Do we actually need a resultative construction? A reappraisal based on data from Dutch

This paper argues that the resultative construction (e.g. Goldberg 1995; Boas 2003; Broccias 2003; Goldberg and Jackendoff 2004) does not exist as an argument-structure construction in Goldberg's (1995) sense, i.e. as a structural template of the form V NP AP/PP meaning 'CAUSE X to BECOME Y by Z-ing' and allowing practically any verb to fill the verb slot. We start from Kay's (2005; to appear) objections to the existence of a productive caused-motion construction which is assumed to give rise to e.g. He sneezed the napkin off the table, a core example in Construction Grammar. Kay argues that such rare occurrences are just nonce, analogical uses (cf. also Herbst (2010: 244): they are "conscious and intended violations of the norm") and moreover, that other more frequently attested cases of an otherwise intransitive verb combining with an object and a path argument, such as They laughed him off the stage, are more or less conventionalized (cp., e.g., ??They snored him off the stage). So, there are only some rather lexicalized resultative expressions in the language and these are only very occasionally taken as models for creative extension. We extend Kay's objections by claiming that the resultative 'mother construction', which the caused-motion construction is a subtype of, can be done without (cf. also Boas 2003). While we make this claim with respect to the oft-studied English resultative, we adduce supporting evidence from Dutch, a closely-related language abounding with conventionalised resultatives taking an unselected (often fake reflexive) object, like (1) (Cappelle to appear): (1) a. Het vriest de stenen uit de grond. b. We betalen ons blauw. c. Ik lach me rot! In addition, Dutch has similarly hyperbolic fake-object resultatives with an NP (rather than an AP/PP) as the resultative phrase, requiring some kind of 'HAVE' or 'COME INTO BEING' rather than a 'BECOME' semantic analysis (cf. cry me a river for a rare, perhaps unique case in English). Some examples are given in (2): (2) a. Ik schrik me een hoedje. b. We verveelden ons de tering. c. Ik lach me een bult! Our argumentation for treating the (English) resultative construction as superfluous is based on the following considerations/observations: 1. English unselected object resultatives similar to the examples in (1) (e.g. laugh one's head off, talk oneself blue in the face) are not always accepted as true resultatives, on account of their excess semantics (Jackendoff 1997; 2002; for other views, see Sawada 2000; Goldberg & Jackendoff, 2004; Kudo to appear). 2. A distinction between 'excess resultatives' and ordinary resultatives (like hammer the metal flat) is supported by the fact that the former come in two subtypes in Dutch (cf. (1) and (2) above) and that there do not seem to be (m)any convincing ordinary resultatives sharing the structure of the excessive resultatives in (2), e.g. *Ze vergiste zich een mislukte taart 'She miscalculated herself a ruined cake'. This warrants positing a verb intensification construction, distinct from ordinary resultatives. Even if this is not accepted, it remains a fact that most instances of the excess construction exemplified in (1) and (2) are fairly idiomatic: they are not created on the fly but form part of the lexico-grammar of Dutch. The same goes for structural counterparts of (1) in English (e.g. cry/?sob one's eyes red; drink somebody under the table/??off his chair); novel ones are stylistically marked. 3. Ordinary resultatives, of the type hammer the metal flat, may not require a special argument structure construction, since, plausibly, they are created analogically or since the concerned verb's lexical entry already includes a NP+AP/PP valency. Glosses (1) a. Het vriest de stenen uit de grond. it freezes the stones out the ground 'It's freezing very hard.' b. We betalen ons blauw. we pay us (= ourselves) blue 'We're paying an awful lot of money.' c. Ik lach me rot! I laugh me (= myself) rotten 'I'm rolling on the floor laughing!' (2) a. Ik schrik me een hoedje. I start me (=myself) a little-hat 'I'm startled out of my wits.' b. We verveelden ons de tering. we bored us (= ourselves) the phthisis 'We're bored to death.' c. Ik lach me een bult! I laugh me (= myself) a hunch 'I'm rolling on the floor laughing!' References Boas, Hans C. (2003) A Constructional Approach to Resultatives. Stanford: Center for the Study of Language and Information. Broccias, Cristiano (2003) The English Change Network. Berlin / New York: Mouton de Gruyter. Cappelle, Bert (to appear) Erop los intensifiëren, deel 2. Over Taal. Goldberg, Adele E. (1995) Constructions. Chicago: University of Chicago Press. Goldberg, Adele E. & Ray Jackendoff (2004) The English resultative as a family of constructions, Language 80(3), 532-568. Herbst, Thomas (2010) Valency constructions and clause constructions or how, if at all, valency grammarians might sneeze the foam off the cappuccino. Hans-Jörg Schmid & Susanne Handl (eds.), Cognitive Foundations of Linguistic Usage Patterns, 225-255, Berlin / New York: Mouton de Gruyter. Jackendoff, Ray (1997) Twistin' the night away. Language 73(3), 534-59. Jackendoff, Ray (2002) Foundations of language. Oxford: Oxford University Press. Kay, Paul (2005) Argument structure constructions and the argument/adjunct distinction. Mirjam Fried & Hans C. Boas (eds.), Grammatical constructions: Back to the roots, 71-98. Amsterdam / Philadelphia: John Benjamins. Kay, Paul (to appear) The Limits of Construction Grammar. Graeme Trousdale and Thomas Hoffmann (eds.), The Oxford Handbook of Construction Grammar. Oxford: Oxford University Press. [Previously circulating as the 2002 ms. "Patterns of Coining"] Kudo, Shun (to appear) A comparative study between Resultative Constructions and Body Part Off constructions. Tsukuba English Studies 29. Sawada, Shigeyuki (2000) The Semantics of the 'Body Part Off' Construction, English Linguistics 17(2), 361-385