Seth Cable Semantics and Generative Grammar Fall 2011 Ling610 1 The Conceptual Foundations of Truth-Conditional Semantics1 1. The Big Questions Semantics is the study of meaning. But what does that mean? Well, consider the following, opening passage from Stephen Pinker’s The Language Instinct: As you are reading these words, you are taking part in one of the wonders of the natural world. For you and I belong to a species with a remarkable ability: we can shape events in each other’s brains with exquisite precision… …Simply by making noises with our mouths, we can reliably cause precise new combinations of ideas to arise in each other’s minds. The ability comes so naturally that we are apt to forget what a miracle it is. Asking you only to surrender your imagination to my words… I can make you think some very specific thoughts… (1) The Over-Arching Miracle: The Productivity of Language (Semantics) The speaker of a language L can understand all the countably infinite sentences of L. How do we do this? Since brains/lifetimes are finite, this knowledge must be represented in our brains as some kind of combinatoric system, one that comprises: a. A finite number of primitive meaningful units (lexemes, lexical items) b. A finite set of ‘rules’ for deriving the meaning of a complex expression from the meanings of the primitives and the structure of the complex expression (2) A Handy – and Common – Locution for this Property The meaning of complex linguistic expressions is (generally) compositional, in the following sense… a. The Principle of Compositionality (cf. Partee 1995) The meaning of a complex expression can be effectively computed from the meaning of its component expressions 1 These notes are based on material in Heim & Kratzer (1998: 1-12, 13-26), Chierchia & McConnell-Ginet (2000: 1-33, 53-73, 99-104), Larson (1995: 361-368), and Partee (1995: 311-316).Seth Cable Semantics and Generative Grammar Fall 2011 Ling610 2 So, since we are speakers of English, we all know and employ some system for computing the meanings of complex expressions… What’s especially interesting is that our knowledge of this sytem is tacit, unconscious… …therefore, if we want to find out what this ‘combinatory system’ is, we need to do SCIENCE! (3) The Over-Arching Research Question What is the combinatoric system (algorithm) that our cognitive systems employ to derive/compute the meanings of complex expressions from the meanings of their component parts? (4) Some Related, More Specific Research Questions a. How does a human being acquire this combinatoric system? How much is already specified by the biology of the organism? b. How does this combinatoric system vary across languages? Do languages differ in how they ‘compute meanings’, and if so, in what ways? Fact: These questions don’t receive nearly as much attention within semantics that they do in syntax and phonology (but this is improving…) 2. The Meaning of ‘Meaning’ (5) An Immediate Problem for Our Project What is a ‘meaning’? • To answer the question in (3), we want to develop some hypotheses about what the system is like, and then test them… • So, we’ll want to design a hypothetical formal system that will manipulate primitive ‘meanings’ to derive the ‘meanings’ of more complex sentences. • But, how do we formally represent the ‘meaning’ of a sentence or its component phrases? • What is the ‘meaning’ of “Barack” and the ‘meaning’ of “smokes” such that ‘combining them together’ gives us the ‘meaning’ of “Barack smokes”?Seth Cable Semantics and Generative Grammar Fall 2011 Ling610 3 Over-Arching Problem: • The word ‘meaning’ is a vague, pre-theoretic term from every-day discourse. (like ‘alive’, ‘hot’ and ‘heavy’) • Thus, it may not be an appropriate term for a precise, scientific study of human language. (e.g. biology doesn’t actually employ terms like ‘alive’ or ‘dead’ or ‘life-form’) (e.g. physics doesn’t actually employ terms like ‘hot’, ‘heavy’, ‘fast’, etc.) (6) New, Preliminary Objective Let’s replace our everyday, pre-theoretic concept of ‘meaning’ with something more precise, so that we might work towards a formal system that can manipulate ‘meanings’… Side-Benefit of Formalization: Better Description of ‘Meanings’ A side-benefit of developing (and working within) a formalized semantics is that it forces a richer and more precise documentation of linguistic meanings. • In traditional grammars, particularly of ‘minority languages’, one often finds frustratingly vague descriptions of the meanings of certain (functional) elements. • By employing a formal semantics, the researcher is forced to ask and answer deeper semantic questions about the language, resulting in greater depth of descriptive coverage… • Without a formal semantics, certain deeper semantic questions about a given language cannot even be framed. … so how can we better pin down the phenomena/properties that we are interested in, those that are typically, loosely categorized under the general umbrella of ‘meaning’?...Seth Cable Semantics and Generative Grammar Fall 2011 Ling610 4 (7) ‘Meaning’ is as ‘Meaning’ Does “In order to say what a meaning is, we may first ask what a meaning does, and then find something that does that” (David Lewis; “General Semantics”) What (exactly) do we know when we know ‘the meaning’ of a sentence? Well, we know a lot of things! (a) Social Appropriateness of the Statement: What social contexts the statement is appropriate in. “That’s wonderful.” vs. “That kicks ass!” (b) ‘Emotional Content’ of the Statement: What the statement reveals about the emotional state of the speaker. “I disagree with Dave’s judgment.” vs. “Dave is a damn fool!” (c) The Informational Content of the Statement What information about the world the sentence ‘conveys’. Rightly or wrongly, (c) has received by-far-and-away the greatest attention over the centuries It will