Semantics in Generative Grammar. Scope acrobatics

Transcription

1 1 Semantics in Generative Grammar Angelika Kratzer November 19 and 21, 2007 Scope acrobatics Supplementary readings for today s lecture: Most relevant: Janet D. Fodor & Ivan A. Sag (1982): Speaker s Reference and Quantificational Indefinites. Linguistics & Philosophy 5, Lisa Matthewson (1999): On the Interpretation of Wide-Scope Indefinites Natural Language Semantics 7, (Access to electronic copies via the course website). I m Nobody! Who are you? Are you - Nobody - too? Then there s a pair of us! Don t tell! they ll banish us - you know! How dreary - to be - Somebody! How public - like a Frog - To tell your name - the livelong June - To an admiring Bog! Emily Dickinson 1. Warm-up exercise on identifying scope ambiguities (1) We showed every subject two of the pictures. (a) For every subject x, there were 2 pictures y such that we showed y to x. (b) There were 2 pictures y such that we showed y to every subject x.

2 2 (2) We showed almost every subject two of the pictures. (a) For almost every subject x, there were 2 pictures y such that we showed y to x. (b) There were two pictures y such that we showed y to almost every subject x. Observation: 1(b) logically implies 1(a), hence we seem to have no case for positing a scope ambiguity for (1). On the other hand, 2(b) does not logically imply 2(a), hence we do have a case for positing a scope ambiguity for (2). But if there is a scoping mechanism that produces reading 2(b) for a sentence like (2), would we actually want to block that mechanism from applying to (1)? Or would we want to say that whatever scoping mechanism has to be invoked to produce reading 2(b) for (2) should be free to apply to (1) as well? The only thing that would be special about (1) would then be that it doesn t actually provide us with an argument for the existence of such a scoping mechanism. Scenario: 10 subjects, 1 to 10, and two pictures, A and B. A B Only 6 out of 10 subjects saw two of the pictures 2. The battle about indefinites The most famous battle concerning the semantic type of DPs was about definite descriptions, and it was fought between Russell and Strawson. In recent years, the semantic type of indefinite noun phrases has been the main focus of attention in the semantics of DPs. While there is no consensus as to the correct analysis yet, what has emerged very clearly is that there are different types of indefinites. All indefinites are not the same across

3 3 languages, and even within a single language, we usually find different kinds of indefinites. Some of the apparently exceptional properties singling out indefinites are the following: a. Scope, as shown by the ability to bind pronouns ( donkey-anaphora ). b. Chamelion-like behavior: Apparent quantificational variability effect. Some occurrences of some kinds of indefinites seem to have universal force, for example. Again, donkey-sentences show this phenomenon. c. Scope with respect to other quantifiers, negation, modals, etc. d. Apparent violations of scope islands (Fodor and Sag data, today s lecture). 3. Fodor and Sag (1) Each teacher overheard the rumor that a student of mine had been called before the dean. Fodor & Sag (1982), example (69), p. 374 (1) has a reading where the indefinite DP a student of mine has scope within the that - clause. What the teachers overheard might have been: A student of Angelika s was called before the dean. This reading is expected if indefinite DPs are quantifiers, and quantifier scope is confined to some local domain. (1) has another reading where a student of mine seems to have widest scope, scope even wider than each teacher. There might have been a student of mine, say Sanders, and each teacher overheard the rumor that Sanders was called before the dean. Fodor and Sag argue that this is not an instance of scope. If indefinite DPs seem to have anomalous scope properties, they are not true quantifiers. The apparent wide-scope reading of a student of mine in (1) is really a referential reading. Indefinite DPs, then, are ambiguous between quantificational and referential readings. If they are quantificational, their scope 1. The prose that follows is adapted from my Scope or Pseudoscope? Are there Wide-Scope Indefinites?. In: Susan Rothstein (ed.): Events in Grammar. Dordrecht (Kluwer Academic Publishers), 1998,

4 4 cannot exceed some local domain. If they are referential, they do not have scope at all. They may be easily confused with widest scope existentials, however. Sentence (1) is important for Fodor and Sag s argument since it offers a third scope possibility for the indefinite DP that doesn t seem to be there. If indefinites were quantifiers that can freely violate scope constraints, we would expect sentence (1) to have a reading where a student of mine has intermediate scope, scope wider than the that - clause, but narrower than each teacher. There might have been different students of mine, say, Sanders, Monroe, and Wilson, and the science teacher overheard that Sanders was called before the dean, the math teacher overheard that Monroe was called before the dean, and the woodshop teacher overheard that it was Wilson who suffered that fate. (1) lacks this reading, thereby supporting Fodor and Sag s proposal. Fodor and Sag s examples and conclusions have been questioned 2. The most serious challenges comes from a recent paper by Dorit Abusch 3 (electronic access via the website). 4. The challenge: Abusch Abusch ( ) presents many cases of indefinite DPs that seem to have the intermediate scope readings that Fodor and Sag claim do not occur. By way of illustration, I want to look at two representative examples. Consider first Abusch s Professor example (2) 4 : The Professor example (2) Every professor rewarded every student who read some book he had recommended. (2) has a reading where every professor picked some book out of all the books he had recommended, and rewarded every student who read that book. Different books might have been picked by different professors. The Professor of French might have picked Madame 2. See e.g. King (1988), Ludlow and Neale (1991), Ruys (1992). An earlier important reference is Farkas (1981). 3. The Scope of Indefinites. Natural Language Semantics 2, 1994, The example is due to Abusch (example (10), p.90), except for the choice of indefinite article. I picked some instead of a.

5 5 Bovary from the list of French novels he had recommended. The Professor of Italian might have chosen I Promessi Sposi. And the Professor of English might have insisted on The House of Mirth. If (2) is understood this way, the indefinite DP some book he had recommended cannot be referential. It does not refer to a particular book. But if it is quantificational, it must have been scoped out of a relative clause, violating a scope constraint. It is not that students were rewarded just because there was some recommended book they read. For every professor, there was a particular book that was singled out for a reward. The second Abusch example I want to present is the Stuttgart example (3): The Stuttgart example (3) Every one of them moved to Stuttgart because some woman lived there. Abusch 1994, example (20), p. 94. (3) has a reading where everyone of them was attracted to Stuttgart by a possibly different woman. That is, neither was it necessarily one and the same woman who attracted everyone to Stuttgart, nor the fact that Stuttgart is not womenless. Here, too, the indefinite DP is not referential. And if it is quantificational, it must have been scoped out of the because-clause, violating again a scope constraint.