Skip to navigation – Site map

HomeNuméros40Against the conversational model ...

Against the conversational model of legal interpretation

On the difference between legislative intent and speaker’s intention
Francesca Poggi
p. 9-26
Translation(s):
Contra el modelo conversacional de la interpretación jurídica [es]

Abstracts

The so-called conversational model claims that legal interpretation does not significantly differ from ordinary understanding: both are inferential processes complying with a pattern that can be traced back to Paul Grice’s thesis. This essay seeks to criticize such a claim, showing that the so-called legislative intent is not able to play the role that speaker’s intention plays in ordinary understanding and the Gricean thesis.

Top of page

Full text

1 Introduction

  • 1 See Dworkin 1986: 50ss.; Scott 2011; Canale 2017: 45ss.

1Currently in legal theory, there exists a very widespread thesis, according to which legislation is a form of intentional communication that is similar, if not identical, to ordinary conversation, and that therefore legal interpretation does not significantly differ from ordinary understanding. In particular, the so-called conversational model1claims that legal interpretation and ordinary understanding are articulated according to an analogous process, governed by similar inferential rules: specifically, both fit with a pattern that can be traced back to Paul Grice’s thesis.

2In this essay, after a brief presentation of Grice’s theory, I will advance some criticisms of the conversational model: I will seek to show that this model is not suitable for legal interpretation, since so-called legislative intent is not able to play the role that speaker’s intention plays in the Gricean model.

2 Grice’s conversational model of ordinary communication

3According to Grice the meaning of an instance of language use is the meaning that the speaker intends to communicate. More precisely, by uttering (and intending to utter) x, a speaker means S, if and only if she intends (a) that the hearer understands S; (b) that the hearer recognizes that the speaker intends (a), at least in part on the basis of utterance x; and (c) that the hearer understands S partly on the basis of the fulfillment of (b).

4Grice’s communicative intention is thus a very complex and reflexive intention: it’s my intention to let you understand what I mean through making you recognize my intention.

  • 2 Grice 1989: 22-40.
  • 3 The clarification ‘at least to a certain extent’ is indispensable in order to take account of cases (...)

5Grice’s conversational maxims fit within this pattern: they are tools, commonly and unconsciously employed, that enable us to fulfill the complex conditions above, especially the condition (b). In fact, Grice claims that conversations are governed by a series of maxims aimed at guaranteeing the efficient and effective use of language with a view toward interaction.2 According to Grice, everything we say is interpreted, as far as possible, in a cooperative manner, i.e. it is interpreted as if it were an appropriate contribution to the communication in which we are engaged. For this to take place, we often need to go beyond the meaning of what is said; we need to interpret what is said in a cooperative manner or, to be more precise, as though it were in accordance (at least to a certain extent)3 with the conversational maxims at stake.

  • 4 Grice 1989: 26.

6Grice identifies four maxims which, taken together, express the general cooperative principle (henceforth CP): ‘Make your conversational contribution such as is required, at the stage at which it occurs, by the accepted purpose or direction of the talk exchange in which you are engaged.’4 The maxims identified by Grice are the following.

Quality:
Try to make your contribution one that is true, specifically:
1. Do not say what you believe to be false;
2. Do not say that for which you lack adequate evidence.

Quantity:
1. Make your contribution as informative as is required (for the current purposes of the exchange);
2. Do not make your contribution more informative than is required.

Relation:
Be relevant.

Manner:
Be perspicuous, specifically:
1. Avoid obscurity of expression;
2. Avoid ambiguity;
3. Be brief (avoid unnecessary prolixity);
4. Be orderly.

7In order to clarify how the conversational mechanism works, let’s consider the following dialogue:

a)
A: What time is it?
B: The local news has already started.

8This dialogue seems perfectly comprehensible. Nevertheless, B’s answer, taken literally, is completely unrelated to A’s question: it is not an answer to A’s question. However, we interpret B’s reply as appropriate to A’s question, and, in order to do so, we attribute an additional meaning to that answer. More precisely, we assume that, in the context of dialogue (a), B is following the CP and the conversational maxims, and, in order to maintain such an assumption, we draw some inferences (some conversational implicatures, in Grice’s terms) from B’s answer (from what is said, in Grice’s terms), i.e. we infer some additional implicated meaning (implicata, in Grice’s terms) from B’s answer. Hence, in dialogue (a), B’s response can be considered appropriate to A’s question insofar as, in such a context, the utterance ‘The local news has already started’ communicates ‘I don’t know what time is right now – because, in accordance with the first maxim of Quantity, if I had known, I would have told you – but I can give you some information that – in accordance with the maxim of Relation – I understand and you can understand as pertinent to your question, i.e. some information from which I believe that you can approximately deduce the time, that is – to be brief, in accordance with the third maxim of Manner – The local news has already started.

9In a nutshell: it is the common (unaware) assumption according to which the CP and the conversational maxims are followed in a certain context which enables the speaker to implicate and the listener to infer a different and/or additional meaning. As Grice states:

  • 5 Grice 1989: 30-1.

A man who, by (in, when) saying (or making as if to say) that p has implicated that q, may be said to have conversationally implicated that q, provided that (1) he is to be presumed to be observing the conversational maxims, or at least the Cooperative Principle; (2) the supposition that he is aware that, or thinks that, q is required in order to make his saying or making as if to say p (or doing so in those terms) consistent with this presumption; and (3) the speaker thinks (and would expect the hearer to think that the speaker thinks) that it is within the competence of the hearer to work out, or grasp intuitively, that the supposition mentioned in (2) is required.5

10As it is clear, the meaning conversationally implicated is, according to Grice, an instance of speaker’s meaning. In a given context, by uttering (and intending to utter) p, a speaker conversationally implicates q if and only if she intends (a) that the hearer understand q; (b) that the hearer recognize that the speaker intends (a), at least in part on the basis of the utterance p, the assumption that the speaker is following the CP and the conversational maxims and the fact that p is consistent with the CP and the conversational maxims if the speaker intends (a); and (c) that the hearer understand q partly on the basis on the fulfillment of (b).

11Indeed the essential elements of Grice’s conversational model are the following: a complex kind of reflexive intention; what is said (i.e. a given utterance-meaning); a mutually known context; the common assumption, shared by speaker and listener, that they will both cooperate, i.e. that they will both follow the CP and the conversational maxims; and what is implicated (or the implicatum).

12Some of the previous elements are very important for my purposes, so we have to look at them more closely.

  • 6 Grice 1989: 25.
  • 7 Levinson 2000: 171. See also Reimer 1998: 598.
  • 8 See Recanati 1989. The theoretical analysis of the processes of pragmatic enrichment (expansion and (...)
  • 9 See Strawson 1973; Bach 1997; Rysiew 2007.

13Firstly, as far as the notion of what is said is concerned, it is worth noting that it does not necessarily correspond to (a kind of) literal meaning – that is, to an alleged a-contextual sentence-meaning. Conversely, ‘for a full identification of what the speaker has said, one would need to know (a) the identity of x [the reference], (b) the time of utterance, and (c) the meaning, on the particular occasion of utterance, of the phrase [uttered]’.6 In other words, what is said is ‘equivalent to the proposition expressed by the use of a sentence or the truth-conditional content of the utterance, and is in turn dependent on reference resolution, indexical fixing and disambiguation.’7 According to some authors, what is said is also determined by phenomena of pragmatic enrichment and, therefore, it is equivalent to a pragmatically enriched content, not fully determined by semantic and syntactic determinants.8 Obviously enough, it is not always the case that what is said implicates something else: in particular, what is said does not conversationally implicate anything when it is, by itself, fully cooperative – i.e. we have to refer to the CP and the conversational maxims in order to conclude that the speaker said what she said without implicating something else.9

  • 10 See Malpas 2002.

14Secondly, the importance of the context mutually known by the speaker and the listener has to be stressed. The notion of context is very complex and indeterminate.10 A deep analysis of such a notion is clearly beyond the scope of this essay. Instead, I will employ a broad and intuitive notion of context, just as Grice does. The important point is that conversational maxims are all context-sensitive. In this regard, Grice distinguishes between generalised implicatures and particularised implicatures: generalised implicatures are always produced, except in special circumstances, while particularised implicatures are produced only within particular contexts. So, for example, the utterance

(b) The flag is red.

15produces – in accordance with the first maxim of Quantity – the generalised implicature ‘The flag is only red’ (because, generally, if I knew that it was not only red, I should say it), while it produces the particularized implicature ‘You shall not swim’ only within the dialogue

(c)
A: Mummy, shall I swim?
B: The flag is red.

16This distinction could suggest that only particularized implicatures are context-sensitive: however, this conclusion is not correct.

17First, generalised implicatures are produced in standard contexts, but in special contexts they do not arise. Imagine the following scenario: my cousin and I are in front of a flag (and we are both aware of this fact); the flag is black and red; my cousin is colourblind, and he cannot distinguish red from green and brown, while he sees black without a problem; I know that he is colourblind, he knows that I know it, I know that he knows that I know it, and so on. If in that context my cousin asks me ‘Is the flag red?’ and I answer (b), my answer does not implicate ‘only red’, because in that scenario the relevant information is that the flag is red and not that the flag is only red – my cousin and I are both aware that it is also black (this is our common knowledge).

18Second, as the previous example shows, the same maxim can produce either generalised or particularised implicatures in different contexts: it is always the first maxim of Quantity that causes (b) to implicate ‘only red’ in standard contexts and not to implicate it in my cousin-scenario.

19Grice’s conversational model is not exempt from criticism, but assuming that it works as an explanatory model of ordinary conversation, we can ask whether it also works in the legal domain.

3 The conversational model of legal interpretation

  • 11 Raz 2009: 274.

20It is reasonable to believe that legislation is an intentional phenomenon. On one hand, statutes are made by persons empowered to change the law, and, as Raz observes, ‘it makes no sense to give any person or body law-making power unless it is assumed that the law they make is the law they intended to make’.11 On the other hand, legislation can play its typical function to direct behaviours only if the law-addressees understand what the legislature intended to say and become aware of being under the obligation to behave accordingly. As a consequence, it may be reasonably argued that the enactment of a statute is a communicative behavior.

21If law is a form of intentional communication, why do we not apply to legal statutes the same maxims that we apply in ordinary conversation? More generally, is the pattern sketched by Grice suitable for legal interpretation? Can we interpret legal statutes applying that pattern or a similar one?

  • 12 See Sinclair 1985; Miller 1990; Morra 2011; Ekins 2012; Carston 2013; Morra 2016; Slocum 2016; Sbis (...)
  • 13 See Sinclair, 1985; Miller 1990; Slocum 2016.
  • 14 See Chiassoni 1999; Poggi 2018.
  • 15 Sinclair 1985.

22Currently a lot of authors give an affirmative answer to those questions.12 In particular, while some authors maintain that it should be correct to apply Grice’s model also to legal interpretation, other theorists argue that it already happens: according to this last position, in fact, some interpretative criteria, some canons of interpretation or construction, usually employed by judges and jurists, are exactly the legal analogues of conversational maxims.13 Actually, different authors identify in different ways the legal interpretative criteria that, according to them, correspond to Grice’s maxims.14 Thus, for example, Sinclair claims that Grice’s four maxims correspond to the following well-known legal criteria:15

Quality:
1. Do not enact a provision that can be shown not to further the legislative purpose.
2. Do not enact a provision when there is no adequate evidence that it furthers the legislative purpose.

Quantity:
1. Make each provision cover all the persons and actions you intend it to.
2. Make each provision cover only the persons and actions you intend it to and no more.

Relevance:
Be relevant to the scope of the statute.

Manner:
Use the words the literal meaning of which gets closest at what you mean.

  • 16 Miller 1990.

23Miller, instead, proposes the following different, more complicated, and less clear matching scheme:16

Quantity:
1.
expressio unius est exclusio alterius; principle of the negative pregnant; statutes in derogation of common law should be strictly construed; repeals by implication are not favoured; a court should defer to long-standing judicial or administrative interpretation; legislative re-enactment of a statute endorses long-standing judicial or administrative interpretation; a clear statement is required for federal regulation of matters traditionally governed by state law; administrative agencies have greater discretion to interpret general statutes than specific statutes; the plain meaning of a statute ordinarily governs.
2.
Ejusdem generis.

Quality
1. Interpret statutes to avoid absurdity; interpret statutes to give agencies reasonable and necessary powers to effectuate their statutory duties.
2. In cases of economic regulation, courts should defer to legislative factfinding and should uphold statutes if the statute can be characterized as reasonable; in cases involving fundamental rights or suspect classifications, courts should not defer to legislative factfinding, but should instead require a close fit between legislative means and ends.
3. Give effect if possible to every word of the statute.
4. Reconcile conflicting statutes if reasonably possible.

Relation
1.
In pari materia.
2. Statutes should be interpreted in light of the legislature’s purposes.

Manner
1. Court should interpret words according to their ordinary, common senses; Court should give legal words their established technical meanings.
2. The plain meaning of a statute ordinarily governs.
3. Give effect to every word the legislature used;
expressio unius est exclusio alterius.
4. Be orderly.

  • 17 More recently Slocum (2016) has argued that the interpretive criterion called eiusdem generis is a (...)

24It is undeniable that, with the exception of some maxims proposed by Miller – which apply only in United States – the criteria above are employed in legal interpretation and legislators are supposed to know that they are employed.17 Is this enough to support the homogeneity between ordinary conversation and legal interpretation? I will argue that it doesn’t.

25My point is that, even assuming that some legal interpretative criteria are similar, in their content, to Grice’s maxims, the conversational model does not operate within legal interpretation: one of the reasons why it does not operate and that, in turn, makes evident that it does not operate is that in legal interpretation there is nothing that can play the fundamental role that the speaker’s intention plays within the conversational model.

4 Speaker’s intention vs. legislative intent

26As we have seen, Grice’s conversational model is grounded in a complex and reflexive type of communicative intention: by uttering (and intending to utter) x, a speaker means S if and only if she intends (a) that the hearer understand S; (b) that the hearer recognize that the speaker intends (a), at least in part on the basis of the utterance x; and (c) that the hearer understand S partly on the basis of the fulfillment of (b).

27For our purposes, it is useful to distinguish four levels of speaker’s intention:

  • 18 See Canale & Poggi 2019, and, on the two last levels, Asgeirsson 2017. Notice that according to Gri (...)

(i) the intention to utter (say, write) something;
(ii) the intention to utter (say, write) what is said (
x) and not something else;
(iii) the intention to mean (implicate, communicate) something (other/different) by uttering
x; and
(iv) the intention to mean (implicate, communicate)
S and nothing else by uttering x.18

28In the next paragraphs I will try to show that each one of these levels poses serious problems within legal interpretation.

4.1 The intention to utter something and the intention to utter what is said

29The first level – the speaker’s intention to utter something – is very important: it is what characterizes the communication and activates the interpretive activity as an activity aimed at understanding what the speaker intends to communicate. To clarify this point, consider the following example:

(d) These spots mean measles.

  • 19 Grice 1989: 213 ff.

30Here the verb ‘to mean’ refers to causal connections among spots and measles. It is a matter of regularity between certain fact and other facts of which the former constitutes the symptoms or the indications: there is no intention to communicate anything. This is Grice’s natural meaning.19 Similarly, imagine that the sea has washed up some shells on the beach: I can certainly argue that those shells have taken a shape that means ‘cat’, but here the verb ‘to mean’ does not refer to someone’s intentional communication: the sea has no communicative intention that those shells reveal. So even if I make random sounds or scrawl signs on a piece of paper, I cannot say that I intend to say something, because, again, there is no communicative intention. In other words, in the conversational model, the first level of intention is a necessary condition for communication to take place and it is also the conditio sine qua of any interpretative activity, seen as the activity that aims at understanding what a speaker intends to communicate.

  • 20 See Raz 2009: 284: “A person is legislating (voting for a Bill, etc) by expressing an intention tha (...)

31As far as this first fundamental level, the intention to utter something does not seem difficult to recognize in the legislative procedure. As Raz states “Only acts undertaken with the intention to legislate can be legislative acts” (Raz 2009: 282) and the very idea of an unintentional legislation is inconceivable. In particular, MPs intend to utter, to enact something, because they vote and they intend to vote in its favour.20 However, a problem arises.

  • 21 See below in the text and for a more detailed discussion Kutz 2000, and Ekins 2012. I thank my anon (...)

32The minority, who votes against the text, does not intend to enact anything. Only the majority intends to enact something, but it seems odd to conceive of legislation as the product of the intentional actions of the majority only. It is contrary to our institutional conventions that ascribe the enacting of statutes to the whole parliament and not to the individual MPs who voted in its favour. Perhaps this problem is not so serious as it appears at first. Joint action theory and some accounts of collective intentionality can explain how the intentional commitment – or the intentional participation – into the legal procedure suffices in order the enacting be seen as the intentional product of all MPS, including those who voted against (and even those who were not present at the vote).21 The second level of speaker’s intention – i.e. the intention to say what is said – poses greater problems.

  • 22 Surely this point is very discussed within post-Gricean and neo-Gricean literature: however, if we (...)

33We have already observed that the Gricean notion of what is said does not (necessarily) correspond to an alleged a-contextual sentence-meaning. Conversely, what is said is a complete meaning whose determination requires reference resolution, indexical fixing, disambiguation, and, at least according to some authors, pragmatic enrichments. It is important to stress that, according to Grice’s original account, it is the speaker’s intention that sets what is said.22 Now, the problem is that, even if the legislative intention to approve and enact something is recognizable, it is doubtful that all those who voted in favour of that text understand it in the same way.

  • 23 Greenberg 2011: 239.
  • 24 On this point see also Raz 2009: chap. 11; Lifante 1999.
  • 25 This is just one aspect of the more general problem of determining who are the subjects to be taken (...)

34Firstly, as Greenberg claims, ‘It is uncontroversial that most legislators do not read most of the text of the statutes on which they vote’:23 an intention to utter a certain (complete) meaning can be ascribed only to those who have read the text that expresses that meaning.24 Instead, it is difficult to argue that those who voted in favour of the text without reading it have this intention: they certainly intend to vote for a text whatever it is, but they do not have the intention to ‘utter and intend to utter x’ because they do not know the text they are voting for.25

  • 26 Greenberg 2011: 253.
  • 27 Surely this also happens in ordinary conversation. E.g., a speaker can utter an ambiguous or vague (...)
  • 28 See Smith v. United States508 U.S. 223 (1993).
  • 29 Actually, according to some authors, ‘to use a firearm in a way whatsoever’ is not the enriched mea (...)

35Secondly, the legislative assemblies often approve texts that are vague, ambiguous and indeterminate. Legislation is a question of negotiation and compromise: ‘As a result, the language is often chosen not in order to implement anyone’s communicative intention, but because, for example, it is unclear enough for a majority to accept’.26 Therefore, very often, the Parliament (and/or the majority that votes for the text) did not have a unique intention about what the text said.27 E.g., let’s consider the famous legal provision ‘Whoever, during and in relation to any crime of violence or drug trafficking […] uses [..] a firearm, shall, in addition to the punishment provided for such crime of violence or drug trafficking crime, be sentenced to imprisonment for five years’.28 The expression ‘to use a firearm’ can be enriched in a number of ways, like ‘to use a firearm as a weapon’, ‘to use a firearm in a way whatsoever (including scratching one’s head)’,29 ‘to use a firearm to commit a series of activities listed in another legal provision’. How is it possible that the majority had a single intention on how to enrich this legal provision, given that this supposed intention is not at all transparent? Is not more plausible to think that the majority agreed on the text and not on what it says?

36At this point one can wonder whether some theories of collective intention cannot solve also the above problems. I think they cannot.

37The most popular and plausible theories of collective intentionality usually require that collective intentionality must rise (emerge, etc.) from individual intentions with some (at least partially) coincident propositional content. But here the point at stake is precisely whether the parliamentarians’ individual intentions are directed to the same meaning. In other words, theories of collective intentionality do not suppose the existence of a collective mind capable of bearing collective intentions that are independent from the individual ones: collective intention is always the result of particular individual intentions related to each other in specific ways.

  • 30 This is the core of Bratman’s continuity thesis: ‘The conceptual, metaphysical, and normative struc (...)
  • 31 A detailed analysis of the theories on collective intention is beyond the scope of this essay, but (...)

38Consider for example Bratman’s famous theory. According to Bratman, collective intentionality derives from individual planning agency, so that we do not need any new basic elements in our metaphysics or philosophy of mind in order to account for robust forms of collective agency.30 It follows that there can be a legislative (collective) intention to utter x (and nothing else) only if each member of Parliament bears the intention ‘I intend that we intend to utter x (and nothing else)’. However, as we have seen, the existence of this intention is precisely the point at stake. 31

  • 32 In Ekins’ more sophisticated terms: ‘Each legislator forms the intention ‘I intend that we legislat (...)
  • 33 This point is even clearer if one adopts Kutz’s minimalist account of collective action. According (...)

39It is worth noting that this problem does not arise with respect to the intention to utter something because such an intention follows from the MPs single intentions to participate into the legislative procedure, accepting the rules that govern it. There is a legislative (collective) intention to utter (to enact) something only if each member of Parliament bears the intention ‘I intend that we intend to participate into the legislative procedure, by accepting the rules R, R1, Rn, that establish the conditions to enact a bill as law’.32 A similar move is not possible as far as the intention to utter x (what is said), and not something else, is concerned. We cannot maintain that the intention to utter x (what is said) follows from the intention to enact a given text (whatever it is) if that text does not express a precise meaning – if it is vague, ambiguous, indeterminate and the majority does not bear a unique intention about how to solve the ambiguities, vagueness, indeterminacies of the text approved.33

40One can object that, as far as legal statutes are concerned, the meaning of what is said is very often clear enough: very often it not ambiguous, vague or indeterminate but rather it amounts to the literal meaning of the legal provision and, therefore, is perfectly clear and transparent to both legislators and interpreters. This objection opens up a factual question that only an empirical investigation might solve. Anyway, I will now try to show that the last two levels – viz. the intention to implicate something and the intention to implicate S and not something else – raise further problems.

4.2. The intention to implicate something and the intention to implicate the implicatum

41The third and fourth levels of the speaker’s intention are decisive. As we have seen, according to Grice, the meaning of a certain utterance is the meaning that the speaker intends to communicate. Therefore, if the speaker, uttering (and intending to utter) x, does not intend that the listener understand S, then the speaker has not meant S and S is not the meaning communicated by x.

  • 34 On the indeterminacy and opacity of legal context see, e.g., Marmor 2008; Poggi 2013; Asgeirsson 20 (...)

42I hope it’s clear from what has been previously argued that attributing this intention to parliamentary assemblies is very problematic. All MPs, or at least the majority, should bear the intention to implicate S – viz. the intention (a) that the addressees understand S, (b) that the addressees recognize that they intend (a) at least in part on the basis of (i) what is said (whose identification, as we have seen, is problematic as well), (ii) the conversation legal maxims (whose identification has been discussed) and (iii) the legal context (which, according to many authors, is indeterminate and opaque),34 and finally (c) that the addressees understand S partly on the basis of the fulfillment of (b).

  • 35 See also Sperber & Wilson 1986: 63-4. I thank my anonymous reviewer for pointing out to me this ref (...)

43However, within the neo-Gricean literature, some authors have cast doubts over the centrality of the intention to implicate.35 Let’s consider the following dialogue:

(e)
A: When I arrived at the party, some guests had already gone.
B: Who remained?
A: Nobody.

  • 36 See Levinson 2000; Atlas 2005; Sbisà 2007; Morra 2011; Poggi 2011; Morra 2016; Poggi 2016; Sbisà 20 (...)
  • 37 For a discussion see, e.g., Poggi 2018.
  • 38 See Atlas & Levinson 1981; Macagno, Walton & Sartor 2018.
  • 39 Actually Greenberg speaks of ‘communicative content’, a notion that includes that of implicatum.

44Intuitively, it seems that, by uttering ‘some guests had already gone’, A implicates ‘not all guests had already gone’, regardless of what she intends to implicate. Different theoretical models try to account for this intuition.36 Broadly speaking, in those models the intention to implicate loses relevance in so far as implicatures occur even if the speaker does not intend them to. I will not discuss those models here.37 What is important to highlight is that, within those models, the process of understanding appears as an inference to the best explanation of what a cooperative speaker – i.e. a speaker who follows the conversational maxims – intends to communicate in a given context:38 as Greenberg claims, this seems to modify the notion of implicatum.39

  • 40 Greenberg 2011: 231. See also Bach 2005.
  • 41 Greenberg 2011: 231.

45According to classical Gricean theory, the meaning implicated ‘is the content of a certain kind of complex intention’;40 according to this new notion, which Greenberg labels ‘objective notion’, the meaning implicated is ‘what a member of the audience would reasonably take a speaker who had uttered the relevant sentence under specified conditions to have intended to communicate’.41

  • 42 Greenberg 2011: 248.

46In regards to these two notions of implicatum, conversational maxims play different roles: while according to the first notion maxims are instruments to infer what the speaker intends to implicate, according to the second notion they are constitutive of what a speaker implicates,42 since a rational speaker implicates what her utterance implicates according to the maxims.

  • 43 Dascal & Wróblewski 1991; Capone 2013; Macagno, Walton & Sartor 2018.

47The objective notion of implicatum seems suitable for legal interpretation: indeed, it could be argued (and it has been argued) that a legal provision implicates what a cooperative speaker (or a rational speaker43) should implicate according to the maxims (or, better, to their legal analogues), and the contextual assumptions of the listeners. Therefore, it would not be necessary to have a real legislative intention to implicate (something and not something else), and the contextual assumptions related to the legislators, and to what the legislators consider an object of mutual knowledge with the interpreters, would not be relevant.

  • 44 I think that this model can be labelled as listener-based because it leaves aside speakers’ actual (...)

48I will argue that this listener-based model44 represents a sophisticated reformulation of the old dogma of the rational legislator and of its more recent incarnations hinged on the concept of counterfactual legislative intention, and is based on a fundamental misunderstanding.

49I think that we have to carefully distinguish two notions often confused – even by Greenberg – namely the intention to implicate and the meaning implicated. In Grice’s classical approach, the intention to implicate plays two interconnected roles: it determines (and amounts to) the meaning implicated (the implicatum) and, consequently, it is the parameter to determine the conversational success, in the sense that communication is successful if what the listener understands is equivalent to what the speaker has intended her to understand. The models that argue that implicatures are independent from the speaker’s actual intention to implicate can give up the first role, not the second.

50In other words, we can concede that a speaker implicates what her utterance implicates according to the maxims and the context, but, if the real speaker’s intention does not coincide with what the speaker should intend (according to the maxims and the context), then we have to admit that the communication fails. This communicative crash is not a purely pragmatic phenomenon, but it characterizes every level of language. If I say ‘Pass me the pan’, while I intend to communicate ‘Pass me the pot’, I violate language semantic rules, and therefore the communication fails – and it fails because of me. Similarly, if A says ‘When I arrived at the party, some guests had already gone’, while she intends ‘all the guests’, A violates the first maxim of Quantity and she implicates ‘not all the guests’; therefore the communication fails – and it fails because of A.

  • 45 See Levinson 2000.
  • 46 See Poggi 2011; Poggi 2016. Maxims are hermeneutic technical rules in the sense that they ensure mu (...)

51It is worth noting that this does not amount to saying that conversational maxims are constitutive of what is implicated: maxims and context (to which all maxims are sensitive) are always instruments to infer what the speaker intends to implicate. They are very strong presumptions,45 or better still, they are hermeneutic technical rules based on a presumption of conformity that, to be overcome on a single occasion, requires considerable evidence to the contrary.46 What really matters is that the intention to implicate continues to play the fundamental role of parameter of communicative success: renouncing this role would mean renouncing it to explain pragmatic communication failures (those failures that do not depend on semantics or syntactic), and it would involve a radical distortion of the conversational model. However, in legal interpretation, this parameter is absent.

5 Conclusions. About the rational legislator (again)

52As we have seen, according to the objective notion of implicatum – as drawn by Greenberg (who nevertheless criticizes it) – a statute implicates what a cooperative (or rational) speaker should implicate according to the maxims, or, better, to the legal interpretative criteria that are supposed to be the equivalents to conversational maxims, and the listeners’ contextual mutual assumptions.

53This model presents the undeniable advantage of not requiring any effective legislators’ intention: however, this is exactly the main reason for its collapse.

  • 47 On this point it has been argued that in order to establish a hierarchy between legal interpretativ (...)

54Firstly, as repeatedly stated, there is not a unanimous consent about which legal criteria are to be considered as the analogues to conversational maxims or about the hierarchy between them: in the legal field, there is not that general expectation of conformity that founds the functioning of maxims in ordinary conversation.47

  • 48 See Asgeirsson 2017.

55Secondly, even if we limit the relevant context to the listeners’ contextual mutual assumptions, it is doubtful whether this is univocal, and sufficiently determined and rich.48

56Finally and above all, in statutory interpretation, the objective notion of implicatum does not find a counterbalance in any (univocal, determined) intention to implicate on the part of the legislators: i.e., a criterion to establish the success of the communication is missing. This is the reason why this model ends up being a sophisticated variant of the normative criterion of the rational legislator: this model does not describe what happens in statutory interpretation, but it prescribes to interpret the statutes as if they were the product of a cooperative speaker. But there is neither a cooperative speaker nor the model assumes that there is. Precisely because there is no parameter to establish communicative success, and because, in fact, there is no consensus either on the relevant context or on the interpretative criteria to be employed, each interpretative proposal has theoretically the same validity as the others. But is this not what actually happens in statutory interpretation? It happens in statutory interpretation, but not in ordinary conversation, and this shows that these are two profoundly different practices.

-Acknowledgments.- I thank Lorena Ramírez-Ludeña, Carmen Vázquez and all the organizers of the conference “En teoría hay mujeres (en teoría)”. I also thank all the members of the audience for discussion, and in particular Lucila Fernandez Alle, Isabel Lifante and José Juan Moreso. Finally, I want to thank the anonymous reviewers for their helpful criticisms, comments and corrections.

Top of page

Bibliography

Asgeirsson, H. (2017). On the Possibility of Non-literal Legislative Speech. In Poggi, F., & Capone, A. (Eds), Pragmatics and Law. Practical and Theoretical Perspectives (pp. 67–101). Cham, Switzerland: Springer.

Atlas, J. D. (2005). Logic, Meaning and Conversation. Oxford, England: Oxford University Press.

Atlas, J. D., & Levinson, S.C. (1981). It-Clefts, Informativeness, and Logical Form. Radical Pragmatics. In Cole, P. (Ed.), Radical Pragmatics (1–61). New York, NY: Academic Press.

Bach, K. (1997). The Semantic-Pragmatic Distinction. What It Is and Why It Matters. In Turner, K. (Ed.), The Semantics-Pragmatics Interface from Different Points of View (pp. 65–84). Amsterdam, the Netherlands: Elsevier.

Bach, K. (2005). The Top 10 Misconceptions about Implicature. In Birner, B.G., & Ward, G.L. (Eds.), Drawing the Boundaries of Meaning. Neo-Gricean Studies in Pragmatics and Semantics in Honor of Laurence R. Horn (pp. 21–30). Amsterdam, the Netherlands: John Benjamins.

Bratman M.E. (2014). Shared Agency: A Planning Theory of Acting Together. Oxford, England: Oxford University Press.

Canale, D. (2017). What Inferentialism Tells Us About Combinatory Vagueness in Law. In Poggi, F., & Capone, A. (Eds.), Pragmatics and Law. Practical and Theoretical Perspectives (pp. 43–66). Cham, Switzerland: Springer.

Canale, D., & Poggi, F. (2019). Pragmatic Aspects of Legislative Intent. American Journal of Jurisprudence, 64(1), 125–138.

Capone, A. (2013). The Role of Pragmatics in (Re)Constructing the Rational Law-Maker. Pragmatics & Cognition, 21(2), 399–414.

Carston, R. (2013). Legal Texts and Canons of Construction: A View from Current Pragmatic Theory. In Freeman, M., & Smith, F. (Eds.), Law and Language (pp. 8–33). Oxford, England: Oxford University Press.

Chiassoni, P. (1999). Interpretative Games: Statutory Construction through Gricean Eyes. Analisi e diritto 1999, 7999.

Dascal, M. & Wróblewski, J. (1991). The Rational Law-Maker and the Pragmatics of Legal Interpretation. Journal of Pragmatics, 15, 421–444.

Dworkin, R. (1986). Law’s Empire. Cambridge, Mass.: Harvard University Press.

Ekins, R. (2012). The nature of legislative intent. Oxford, England: Oxford University Press.

Gizbert-Studnicki, T. (2015). The Normativity of Rules of Interpretation. In Ed. Araszkiewicz, M., Banás, P., Gizbert-Studnicki, T., & Krzysztof Pleszka (Eds.). Problems of Normativity, Rules and Rule-Following (pp. 243–264). Dordrecht, the Netherlands: Springer.

Greenberg, M. (2011). Legislation as Communication? Legal Interpretation and the Study of Linguistic Communication. In Marmor, A., & Soames, S. (Eds.), Philosophical Foundations of the Language in the Law (pp. 217–264). Oxford, England: Oxford University Press.

Grice, P.H. (1989). Studies in the Way of Words. Cambridge, Mass.: Harvard University Press.

Kutz, C. (2000). Acting together. Philosophy and Phenomemological Research, 61(1), 1–31.

Levinson, S. (2000). Presumptive Meaning. Cambridge, Mass. & London, England: MIT Press.

Lifante Vidal, I. (1999). La interpretación jurídical en la teoría del derecho contemporanea. Madrid, Spain: Centro de Estudios Politicos y Constitucionales.

Macagno, F., Walton, D., & Sartor, G. (2018). Pragmatic Maxims and Presuppositions in Legal Interpretation. Law and Philosophy, 37, 69–115.

Malpas, J. (2002). The Weave of Meaning. Holism and Contextualism. Language & Communication, 22(4), 403–419.

Marmor, A. (2008). The Pragmatics of Legal Language. Ratio Juris, 21(4), 423–452.

Marmor, A. (2016). Defeasibility and Pragmatic Indeterminacy in Law. In Capone, A., & Poggi, F. (Eds.), Pragmatics and Law. Philosophical Perspectives (15–32). Cham, Switzerland: Springer.

Miller, G.P. (1990). Pragmatics and Maxims of Interpretation. Wisconsin Law Review, 1179–1227.

Morra, L. (2011). Implicature conversazionali nei testi di legge. Esercizi filosofici, 6(1), 214–231.

Morra, L. (2016). Widening the Gricean Picture to Strategic Exchanges. In Capone, A., & Poggi, F. (Eds.), Pragmatics and Law. Philosophical Perspectives (pp. 201–229). Cham, Switzerland: Springer.

Poggi, F. (2011). Law and Conversational Implicatures. International Journal for Semiotics of Law, 24(1), 21–40.

Poggi, F. (2013). The Myth of Literal Meaning in Legal Interpretation. Analisi e diritto 2013, 313-335.

Poggi, F. (2016). Grice, the Law, and the Linguistic Special Case Thesis. In Capone, A., & Poggi, F. (Eds.), Pragmatics and Law. Philosophical Perspectives (pp. 231–248). Cham, Switzerland: Springer.

Poggi, F. (2018). Conversational Implicatures and Legal Interpretation. On the Difference between Conversational Maxims and Legal Interpretative Criteria. Analisi & diritto, 18(2), 39–66.

Radin, M. (1930). Statutory Interpretation. Harvard Law Review, 43(6), 863–885.

Raz, J. (2009). Between Authority and Interpretation. Oxford, England: Oxford University Press.

Recanati, F. (1989). The pragmatic of what is said. Mind and Language, 4(4), 295-329.

Reimer, M.M. (1998). What Is Meant by ‘What Is Said’. A Reply to Cappelen and Lepore. Mind & Language, 13(4), 598–604.

Roversi, C. (2016). Intezionalità collettiva e realtà del diritto. In Bongiovanni, G., Pino, G. & Roversi, C. (Eds.), Che cosa è il diritto. Ontologie e concezioni del giuridico (pp. 255–294). Torino, Italy: Giappichelli.

Rysiew, P. (2007). Beyond Words: Communication, Truthfulness, and Understanding. Episteme, 4(3), 285–304.

Sbisà, M. (2007). Detto, non detto. Le forme della comunicazione implicita. Roma-Bari, Italy: Laterza.

Sbisà, M. (2017). Implicitness in Normative Texts. In Poggi, F., & Capone, A. (Eds.), Pragmatics and Law: Practical and Theoretical Perspectives (pp. 23–42). Cham, Switzerland: Springer.

Schweikard, D.P., & Schmid, H.B. (2013). Collective Intentionality. The Stanford Encyclopedia of Philosophy. Ed. Edward N. Zalta. URL = <https://plato.stanford.edu/archives/sum2013/entries/collective-intentionality/>.

Scott, A. (2011). Legal Interpretation: Taking Words Seriously. Claremont, CA: CMC senior thesis.

Skoczén, I. (2019). Implicatures Within Legal Language. Dordrecht, the Netherlands: Springer.

Sinclair, M.B.W. (1985). Law and Language. The Role of Pragmatics in Statutory Interpretation. University of Pittsburgh Law Review, 46, 373–420.

Slocum, B.C. (2016). Conversational Implicatures and Legal Texts. Ratio Juris, 29(1), 23–43.

Strawson, P.F. (1973). Austin and ‘Locutionary Meaning’. In Berlin, I. et al. (Eds.), Essays on J.L. Austin (pp. 46–68). Oxford, England: Clarendon Press.

Waldron, J. (1999). Law and Disagreement. Oxford, England: Oxford University Press.

Top of page

Notes

1 See Dworkin 1986: 50ss.; Scott 2011; Canale 2017: 45ss.

2 Grice 1989: 22-40.

3 The clarification ‘at least to a certain extent’ is indispensable in order to take account of cases involving the so-called exploitation or flouting of the maxims: cases in which a maxim is overtly violated in order to intentionally exploit the evidence of the violation and convey a certain message. In Grice’s opinion, irony, metaphors, meiosis and hyperbole fall into this category: Grice 1989: 53ff; Levinson 1983: chap. 3.

4 Grice 1989: 26.

5 Grice 1989: 30-1.

6 Grice 1989: 25.

7 Levinson 2000: 171. See also Reimer 1998: 598.

8 See Recanati 1989. The theoretical analysis of the processes of pragmatic enrichment (expansion and completion and the distinction between them) is very complex and will not be addressed here. To give an example, consider the sentence (f) ‘I did not have breakfast’. If today I utter (f), according to this view, what is said is not equivalent to ‘I [Francesca] did not have breakfast’, but to ‘I [Francesca] did not have breakfast today’.

9 See Strawson 1973; Bach 1997; Rysiew 2007.

10 See Malpas 2002.

11 Raz 2009: 274.

12 See Sinclair 1985; Miller 1990; Morra 2011; Ekins 2012; Carston 2013; Morra 2016; Slocum 2016; Sbisà 2017; Macagno, Walton & Sartor 2018; Skoczén 2019.

13 See Sinclair, 1985; Miller 1990; Slocum 2016.

14 See Chiassoni 1999; Poggi 2018.

15 Sinclair 1985.

16 Miller 1990.

17 More recently Slocum (2016) has argued that the interpretive criterion called eiusdem generis is a legal equivalent of a Gricean maxim or, better, of one of Levinson’s heuristics: see Levinson 2000. Similarly, Skoczén (2019) argues that there are two legal interpretative criteria that correspond to Gricean maxims: expression unius exclusion alterius and eiusdem generis. However, Skoczén's approach is much more sophisticated than that of the other authors: according to Skoczén, legal interpretation is a non-cooperative, strategic form of interaction, regulated by the principle ‘Assume that the speech is cooperative, infer all the implicatures and then apply the second-order strategic maxims to the inferred implications (anticipate which implications your hearers/speaker will choose/ignore)’. Briefly, Skoczén opines that, as far as legal interpretation is concerned, after a first phase compliant with the Gricean model, a second strategic phase (a set of second-order strategic maxims, summarized in the super-maxim ‘Pursue your goal through selection conforming implicatures’) comes into play.

18 See Canale & Poggi 2019, and, on the two last levels, Asgeirsson 2017. Notice that according to Grice the speaker’s intention, even if complex and reflective, is in fact a unitary intention. Therefore, the four levels distinguished in the text should not be confused with different kinds of intention. They simply identify the internal articulation of a communicative intention.

19 Grice 1989: 213 ff.

20 See Raz 2009: 284: “A person is legislating (voting for a Bill, etc) by expressing an intention that the text of the Bill on which he is voting will […] be law […] Legislators who have the minimal intention know that they are, if they carry the majority, making law”.

21 See below in the text and for a more detailed discussion Kutz 2000, and Ekins 2012. I thank my anonymous reviewers for this suggestion.

22 Surely this point is very discussed within post-Gricean and neo-Gricean literature: however, if we admit that, according to Grice's original account, the meaning implicated is the meaning that the speaker intends to implicate, then we have to admit that what is said is what the speaker intends to say, since what is implicated depends on what is said.

23 Greenberg 2011: 239.

24 On this point see also Raz 2009: chap. 11; Lifante 1999.

25 This is just one aspect of the more general problem of determining who are the subjects to be taken into account in order to establish the legislative intention: on that point see, for example, Radin 1930; Dworkin 1986: 314ff.; Marmor 2008.

26 Greenberg 2011: 253.

27 Surely this also happens in ordinary conversation. E.g., a speaker can utter an ambiguous or vague sentence without having any precise intention on how to solve the ambiguity or vagueness. Even in this case what is said is and remains ambiguous or vague. In Grice’s original approach, the speaker’s intention sets up what is said: if there is not such an intention then what is said is and remains indeterminate. My point is that this happens more frequently in law because there is not just one speaker here.

28 See Smith v. United States508 U.S. 223 (1993).

29 Actually, according to some authors, ‘to use a firearm in a way whatsoever’ is not the enriched meaning of ‘to use a firearm’ but its minimal semantic content. I will not discuss this point here.

30 This is the core of Bratman’s continuity thesis: ‘The conceptual, metaphysical, and normative structures central to such modest sociality are […] continuous with structures of individual planning agency’ (Bratman 2014: 8).

31 A detailed analysis of the theories on collective intention is beyond the scope of this essay, but I claim that the same hold for other theories of collective intentionality, such as Tuomela’s and Gilbert’s. On this point see Schweikard & Schmid 2013; Roversi 2016.

32 In Ekins’ more sophisticated terms: ‘Each legislator forms the intention ‘I intend that we legislate by means of the relevant set of procedures, in which majority vote for a complete statutory proposal on the third reading counts as the legislature enacting the proposal’. The intention interlocks with the intention of other legislators to similar effect, so that ‘I intend that we legislate …because of and in accordance with your intention that we legislate’. (Ekins 2012: 221).

33 This point is even clearer if one adopts Kutz’s minimalist account of collective action. According to Kutz ‘the core, minimalist, notion of collective action as such requires only that individuals act on overlapping participatory intentions’ (Kutz 2000: 4), and a participatory intention is ‘an intention to do my part of a collective act, where my part is defines as the task I ought to perform if we are to be successful in realizing a shared goal’ (Kutz 2000: 10). The shared goal surely can be identified with the enacting of a text of law – or, better, with the result of the legislative procedure. However, it cannot be identified with uttering what is said (viz. with enacting a certain statutory meaning) if the majority does not agree on the meaning of the statute. According to Grice’s original account, what is said is a function of the speaker’s intention: if the majority does not agree on what is said – if there is not a common intention to say x - what is said is not determined, and therefore there is not a common goal – but as much goals as the different intended meanings of the text. I thank my anonymous reviewer for pointing out Kutz’s theory.

34 On the indeterminacy and opacity of legal context see, e.g., Marmor 2008; Poggi 2013; Asgeirsson 2017; Canale & Poggi 2019.

35 See also Sperber & Wilson 1986: 63-4. I thank my anonymous reviewer for pointing out to me this reference.

36 See Levinson 2000; Atlas 2005; Sbisà 2007; Morra 2011; Poggi 2011; Morra 2016; Poggi 2016; Sbisà 2017.

37 For a discussion see, e.g., Poggi 2018.

38 See Atlas & Levinson 1981; Macagno, Walton & Sartor 2018.

39 Actually Greenberg speaks of ‘communicative content’, a notion that includes that of implicatum.

40 Greenberg 2011: 231. See also Bach 2005.

41 Greenberg 2011: 231.

42 Greenberg 2011: 248.

43 Dascal & Wróblewski 1991; Capone 2013; Macagno, Walton & Sartor 2018.

44 I think that this model can be labelled as listener-based because it leaves aside speakers’ actual communicative intention and focuses on ‘what a member of the audience would reasonably take a speaker who had uttered the relevant sentence under specified conditions to have intended to communicate’ (Greenberg 2011: 231).

45 See Levinson 2000.

46 See Poggi 2011; Poggi 2016. Maxims are hermeneutic technical rules in the sense that they ensure mutual understanding. More exactly, they are rules that establish the means trough which we can rich mutual understanding. So, according to Poggi (2016), the CP and conversational maxims then can be formulated in the following way: (TR) ‘If you want to cooperate and you want to understand and be understood, then you must follow the CP, that is m, m1 ,…’ where ‘m, m1 ,…’ stands for the contents of each maxim.

47 On this point it has been argued that in order to establish a hierarchy between legal interpretative criteria it is necessary to appeal not to the existence of a generalized expectation (as is the case with conversational maxims), but rather to normative and political arguments: see Greenberg 2011; Gizbert-Studnicki 2015.

48 See Asgeirsson 2017.

Top of page

References

Bibliographical reference

Francesca Poggi, Against the conversational model of legal interpretationRevus, 40 | 2020, 9-26.

Electronic reference

Francesca Poggi, Against the conversational model of legal interpretationRevus [Online], 40 | 2020, Online since 10 June 2020, connection on 29 July 2021. URL: http://journals.openedition.org/revus/5694; DOI: https://doi.org/10.4000/revus.5694

Top of page

About the author

Francesca Poggi

Associate Professor of Law, University of Milan (Italy)

Address: University of Milan Department of Law – Via Festa del Perdono, 3 – 20122 Milan – Italy

E-mail: francesca.poggi@unimi.it

By this author

Top of page

Copyright

All rights reserved

Top of page
  • OpenEdition Journals
Search OpenEdition Search

You will be redirected to OpenEdition Search