(Updated January 14, 2023)
- These are the comments I gave at Yale in June 2022 on Michael della Rocca’s paper ‘The Original Sin of Analytic Philosophy’.
- These are the comments I gave at Syracuse in August 2011 on Mike Nelson’s paper ‘Time and Person in Thought’.
- These are the comments I gave at Michigan in November 2008 on Franke, Jager, and van Rooij ’s paper on game-theoretic pragmatics.
- These are the comments I gave at Ohio State in October 2006 on Kai von Fintel’s paper on presupposition accommodation.
- These are the comments I gave at Rutgers in September 2005 on John MacFarlene’s paper on non-indexical contextualism.
- These are the comments I gave at Michigan in November 2004 on Pauline Jacobson’s paper on direct compositionality.
Published papers (penultimate drafts; for citation, please consult the published version):
This is a review of Barbara Abbott’s book Reference. It came out in Lingua (2011) 121: 1859 – 1861.
In this paper, I argue that although the behavior of adjectives in context poses a serious challenge to the principle of compositionality of content, in the end such considerations do not defeat the principle. The first two sections are devoted to the precise statement of the challenge; the rest of the paper presents a semantic analysis of a large class of adjectives that provides a satisfactory answer to it. In section 1, I formulate the context thesis, according to which the content of a complex expression depends on the context of its utterance only insofar as the contents of its constituents do. If the context thesis is false, the content of some complex expression is not compositionally determined. In section 2, using an example due to Charles Travis, I construct an objection to the context thesis based on the behavior of the adjective ‘green’. In sections 3 and 4, I look at some of the difficulties surrounding the semantics of ‘good’, which provide the motivation for the thesis that most adjectives are contextually incomplete one-place predicates. In section 5, I discuss how ‘green’ and other color adjectives can be treated within such a semantic theory. Since this theory is compatible with the context thesis, the objection against the compositionality of content looses its force. The paper came out in a Festschrift for the Hungarian linguist Ferenc Kiefer, entitled Perspectives on Semantics, Pragmatics, and Discourse (I. Kenesei and R. M. Harnish eds.; Amsterdam: John Benjamins, 2001).
According to the traditional doctrine of logical form, sentences have an underlying structure which lays bare their inferential profiles. Logical form is supposed to be inherent in the sentence (not ascribed to it as a result of formalization) and it is supposed to
capture its logical (not merely syntactic or semantic) features. The paper argues against the existence of logical form in this sense by casting doubt on the coherence of the idea that certain inferences are valid solely in virtue of their form.
And while the idea that some inferences are valid in virtue of their form and facts of logic is perfectly coherent, these inferences are arguably unavailable in natural languages. Moreover, appeal to semantic competence is no help because logical competence
is distinct from our knowledge of language. The upshot of the argument is that we should abandon the Davidsonean hope that an adequate compositional semantics could explain why logical validities are valid or how we know that they are. The paper appeared in 2012 in G. Preyer ed., Donald Davidson on Truth, Meaning, and the Mental. Oxford: Oxford University Press, pp. 105 – 126.
Frege famously gets himself into a contradiction regarding the question whether the referent of ‘is a horse’ is a function. One of the aims of this paper is to see whether any of the assumptions that lead to this contradiction can be dropped without risking the integrity the Fregean program in natural language semantics. I will argue that the answer is negative: to avoid the problem we must embark on a radical revision. I think we should abandon the singularist conception of semantics, which demands that we interpret linguistic expressions via assignment of a unique semantic value. There is an alternative, going back to medieval logicians of nominalist leaning, according to which predicates signify multiple particulars rather than a single universal and syncategoremata signify nothing at all. I will try to make the case that natural language semantics could and perhaps should return this tradition. The Hungarian version appeared in Általános Nyelvészeti Tanulmányok (2014) 26: 241 – 68. For those of you whose Hungarian is rusty, here is an English draft: Reference and the concept horse.
Let’s say that a quantifier occurs bare within a sentence just in case its domain is not constrained by any predicate within the sentence. A quantifier is bare just in case some of its occurrences are bare. I argue that there are perfectly ordinary sentences of English that express bare quantification. I begin by clarifying the notion of bare quantification and explaining why the presence of bare quantifiers in natural languages is far from obvious. Then I show how non-extensional scope-bearing expressions can be used to identify bear readings for many quantified sentences. The paper was published in the Philosophical Review (2011) 120: 247 –283.
This paper is about ontological commitment. I argue that being ontologically committed to – i.e. believing in F’s – is more than believing that there are F’s. If F’s exist but the conception of F’s (the conjunction of propositions whose acceptance constitutes expert competence with the term) is false, then someone who believes that F’s exist would be right while someone who believes in F’s would be wrong. This opens up the possibility to express ontological reservations without challenging the soundness of cheap arguments to the existence of those entities. This paper came out in Philosophy and Phenomenological Research (2003) 66: 584 – 611.
This is a brief entry for The Dictionary of American Philosophers.
Available at the Stanford Encyclopedia of Philosophy.
I argue that the best way to capture the intuitive content of the principle of compositionality is to construe it as the following strong supervenience claim: For all possible human languages L, for any meaning property M and any complex expression e in L, if e has M in L, then there is a constitution property C such that e has C in L, and for any possible human language L’ if any complex expression e’ in L’ has C in L’ then e’ has M in L’. If it is accepted that the principle of compositionality is an intricate thesis about the nature of possible languages, one will be less inclined to think of it as a triviality or as a mere methodological assumption. It will more likely to be taken as a significant, though extremely general, empirical assumption. This paper appeared in Linguistics and Philosophy (2000) 23: 475 – 505.
I present two challenges to fictionalism. According to the first, the reasons fictionalists offer for acceptance without belief often warrant a somewhat different attitude. According to the second, the possibility of fictionalist acceptance rests on the poorly supported hypothesis that there is a clear distinction between philosophical and ordinary contexts. This appeared in Noûs (2011) 45: 375 – 85.
A defense of presentism against what I call semantic arguments. Most of the discussion focuses on a response to the argument from quantification, recently advanced by David Lewis in his posthumous paper Tensed Quantifiers. It appeared in Philosophical Perspectives 20: Metaphysics. (2007): 399 – 426.
I argue for a simple unitary semantics for singular definite and indefinite descriptions: they all contribute existential quantification to the logical form of sentences in which they occur in extensional contexts. Uniqueness implications of definite descriptions (as well as non-uniqueness implications of indefinites) arise as a result of reasoning about what the mental file of our conversational partner might contain. I claim that the best alternative to my views are unitary Russellian truth-conditions and argue that my account is preferable to such a view on both empirical and methodological grounds. This paper was published in Philosophical Studies (2000) 101: 29 – 57.
This is a rejoinder to Barbara Abbott’s reply to my ‘Descriptions and Uniqueness’. Abbott’ paper is in Philosophical Studies (2003) 103: 223 – 231; my rejoinder is in Philosophical Studies (2003) 104: 279 – 291. I present some considerations that provide further support for my view.
Semantic theories usually employ a triadic notion of truth: they specify the conditions under which a sentence is true in a context c at an index i. Yet philosophical orthodoxy holds that truth is absolute. The compromise is forged by two claims. First, that truth belongs primarily to propositions and the reason sentences can have different truth-values in different contexts is that they express different propositions in them. So, propositional truth does not depend on context. Second, talk of truth at an index should be paraphrased away: a proposition it true at a time iff it is, was or will be true when that time is present and it is true at a possible world iff it would be true if that world were actual. So, propositional truth does not depend on index either. Following Austin, I suggest that the fundamental notion of truth is neither triadic nor monadic. Truth is dyadic: to be true is to be true of a situation. The paper was published in a festschrift for László Kálmán and András Kornai and is available here: http://clara.nytud.hu/~kk120/
This is an entry, co-authored with Jason Stanley, for the new edition of the International Encyclopedia of Linguistics. This piece is supposed to present the issues briefly and neutrally. For a more detailed and opinionated piece see ‘On Quantifier Domain Restriction.’ (with Jason Stanley). Mind and Language (2000) 15: 219-61.
Epistemic Comparativism (with Jonathan Schaffer):
We articulate and defend an explicit contextualist semantics for ‘know’. Essentially our semantics integrates a conception of knowledge—fairly orthodox among epistemic contextualists—as requiring the elimination of the relevant alternatives, with a treatment of adverbial quantifiers (A-quantifiers)—fairly orthodox among semanticists—as quantification over a contextually variable domain of situations. We treat ‘know’ as having an A-quantificational aspect. We call this a comparativist semantics, since it treats ‘know’ as comparing a contextually constrained domain proposition with an explicit scopal proposition. It was published in Philosophical Studies (2014) 168: 491 – 543.
The main thesis of this paper is that linguistic tokens are representations of their types. This view is defended against the traditional view, according to which the relation that linguistic tokens bear to their types is that of instantiation. The main argument against the traditional view is the inverted word argument. Consider the case of seventeen-year-old Karel who recently began to study English. In his book, due to a series of unfortunate misprints, there is the following chart:
13 : thirty, 14 : forty, 15 : fifty, etc.; 30 : thirteen, 40 : fourteen, 50 : fifteen, etc.
He interprets this chart as any of us would and consequently comes to believe that ‘seventeen’ refers to 70, and that ‘seventy’ refers to 17. The mistake is perfectly manifest in his writing. But it is not detectable when he speaks because in pronunciation he makes the opposite mistake; he pronounces ‘seventy’ as a normal English speaker would pronounce ‘seventeen’ and he pronounces ‘seventeen’ as a normal English speaker would pronounce ‘seventy’. Suppose one day the English teacher asks Karel how old he is, and as a response Karel makes a linguistic utterance which sounds like a normal English utterance of “I am seventeen years old.” Intuitively, Karel does not know what the token numeral refers to in his utterance, but – unlike the representationalist view – the instantiation view has a hard time explaining why. This paper appeared in Philosophical Quarterly (1999) 49: 145 – 63.
Many philosophers strive for a thin ontology but are nevertheless unwilling to curtail ordinary and scientific talk that carries apparent commitment to the entities they reject. As Carnap put it, such a philosopher speaks with an uneasy conscience, “like a man who in his everyday life does with qualms many things which are not in accord with the high moral principles he professes on Sundays.” To appear less hypocritical he may, of course, tell us openly what he is doing and invite us to join him. But then it is hard to see why he is not advocating the absurd position that we should assent to sentences of the form ‘There are F’s but I don’t believe that there are F’s.’ This is a simple objection, and there is a simple answer to it. But the answer is not available to everyone. I will argue that defenders of a particular version of fictionalism are in trouble with Moore’s paradox. The bad news for fictionalism in general is that this version is the one that best deals with the Quine-Putnam challenge. This paper appeared in Canadian Journal of Philosophy (2001) 31: 293 – 308.
In my commentary on Stephen Yablo’s Aboutness, I try to do two things: present Yablo’s view on subject matter as a way of fixing a problem within the classic Fregean theory of meaning, and then argue that the proposal needs one additional repair. This appeared in Philosophical Studies (2017) 174: 779 – 86.
This is a review of The Compositionality Papers by Jerry Fodor and Ernie Lepore. It appeared in Mind (2004) 113: 340 – 344.
This is part of a symposium on Ernest Lepore and Matthew Stone’s Imagination and Convention. I argue that even though Lepore and Stone are successful in arguing that the scope of conversational implicatures is narrower than Grice thought, they are mistaken in thinking that the entire phenomenon is illusory. The paper is in Inquiry (2016) 59: 163 – 74.
This is my contribution to a forthcoming PPR symposium on Paul Pietroski’s Conjoining Meanings.
This is the introduction to a collection of essays on the distinction between semantics and pragmatics I am editing. The volume appeared in 2005, it was published by Oxford University Press.
TMy contribution to a special edition of Disputatio devoted to logical form. It was the basis of my comments on Andrea Iacona’s Logical Form at the Pacific APA in Vancouver.
It is an old charge against Locke that his commitment to a common substratum for the observable qualities of particular objects and his empiricist theory about the origin of ideas are inconsistent with one another. How could we have an idea of something in which observable qualities inhere if all our ideas are constructed from ideas of observable qualities? In this paper, I propose an interpretation of the crucial passages in Locke, according to which the idea of substratum is formed through an elaborate mental process which he calls supposition. It is the same process we use when we form the idea of infinity – another problematic idea for an empiricist. In the end, Locke was more liberal than most empiricists in subscribing to the existence of ideas far removed from experience, because he accepted supposition as a legitimate way of constructing new ideas. The paper came out in The Locke Newsletter (2000) 31: 11 – 42.
This is a review of Ofra Magidor’s book Category Mistakes. It appeared in Philosophical Review (2015) 124: 289 – 92.
According to the contemporary consensus, when reaching in the lexicon grammar looks for nouns, verbs, adjectives and prepositions, while logic sees predicates, connectives, quantifiers and individual constants. In fact, there doesn’t seems to be a single lexical category contemporary grammar and logic both make use of. I hope to show that this sort of radical dissociation between grammar and logic is unnecessary and ill-advised. The view I defend is this. Nouns, verbs, adjectives and adverbs are the broadest open lexical categories. Nouns and verbs are constants – their interpretation is insensitive to linguistic or extra-linguistic context. Adjectives and adverbs are variables – they perform their semantic function only when context supplements a way in which they apply. The semantic function of nouns and adjectives is reference – to call things something or other. Verbs and adverbs have a different job – they are satisfied by things. The paper was published in Erkenntnis (2015) 80: 3 – 29.
Modals with the Taste of the Deontic (with Joshua Knobe):
The aim of this paper is to present an explanation for the impact of normative considerations on people’s assessment of certain seemingly purely descriptive matters. A number of experiments in the last few years have shown that people’s judgments about whether an action was free or forced, whether it caused a certain outcome, and whether was performed intentionally often depend on whether the action violates a salient norm. The explanation we provided is unified and charitable: we argued that there is a common core of the phenomenon and that these judgments are not in error.At the heart of our account is a claim about modality. We argue that each of the judgments that show this surprising impact of normativity is modal and the modality in question, though primarily circumstantial or teleological, nonetheless shows a taste of the deontic. The paper came out in Semantics & Pragmatics (2013) 6: 1 – 42.
This is a general survey about nominalism in metaphysics. It differs from some others in spending more time on general questions regarding ontological commitment. Although I tried to be evenhanded, my anti-nominalist biases no doubt shine through. It appeared in M. J. Loux and D. Zimmerman eds., The Oxford Handbook of Metaphysics. Oxford: Oxford University Press, 2003, pp. 11 – 45.
This is a paper on the semantics of as-phrases. I argue that sentences like ‘John as a judge earns (exactly) $50,000’ involve two-layered predication where the qualified predicate (here: ‘earns (exactly) $50,000’) is ascribed to the subject insofar as the qualifying predicate (here: ‘judge’) is. Predication under a description is given a semantics within a neo-Davidsonean framework. I also argue that the analysis can shed light on the way qualification is used i n discussing questions of sameness and difference in metaphysics. The paper appeared in Philosophical Perspectives 17: Language and Philosophical Linguistics. (2003): 409 – 438.
On Quantifier Domain Restriction (with Jason Stanley):
In this the we survey the various possibilities for handling quantifier domains and argue for a specific approach. The paper was published in Mind and Language (2000): 219 – 261.
This is a paper about the semantics of the progressive and the perfective. The key thesis is that even though the project of trying to analyze the former in terms of the latter fails, the reverse analysis can succeed. The paper appeared in Noûs (2004) 38: 29 – 59.
This is a review of Barry Schein’s ‘And’: Conjunction Reduction Redux. It appeared in the Journal of Philosophy (2019) 141: 119 – 24.
This is a review of Scott Soames’s Philosophy of Language. It came out in the Notre Dame Review of Philosophy (http://ndpr.nd.edu/review.cfm?id=22711).
This paper is a defense of substantive explanations in semantics. I begin by offering a diagnosis of why the view that semantic theories are merely descriptive has been widely accepted in philosophy and I suggest that these grounds are not compelling. Then I argue that semantic explanations don’t have a uniform direction – upwards or downwards the syntactic tree. There is an explanatory division within the lexicon: the meanings of content words (nouns, verbs, adjectives, and adverbs) are semantically fundamental, while the meanings of function words (auxiliaries, connectives, copulas, derivational morphemes, determiners, expletives, prepositions, etc.) are derivative. In came out in E. Lepore and D. Sosa ed., Oxford Studies in the Philosophy of Language. Oxford: Oxford University Press, pp. 240 – 275, 2019.
A short critical piece on Cappelen and Lepore’s Insensitive Semantics. I defend a moderate version of contextualism against their arguments. It appeared (along with their response) in Mind and Language (2006) 21: 31 – 38.
Sententialism is the view that intensional positions in natural languages occur within clausal complements only. According to proponents of this view, intensional transitive verbs – such as ‘want’, ‘seek’, or ‘resemble’ – are actually propositional attitude verbs in disguise. I argue that ‘conceive’ (and a few other verbs) cannot fit this mold – conceiving-of is not reducible to conceiving-that. The path of the argument is somewhat unusual. I offer a new analysis of where Berkeley’s Master Argument goes astray, analyzing what exactly is odd about saying that Hylas conceives a tree which in not conceived. It turns out that a sententialist semantics cannot adequately account for the source of absurdity in attitude ascriptions of this type; to do that, we need to acknowledge irreducibly non-propositional (but nonetheless de dicto) conceiving. The paper appeared in Philosophical Quarterly (2005) 55: 462 – 474.
In her dissertation, Janet Fodor has argued that the quantificational force and the intensional status of certain quantifier phrases can be evaluated independently. The proposal was only halfway accepted: the existence of non-specific transparent readings is well-established today, but specific opaque readings are deemed illusory. I argue that they are real and outline a semantic framework that can generate them. The idea is to permit two types of quantifier raising: one that carries the restrictor of the determiner along and another that does not. When the second is applied the restrictor can be stranded within the scope of an intensional operator as the quantificational determiner itself takes wider scope. This has appeared in the proceedings of the 17th Amsterdam Colloquium, where it was delivered as one of the invited talks.
Part of a symposium on Wayne Davis’s Meaning, Expression, and Thought. I argue that there is no good reason to believe in structural conventions, i.e. conventions attaching to syntactic rules. The paper appeared in Philosophical Studies (2008) 137: 399 – 408.
Can we count the primes? There is a near unanimous consensus that in principle we can. I believe the near-consensus rests on a mistake: we tend to confuse counting the primes with counting each primeTo count the primes, I suggest, is to come up with an answer to the question “How many primes are there?” as a result of counting each prime. This, in turn requires some sort of dependence, presumably a causal one, of outcome on process. I argue that – barring very odd laws of nature – such dependence cannot obtain. This has appeared in a special English language issue of Magyar Filozófiai Szemle.
I argued for three main claims. The first one is that the usual statement of the compositionality principle is massively ambiguous. One of the eight available readings rules out all sources of multiplicity in meaning in complex expressions besides the lexicon and the syntax. Others are more permissive – how much more is not always clear. The second claim is that traditional considerations in favor of compositionality are less powerful than is often assumed: the intrinsicness argument and the inductive argument are inconclusive; the two arguments based on facts of language understanding have a number of shortcomings. In the end, compositionality is best construed as an empirical hypothesis on meanings expressed in natural languages. Finally, the third claim is that even if compositionality is true, most of the debates in philosophy, linguistics, and psychology will remain open. These debates tend to be about theses that are significantly stronger than compositionality. This has appeared in M. Werning, W. Hinzen, and E. Machery eds., The Oxford Handbook of Compositionality. Oxford: Oxford University Press.
I identify a notion of compositionality at the intersection of the different notions philosophers, linguists, and psychologists are concerned with. The notion is compositionality of expression content: the idea that the content of a complex expression in a context of its utterance is determined by its syntactic structure and the contents of its constituents in the contexts of their respective utterances. Traditional arguments from productivity and systematicity cannot establish that the contents of linguistic expressions are compositionally determined in this sense. I present a novel argument for this thesis based on plausible premises about literal use and a detailed defense of the compositionality of speech-act content. This is in Philosophical Studies (2010) 148: 253 – 272.
This is a general survey about the way the distinction between semantics and pragmatics has been drawn in the literature and a proposal to draw it slightly differently. It is in E. Lepore and B. Smith eds., The Oxford Handbook of Philosophy of Language. Oxford: Oxford University Press.
I begin by briefly reviewing the case against the uniqueness entailment Russell’s theory associates with singular definite descriptions. My main concern here, however, is not so much whether the case stands, but why it matters whether it does. Do we lose anything of substance from Russell’s own insights or the subsequent use to which they were put if we drop the uniqueness clause from his analysis of definite descriptions? Russell himself would not be bothered by this argument: he did not believe that an empirically adequate semantics is possible for natural languages and he was not interested in devising theories that are merely close to being adequate. If despite Russell’s contrary intentions, we are determined to view his theory of descriptions as part of the semantics of English, we need to settle what components of the original view we should hold on to. I argue that the current focus on the particular truth-conditions he gave is misguided – the semantic explanations in On Denoting put almost no constraint on what the truth-conditional content of the definite article might be. We can drop uniqueness and remain true Russellians about descriptions, if we wish. But we should not. In the final section I argue that the Russell-inspired view that descriptions could in principle be eliminated from a language that is equipped with standard quantifiers and the identity predicate is mistaken. Whatever descriptions are, they are not mere devices of quantification. The paper appeared in Mind (2006) 114: 1185 – 1222.
This paper is about ontological commitment. I argue that we can abstract away from controversial features of Quine’s proposal and come up with an attractive core conception of what it is for someone to be ontologically committed to certain entities. I use this conception to identify three types of approaches towards reconciling the received ontology with its philosophical rivals, and I argue that the least popular of the three – the attitude-based approach – has the best chance of delivering a satisfactory reconciliation. I argue that the standard version of the attitude-based reconciliation – fictionalism – faces serious difficulties. I propose and defend my own version of the attitude-based reconciliation, and argue that it delivers all the goods the fictionalist approach can without its shortcomings. I close with a brief discussion of a possible historical precedent for the type of view I advocate. This appeared in The Analytic Way, a book containing the papers form the 6th European Conference of Analytic Philosophy. A version of it was delivered there as the keynote address.
I argue that sentences like ‘John is building a house’ entail the existence of some thing John is building, although they do not entail that this thing is a house. It is a house in progress. On the way, I argue against intensional analyses of the progressive. This is a follow-up (and to some extent, a correction) of my earlier paper ‘On the Progressive and the Perfective.’ The paper appeared in Philosophical Perspectives (2008) 22: 499 – 525.
Since Frege’s writings it has become fairly standard to downplay the semantic significance of traditional lexical categories. Verbs, in particular, are typically assigned the semantic type of predicates, leaving intransitive verbs indistinguishable from common nouns and adjectives. At the same time, intransitive and transitive verbs are assigned different semantic types despite their robust syntactic similarity. The paper argues that this is a mistake: we get a better theory if we keep our syntactic categories and semantic types aligned. One way to do so is to deny that anything but a verb can be a predicate (unless it is turned into a verb first, as in ‘is green’ or ‘becomes a professor’) and to treat all verbs as one-place predicates of eventualities (which relate to their syntactic arguments via thematic roles). This came out in In D. Fara and G. Russell eds., The Routledge Companion to the Philosophy of Language. London: Routledge.