Topic: HUMAN SEMANTICS

The Telicity Parameter Revisited

By Hana Filip

The goals of this paper are threefold:
First, to review some recent syntactic accounts of cross-linguistic differences in the expression of telicity in Slavic vs. Germanic languages.
Second, I will argue that the parametric variation in the encoding of telicity cannot be based on a unidirectional specifier head agreement between the verbal functional head linked to the telicity of the VP and the DO-DP in its specifier position, with languages exhibiting two clearly distinct modes of assigning telicity to the functional head. In the simplest terms, in Germanic languages, it is assigned by the DO-DP and in Slavic by the perfective/imperfective aspect of the lexical VP head. Rather, in a given telicity structure in both Slavic and Germanic languages, we actually observe mutual constraints and interactions between the head verb and one of its semantic arguments, namely, the incremental argument.
Third, the variation in the encoding of telicity cannot be limited just to syntactic
factors. Instead, it is semantic (and also pragmatic) factors that ultimately motivate

(i) the phenomena that the syntactic parametric approach tries to capture, and also
(ii) telicity phenomena that are a priori recluded by it, left out or unnoticed. In this connection, I will defend the familiar (though often orgotten) insights of Krifka's (1986 and elsewhere) and Dowty's (1991) mereological theory of telicity.

Source: Semantics Archive

Topic: Interconnections

WHY ARE HUMANS UNLIKE OTHER ANIMALS

What are we supposed to say when the answer to a question is the question itself? This seems to have been the very puzzle of the studies on human behaviour and nature.

Topic: Cognition & Epistemology

What justifies that?

By Patrick Hawley

I clarify and defuse an argument for skepticism about justification with the aid of some results from recent linguistic theory. These considerations shed light on discussion about the structure of justification.

Take a look

Topic: GENERAL LOGIC

Minimalists about Truth can (and should) be Epistemicists, and it helps if they are revision theorists too

By Greg Restall

Minimalists about truth say that the important properties of the truth predicate are revealed in the class of T-biconditionals. Most minimalists demur from taking all of the T-biconditionals of the form 'p' is true if and only if p, to be true, because to do so leads to paradox. But exactly which biconditionals turn out to be true? I take a leaf out of the epistemic account of vagueness to show how the minimalist can avoid giving a comprehensive answer to that question. I also show that this response is entailed by taking minimalism seriously, and that objections to this position may be usefully aided and abetted by Gupta and Belnap’s revision theory of truth.

Topic: HUMAN SEMANTICS

Unexpected Substitutions

Context: The facts below (seem to) show that if (OP₁) and (OP₂) hold, they do not suffice, for they can only to account for the cases of unsuccessful substitutions, and the successful cases remain unexplained:

(OP₁)Opaque context (classic version)
A sentential context φ containing an occurrence of a term t is opaque, if the substitution of co-referential terms is an invalid mode of inference with respect to this occurrence. (See Mckinsey 1998, Quine 1956)

(OP₂)Church’ Substitutability of Identicals
Things are identical if the name of one can be substituted for that of the other without loss of truth.

Failure in Transparent Contexts

The first argument against the strict association between substitutability and opacity is that even in certain transparent contexts, i.e., where no propositional attitude is reported, the application of the substitutability principle does yield odd results. This evidences that one thing is independent from the other.
Consider the famous Fregean examples: since Venus has is styled Venus, morning star and evening star, the three expressions should be perfectly inter-exchangeable in a transparent context. But, if one takes the sentence (1a) and substitutes the evening star for Venus, the result (2b) is odd:

(1)a. Venus appears in the morning.==>
b. #The evening star appears in the morning.

An interpretation of Frege’s work traditionally attributes the oddity of (1b) to the sense versus referent distinction. Regardless of how to characterise it in theoretic terms of a semantic framework, it is clear that what is in question is not whether (1b) can be inserted into a believe clause, but the very presence of the adjective evening that modifies star in contradiction to the predicate appears in the morning.
Indeed there are numerous analogous examples that show called impossible syllogisms in transparent contexts, involving what is herein:

(2)a. Lepidopterans (can) fly.
b. Caterpillars are Lepidopterans. ==>
c. ƒ Caterpillars (can) fly.

In (2) there is no obvious evidence of a (hidden) propositional operator. Rather, what happens to (2) reflects revision of information and the defeasibility property of natural languages. Accordingly (2) states a default rule for the class of Lepidopterans, the exception being the stage of their lives when they are caterpillars. In other words, (2) is susceptible of reviewing, as most statements in natural languages. This same reasoning can be applied to sentences containing proper nouns, like (3):

(3) a. Captain Marvel looses his super-strength and his capacity to fly when he says the magic word.
b. Billy Batson is Captain Marvel. ==>
c. ƒ Billy Batson looses his super-strength and his capacity to fly when he says the magic word.

Sentence (3c) is the wrong depiction of the Comics book mentioned, since Billy Batson actually gains super-strength and the capacity to fly when he says the magic word (and consequently transforms into Captain Marvel).
Of course, in comparison, (6) cannot be explained solely in relation to defeasibility, although one could think of a context where it its uttered as a consequence of someone’s astronomical discovery, which forced him to review his former beliefs respecting the stars. The oddity of (6) has to do with issues of consistency, wherefrom one concludes that there is at least another fundamental semantic property involved.
Anyway, the examples above suffice to show that restrictions on substitutions transcend the case of propositional attitude ascriptions.

Non-Uniformity of Results in Opaque Contexts

The second argument is against the idea that attitude ascriptions always block Leibniz’ substitutability. Indeed, the substitution tests do not uniformly yield false results in all opaque context. On the contrary, some substitutions might yield even true results. Let us give one initial example, where the same object is designated by different names. Yet none of the possible alternative names or expressions to designate the same object changes the truth of the statement:

(4) Every visitor to the Louvre intends to see the Mona Lisa/ La Gioconda/ da Vinci’s most famous painting.

Of course, there are some tricks to make unlikely substitutions work. For instance, a true sentence like even Jameson believes that Peter Parker is a mere photographer does not normally yield a true result, if, as in (5a), one substitutes Spiderman for Peter Parker. But in a context like (5b) the substitution preserves the truth:

(5) a. ƒ Even Jameson believes that Spiderman is a mere photographer.
b. T The disguise is so convincing that even Jameson believes that Spiderman is a mere photographer. (See Berg 1988, Mckinsey 1999)

Topic: HUMAN SEMANTICS

Strict Identity with No Overlap

By Achille C. Varzi

It is natural to think that a standard, Kripke-style semantics for quantified modal logic (QML) is incompatible with the view that no individual can exist in more than one possible world, a view that seems to require a Lewis-style, counterpart-theoretic semantics instead. Strictly speaking, however, this thought is wrong-headed. A standard semantics regards a modal statement such as ?I might have been fat? as true only if I am in the extension of ?is fat? at some other possible world, whereas counterpart theory regards it as true only if a counterpart of mine is in the extension of ?is fat?. But just as the truth conditions of counterpart theory are in principle compatible with the possibility (rejected by Lewis) that some individuals qualify as their own other-wordly counterparts, the truth conditions of a standard semantics are in principle compatible with the possibility (dismissed by Kripke) that all individuals are world-bound. Here is how.
Source: Online Papers in Philosophy

Topic: HUMAN SEMANTICS

ON HIGGINBOTHAM'S LINKING THEORY

The occurence and the behaviour of co-referential nominals in sentences, under the labels of binding and control, have been classically treated as mainly syntactic phenomena, where the syntactic structure was licensed by teleological co-indexation considerations.

Binding and Control theories are the oldest and most resilient component of Generative thought that has survived to present day. The basic assumptions and core concepts of Classical Binding and Control theory have their origins in Langacker (1966), Rosenbaun (1967), Postal (1970), Jackendoff (1972) (Chapters 4 and 5), Chomsky (1973) and Lasnik (1976), who had tackled most of the main issues from a transformational perspective. Thereafter, this conceptual Binding theoretic nucleus has been re-formulated based on the theoretic refinement and enlargement of the application of Langacker's (1966) notion of command in Reinhart (1976) and Chomsky (1980, 1981 and 1982), together and in parallel with Bresnam (1982) and Manzini (1983). The format of Binding theory has remained almost unaltered since that time. Later proposals of reformulation, such as Reinhart and Reuland (1993), Reinhart (1999, 2000), Heim (2004) etc, have kept the essence of the model virtually intact.

One alternative that I consider at least captivating is Higginbotham's (1983) Linking theory. But, as far as I know, it has not been used by many.

If anyone reading this blog works with James Higginbotham's linking theory, please put a comment to let me know about your work. Thanks!

Topic: GENERAL LOGIC

Polynomial Ring Calculus for Logical Inference

By Walter Carnielli

This paper proposes a new "all-purpose" algebraic proof method applicable to general truth-functional sentential logics and to some non-truth-functional logics. The method, based on reducing polynomials over finite fields, is particularly apt for finitely-many-valued logics (and for classical propositional logic PC ). It can be extended to certain non-finitely valued logics and non-truth-functional logics as well, provided they can be characterized by two-valued dyadic semantics. The resulting mechanizable proof method introduced here is of interest for automatic proof theory, and seems also to be appropriate for investigating questions on complexity.

Source: CLE

Posted by Tony Marmo at 20:14 BST

Updated: Sunday, 10 April 2005 20:16 BST

Share This Post

Post Comment | Permalink