09.07.2015 Views

Practical Information - Generative Linguistics in the Old World

Practical Information - Generative Linguistics in the Old World

Practical Information - Generative Linguistics in the Old World

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

A Dual-Source Analysis of Gapp<strong>in</strong>gDavid Potter, Michael Frazier, Masaya YoshidaGapp<strong>in</strong>g constructions (1) have long been known to be ambiguous with respect to <strong>the</strong> scopeof modals and negation [11], with <strong>the</strong> scope-tak<strong>in</strong>g material scop<strong>in</strong>g ei<strong>the</strong>r under <strong>the</strong> coord<strong>in</strong>ator,yield<strong>in</strong>g distributed scope, or above <strong>the</strong> coord<strong>in</strong>ator, yield<strong>in</strong>g wide scope. Extant analysesof Gapp<strong>in</strong>g [2, 5, 4, 6, 9] stumble over this ambiguity, fail<strong>in</strong>g to account for its full distributionand a constra<strong>in</strong>t on split scope: multiple scope-tak<strong>in</strong>g elements cannot be split between wideand distributed scopes. In turn, we propose that Gapp<strong>in</strong>g is a heterogenous phenomenon andthat this scope ambiguity should be reduced to a structural ambiguity between <strong>the</strong> coord<strong>in</strong>ationof CPs (2a) and vPs (2b).(1) Jim can’t eat caviar and Sue can’t eat beans.(2) a. [ CP Jim can’t eat caviar ] and [ CP Sue i beans j [ T P t i can’t eat t j ]]b. Jim can’t [ vP eat caviar ] and [ vP Sue i beans j [ vP t i eat t j ]]Distributed scope follows from <strong>the</strong> <strong>in</strong>terpretation of scope-tak<strong>in</strong>g material with<strong>in</strong> <strong>the</strong> scopeof <strong>the</strong> coord<strong>in</strong>ation; this read<strong>in</strong>g of (1) can be paraphrased as “Jim is not permitted to eatcaviar and Sue is not permitted to eat beans.” In <strong>the</strong> wide scope read<strong>in</strong>g, <strong>the</strong> scope-tak<strong>in</strong>gmaterial scopes over <strong>the</strong> coord<strong>in</strong>ator and can be paraphrased “it is not permitted for Jim to eatcaviar and for Sue to eat beans.” Most o<strong>the</strong>r elements <strong>in</strong> <strong>the</strong> IP and CP doma<strong>in</strong>s participate<strong>in</strong> this ambiguity, <strong>in</strong>clud<strong>in</strong>g high adverbs (3), aspect (4), and epistemic and root modality (5).Fur<strong>the</strong>rmore, scope-tak<strong>in</strong>g material <strong>in</strong> <strong>the</strong> IP and CP doma<strong>in</strong>s must all take ei<strong>the</strong>r distributed orwide scope. (1) cannot be <strong>in</strong>terpreted as <strong>in</strong> (6a) or (6b): split scope is impossible <strong>in</strong> Gapp<strong>in</strong>g.(3) James is probably writ<strong>in</strong>g his term paper and Mary her f<strong>in</strong>al exam(4) James has been work<strong>in</strong>g hard on <strong>the</strong>ir article and Mary on <strong>the</strong>ir presentation.(5) a. James might vote <strong>in</strong>dependent and Mary Democratb. James can cook <strong>the</strong> pasta and Mary <strong>the</strong> chicken.(6) a. It is not <strong>the</strong> case that Jim is permitted to eat caviar and Sue is permitted to eat beans.b. It is permitted for Jim not to eat caviar and for Sue not to eat beans.Material <strong>in</strong>terpreted below <strong>the</strong> IP doma<strong>in</strong>, <strong>in</strong>clud<strong>in</strong>g manner adverbs and adverbial negation (7),can only take distributed scope. However, wh-words at <strong>the</strong> left edge of <strong>the</strong> correlate conjunctare <strong>in</strong>terpreted with a s<strong>in</strong>gle referent, obligatorily tak<strong>in</strong>g wide scope along with can’t (8).(7) a. James quickly ate <strong>the</strong> beans and Mary <strong>the</strong> rice.b. James can’t not eat beans and Mary rice.(8) Who can’t James meet on Monday and Bill on Tuesday?This data suggest that Gapp<strong>in</strong>g is supported <strong>in</strong> exactly two configurations: CP and vPcoord<strong>in</strong>ate structures. In <strong>the</strong> large-conjunct structures, each conjunct conta<strong>in</strong>s a copy of <strong>the</strong>gapped material, result<strong>in</strong>g <strong>in</strong> <strong>the</strong> distributed scope read<strong>in</strong>g. Remnants <strong>in</strong>terpreted <strong>in</strong> <strong>the</strong> CPdoma<strong>in</strong>, <strong>in</strong>clud<strong>in</strong>g epistemic modality (5a)[1], and topicalized elements (9) [10] can receivea distributed <strong>in</strong>terpretation, and so <strong>the</strong> large conjuncts must be full CPs (contra [9, 6, 4]). In<strong>the</strong> small conjunct structures, <strong>the</strong> scope-tak<strong>in</strong>g material <strong>in</strong> <strong>the</strong> IP and CP doma<strong>in</strong>s c-commandsand consequently takes wide scope over <strong>the</strong> low coord<strong>in</strong>ate structure. Material with<strong>in</strong> <strong>the</strong> vP1

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!