Abstract
In his “Meaning and Formal Semantics in Generative Grammar” (Erkenntnis 80(1):61–87, 2015), Stephen Schiffer argues that truth-conditional semantics is a poor fit with generative linguistics. In particular, he thinks that it fails to explain speakers’ abilities to understand the sentences of their language. In its place, he recommends his “Best Bet Theory”—a theory which aims to directly explain speakers’ abilities to mean things by their utterances and know what others mean by their utterances. I argue that Schiffer does not provide good reason to prefer the Best Bet Theory over truth-conditional semantics in the context of generative linguistics. First, his negative arguments against the truth-conditional approach are unpersuasive, and second, the Best Bet Theory involves an explanatory circularity which makes it unfit for linguistic theorizing. I conclude that the Best Bet Theory is thus not even a viable competitor to truth-conditional semantics in generative linguistics.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
As Schiffer makes clear, ‘truth-conditional semantics’ is meant to be understood broadly, encompassing both the Montagovian, model-theoretic approach, and the Davidsonian approach. I will tend to use ‘truth-conditional semantics’ the same way, ignoring the distinctions between the two approaches.
I place ‘tacit knowledge’ in quotation marks just to flag that it is an open empirical question just what this relation comes to. (GGH) is not meant to saddle generative linguistics with any substantial epistemological commitments concerning semantic competence. For detailed presentations of the generative linguistics framework, see Chomsky (1986) and Ludlow (2011).
Thus, I think this is an instance of what Chomsky (1995) calls ‘methodological dualism’: placing constraints on theories of mind and language that everyone agrees are out of place in other naturalistic research programs.
See Yalcin (2014) for relevant discussion.
The ongoing inquiry into the semantics/pragmatics distinction, e.g., suggests that it is not an easy task to say just which phenomena are semantic and which are pragmatic, for instance.
The same sorts of considerations apply to the phrases “linguistic competence” and “semantic competence”.
At least, they do not provide constraints of the sort that would adjudicate between, e.g., truth-conditional semantics and the BBT. They would, of course, rule out geological inquiries, for instance, as being semantic.
That truth-conditional semantics does not explain our communicative abilities might be taken as evidence that the latter are explained by appeal to mechanisms which are not the truth-conditional theorist’s target.
For convenience, I ignore the fact that various dialects of languages will differ from one another in theoretically important ways.
The argument comes from Schiffer (2015, 75).
See, e.g., Matthewson (2004) for a general discussion of semantic methodology, including techniques for distinguishing entailments from implicatures utilizing speakers’ judgments of truth value and infelicity and other indirect sources of evidence. See also Ludlow (2011, Ch. 3) for an overview and defense of the role of speaker judgments in semantic theorizing. Thanks to an anonymous referee for suggesting that I include discussion on semantic methodology.
Schiffer does not endorse the Russellian account of propositions, but adopts it for ease of exposition (2015, 80). The differences between his own view of propositions (2003, Ch. 2) and the Russellian view will not be relevant to my discussion, and so I will treat the BBT as making use of Russellian propositions.
There are linguists (e.g., Speas and Tenny 2003; Rizzi 1997) who argue that information about the speech act type is encoded in syntax. In principle, there is no reason why a truth-conditional semanticist could not accept this view. However, they would still differ from the BBT theorist with respect to the meanings that they assign to the syntactic representation of a sentence. Thanks to an anonymous referee for raising this point.
It is a well-known problem that knowledge (tacit or otherwise) of truth-conditions does not suffice for knowledge of meaning. See, e.g., Foster (2010) and Soames (1992). See Larson & Segal (1995), Heck (2007), and Kölbel (2001) for proposals to the effect that tacit knowledge of truth-conditions, in conjunction with certain interpretive dispositions, might suffice to be able to know what utterances mean. Other truth-conditional theorists will simply acknowledge that for some classes of expressions, meanings need to be understood in a way that is more fine-grained than truth-conditions.
Similar quotes can be found throughout section 3 of Schiffer (2015).
See also Schiffer (2003, 112), where a sentence’s having its meaning is a matter of it having its character*, where a character* is represented “by an ordered pair \({<}A, P{>}\), where A is the kind of speech act that must be performed in a literal utterance of the sentence, and P is the kind of propositional content the speech act must have.”
Consider for example, utterances like “The Broncos destroyed the Panthers”, which would typically be used to mean, roughly, that the Broncos defeated the Panthers in the contextually salient game by a wide margin. Neither the speaker nor hearer will seem to register at any level that the utterance also means that the Broncos organization literally destroyed the Panthers organization. Since in this context we are concerned with a kid of psychological theorizing, we can’t simply stipulate that the utterance also has the literal meaning.
This is not to deny that there will be the further question, “In virtue of what do expressions have the truth-conditional contents they have?” The point is that—except when we are providing a compositional explanation for the contents of complex expressions—this is a question which goes beyond semantic theorizing proper.
Schiffer also cites Heim and Kratzer (1998, 1), who claim that “to know the meaning of a sentence is to know its truth-conditions,” and Partee (1988, 49), who writes that “the real argument for compositional truth-theoretic semantics is not that language users can understand indefinitely many novel utterances, but rather that there are indefinitely many semantic facts to be explained, indefinitely many pieces of basic data about truth-conditions and entailment relations among sentences of any given natural language.”
The semantic content of an utterance of S is, roughly, any proposition communicated by an utterance of S which has the form of \(\Psi \), where \(\Psi \) is the propositional form associated with S by its meaning (81–82). We can therefore think of the literal utterances of S, generally, as the utterances of S where only contents of form \(\Psi \) are expressed.
As I argue in my (Napoletano 2015), something’s having a particular property, in general, does not consist in its bearing a relation to things with that very property. In this case, a sentence S’s having a particular meaning M is (roughly) taken to be a matter of standing in a certain relation to uses of S with that very meaning M.
Note that we might have good reason to think that someone speaks sincerely without knowing what they or the uttered sentence meant, but the sincere/insincere distinction does not track the literal/non-literal distinction.
See Heck (2007) for a similar kind of proposal.
It also seems as though simple expressions can have literal or figurative uses. For instance, ‘point’ in ‘I see your point’. It’s hard to see how we could understand the meanings of simple expressions except by reference to their contributions to the sentences they appear in.
See, e.g., Alston (1999).
In non-literal utterances, the hearer will drop the presupposition that the context is a usual one (on the basis of various sorts of contextual evidence), and then arrive at the correct speaker-meaning (often enough, hopefully) via pragmatic processes.
I don’t think the same sort of adjustment could be made for the proposal in Sect. 4.2. In that case, we would suppose that speakers tacitly assume that utterances of S mean the explanatorily basic use of S. But it seems as though the notion of explanatorily basic use is too theoretical to plausibly suppose that ordinary speakers have even a tacit conception of it.
Even if one thinks that scalar implicatures are semantic phenomena, it wouldn’t be difficult to come up with other examples that are even less plausibly semantic, or to imagine a human linguistic community which, most of the time, employed paradigmatically pragmatic processes with respect to uses of certain sentences of their language (e.g., metaphor).
One other desperate strategy might be to just abolish the liter/non-literal distinction, so that every use would, in effect, be treated as “literal”. The result would be the positing of massive and widespread ambiguity. In that case, it wouldn’t be the BBT that would explain our communicative abilities, but whatever it was that allowed us to disambiguate properly.
I abstract away from details here, like the fact that the truth-conditional theory might not pair sentences with truth-conditions, but something sub-propositional.
See, e.g., Pietroski (2003) for a criticism of the viability of truth-conditional semantics in generative linguistics.
References
Alston, W. (1999). Illocutionary acts and sentence meaning. Ithaca: Cornell University Press.
Chomsky, N. (1986). Knowledge of language. New York: Prager.
Chomsky, N. (1995). Language and nature. Mind, 104(413), 1–61.
Foster, J. (2010). Meaning and truth theory. In D. Byrne & M. Kólbel (Eds.), Arguing about language. London: Routledge.
Heck, R. (2007). Meaning and truth-conditions. In D. Greimann & G. Siegwart (Eds.), Truth and speech acts: Studies in the philosophy of language (pp. 349–376). London: Routledge.
Heim, I., & Kratzer, A. (1998). Semantics in generative grammar. Blackwell: Wiley.
Kölbel, M. (2001). Two dogmas of Davidsonian semantics. Journal of Philosophy, 98(12), 613–635.
Larson, R., & Segal, G. (1995). Knowledge of meaning. Cambridge: MIT Press.
Ludlow, P. (2011). The philosophy of generative linguistics. Oxford: Oxford University Press.
Matthewson, L. (2004). On the methodology of semantic fieldwork. Int J Am Linguist, 70(4), 369–415.
Napoletano, T. (2015). Compositionality as weak supervenience. Synthese, 192(1), 201–220.
Partee, B. (1988). Semantic facts and psychological facts. Mind and Language, 3(1), 43–52.
Partee, B. H. (2011). Formal semantics: Origins, issues, early impact. The Baltic International Yearbook of Cognition, Logic and Communication, 6, 1–52.
Pietroski, P. (2003). The character of natural language semantics. In A. Barber (Ed.), Epistemology of language (pp. 217–256). Oxford: Oxford University Press.
Rizzi, L. (1997). The fine structure of the left-periphery. In L. Haegeman (Ed.), Elements of grammar: A handbook of generative syntax (pp. 281–337). Dordrecht: Kluwer.
Schiffer, S. R. (2003). The things we mean. Oxford: Oxford University Press.
Schiffer, S. (2015). Meaning and formal semantics in generative grammar. Erkenntnis, 80(1), 61–87.
Speas, P., & Tenny, C. (2003). Configurational properties of point of view roles. In A. DiSciullo (Ed.), Asymmetry in grammar (pp. 315–344). Amsterdam: John Benjamins.
Soames, S. (1992). Truth, meaning, and understanding. Philosophical Studies, 65(1–2), 17–35.
Yalcin, S. (2014). Semantics and metasemantics in the context of generative grammar. In A. Burgess & B. Sherman (Eds.), Metasemantics (pp. 17–54). Oxford: Oxford University Press.
Acknowledgments
Many thanks to Lionel Shapiro and an anonymous reviewer for extensive comments on earlier drafts. Any errors that remain in the paper are my own.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Napoletano, T. Why Truth-Conditional Semantics in Generative Linguistics is Still the Better Bet. Erkenn 82, 673–692 (2017). https://doi.org/10.1007/s10670-016-9838-2
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10670-016-9838-2