of the correct “interpretation”: Tan, David, Objective Intentionalism and Disagreement, 27 Legal Theory 327 (2021)CrossRefGoogle Scholar, calling it lawmaker-objectivist. This position is strikingly at odds with standard objective intentionalism, as the inference is not to what meaning-intention it is reasonable to believe the legislature actually held but to what it ought to have held. Many of the arguments for intentionalism—indeed, the informal reasoning in the introduction to this article—are inapplicable to this position. However, I cannot develop this point here and will largely ignore this version in what follows.

29 As Neale warns us, we should not conflate the epistemic question of how meaning is known with the question of what determines linguistic meaning. Stephen Neale, Convergentism & the Nature of Law, sections 3.1–3.3 (March 14, 2013) (unpublished manuscript).

30 Goldsworthy remarks, “In leading cases and treatises in England, Australia, Canada, and the United States, it is almost universally asserted that the most fundamental principle of interpretation is that statutes should be interpreted according to the intentions which they convey, either expressly or by implication given the context in which they were enacted.” [note omitted] Goldsworthy, Jeffrey, Marmor on Meaning, Interpretation, and Legislative Intention 1 Legal Theory 450 (1995)CrossRefGoogle Scholar. Also see Scalia, Antonin, Common-Law Courts in a Civil-Law System: The Role of United States Federal Courts in Interpreting the Constitution and Laws in A Matter of Interpretation: Federal Courts and the Law 16 (1997/2018)Google Scholar: “You will find it frequently said in judicial opinions of my court and others, that the judge's objective in interpreting a statute is to give effect to ‘the intent of the legislature.’ This principle, in some form or another, goes back at least as far as Blackstone.” [note omitted]

31 See Mark Greenberg, The Standard Picture and Its Discontents in Oxford Studies in Philosophy of Law vol. 1 especially 75–80 (Leslie Green and Brian Leiter eds., 2011); Dale Smith, The Practice-Based Objection to the `Standard Picture’ of How Law Works, 10 Jurisprudence 502 (2019); Baude, William and Sachs, Stephen E., The Law of Interpretation 130 Harvard Law Review 1079 (2017)Google Scholar; and Levenbook, Barbara Baum, A Puzzle About Legal Systems and Democratic Theory, 11 Jurisprudence 157 (2020)CrossRefGoogle Scholar. See also Barbara Baum Levenbook, The Law of the Street, in new essays on the nature of legal reasoning (Mark McBride and James Penner eds., 2022).

32 Someone might want to argue from a communicative content theory of law to the claim that the practice of searching for legislative intent in appellate decisions is a practice of discovering the full linguistic meaning of the statute. It is true that appellate courts use language that appears to be a declaration of ex ante law, rather than an unambiguously performative speech act (making it the case that, for example, section such-and-such of a particular statute does not apply to private actors). However, we are entitled to ask for the reason to think that the language is not merely conventional and used indiscriminately whether or not the court is discovering ex ante law. Proponents will find no support from the claim that appellate judges think they are thereby always uncovering ex ante law. Their private writings reveal that they do not agree on this point.

33 I cannot hope to outline all the arguments for intentionalism in its various forms—linguistic and otherwise. A good beginning is Stoljar, Natalie, Survey Article: Interpretation, Indeterminacy and Authority: Some Recent Controversies in the Philosophy of Law, 11 The Journal of Political Philosophy 470, 475–476 (2003)CrossRefGoogle Scholar.

34 Various forms of intentionalism have been subject to criticism, some of it applicable to linguistic intentionalism. Most of it is either metaphysical (calling into question the coherence of the idea of a corporate communicative- or meaning-intention) or epistemic (calling into question the sufficiency of access to information about legislative meaning-intent). See Stoljar, id., at 476–480 for some of these criticisms. See also Levenbook, supra note 31. See also Levenbook, supra note 4 at 79, n. 10 and the criticism, id., 81 of an analogous view about statutory applications.

35 Also, from what context and given what information reasonableness is to be assessed. The point will be briefly revisited in Part IX.

36 When statutory texts use technical terms, or technical legal terms, or ordinary terms with special legal meanings (“possession,” “offer for sale,” etc.), those who are competent in the language in question may be only a portion of those subject to the application of the statute. So there may be limited access to one ground of full linguistic meaning—certain semantic norms. The theory I espouse may be adjusted to fit, but in order to illuminate the general character of my proposal, the adjustments will remain unexplored. I will largely describe the situation in which technical terms are not a factor in full linguistic meaning—including, of course, statutory implicitures.

37 Sandro, following Horner, prefers to speak of “text acts” of the legislature, as opposed to speech acts. Sandro, supra note 24.

38 A point made to me in private conversation by Michael Pendlebury.

39 Bach remarks that many examples of generalized conversational implicatures are cases of generalized conversational implicitures. Bach, supra note 6, at 127. Stephen C. Levinson agrees in his Presumptive Meanings (2000) 197.

40 Doubtless there is also a naughty delight that Marx flaunts social convention—and moral rules—to exploit the second of these implicitures.

41 Unintentional contextual implicitures in signs and advertisements were a staple for decades, along with written examples of garden-variety ambiguity, of the late Johnny Carson's comedy on the American late-night television show, The Tonight Show.

43 Goldsworthy, supra note 26 at 676. Goldsworthy is discussing an example from Jonathan Culler that was intended to show, according to Goldsworthy, “the meaning of an utterance can differ from the meaning intended by the speaker.”

44 Mark Greenberg, Legislation as Communication? Legal Interpretation and the Study of Linguistic Communication in LANGUAGE IN THE LAW 240 (Andrei Marmor and Scott Soames eds., 2011).

45 Supra note 15 at 241–247.

46 It makes sense to talk about a single legislator doing these things in a single-legislator legal system. I am not at all certain it is coherent to talk about these kinds of activities as those in which corporate legislatures can engage, when we do not employ an aggregative account of such activities. The view I favor is that corporate legislative actions are determined by, and limited to, those actions legislatures have procedures for doing, and the causing of consequences of doing so. So, for instance, legislatures can enact statutes because they have procedures for doing so. I am not aware of any legislative procedures for anticipating addressees’ linguistic expectations or applying intention-free pragmatic norms.

47 An anonymous reviewer drew my attention to the fact that many of these norms might be seen as special instances of “inference rules” proposed by Levinson. However, Levinson's rules are proposed for an empirical purpose, to “elucidate” the “heuristics that are actually employed” by recipients of messages and “help to resolve the logical problem of intention-recovery.” Levinson supra note 39 at 34–35. I think of the norms as constitutive, creating statutory implicitures. For further discussion on this point, see remarks in Parts VIII and IX.

48 Statutes, of course, are not the only written instruments containing normative language, exhortations, or instructions aimed at audiences remote in space and time and unknowable to their authors, nor is law the only place we find such instruments. Neale reminds us that there are instruction manuals for electrical appliances. Neale, supra note 29, section 1.3. Think also of cookbooks, gardening books, and first aid books.

49 See United States v. Falvey, 676 F.2d 871 [1982].

50 The impliciture might also be defeated by norms of private international law. I owe this point to Izabela Skoczén, Implicatures Within Legal Language 115 (2019).

51 Protection of Animals Act 1934 (c. 21) §1. Available at https://www.legislation.gov.uk/ukpga/Geo5/24-25/21/section/1/enacted

52 United States v. Falvey, supra note 49, at 877.

53 I am indebted to an anonymous reviewer for the suggestions.

54 This argument is made about all “implicit language” by Marina Sbisà, Implicitness in Normative Texts, in Pragmatics in Law: Practical and Theoretical Perspectives 23 (Francesca Poggi and Alessandro Capone eds., 2017). For present purposes, I do not make a distinction between rules, norms, and conventions.

55 H. L. A. Hart, The Concept of Law (1961), especially Ch. IV.

56 Perhaps the best-known source is Ronald Dworkin, Law's Empire 317–321 (1986), although Dworkin is talking about intentions regarding statutory application.

57 This repeats a point I made in Levenbook, supra note 13.

58 See, e.g., Levenbook, supra note 4 for what might be considered a pro tanto legal application account. Since law tends to be systematic, even more needs to be said if the goal is to guide law subjects to actions required or permitted by the local law, all-legal-things-considered. Recall the comments on statutory legal content in Part IV and see Levenbook, supra note 31.

59 This original version of this argument (with respect to statutory application) can be found in Levenbook, supra note 4 at 80–81.