The Syntax of Wherewithal

The Syntax of Wherewithal Richard S. Kayne New York University April, 2017 1. Introduction English, or at least some English, allows: (1) They don’t have the wherewithal to buy that big a house. in approximately the sense of They don’t have the means/money to buy that big a house. A lexicalist approach to syntax in the spirit of Chomsky (1970), although it might well recognize that wherewithal is composed of three morphemes, where, with and al(l), would, since it takes words to be atomic with respect to Merge, deny that the three morphemes of wherewithal are put together by Merge. On the other hand, if the morphemes that constitute a word can be put together by Merge, as they could be in an updated and generalized version of Chomsky’s (1957) affix-hopping analysis of English forms such as played,1 then the component morphemes of wherewithal could be, too. Yet allowing that the three morphemes of wherewithal are treated as atoms by Merge underdetermines the derivation of (the relevant part of) (1) to a significant degree. We don’t automatically know whether where, with and al(l) are affected solely by external merge or whether they are also subject to internal merge (movement). Nor do we automatically know in what order Merge affects them. Finally we don’t automatically know whether or not there are in (1) additional relevant morphemes that are not pronounced. In this paper, I will try to spell out what an analysis might look like that takes the component morphemes of wherewithal to be manipulated by Merge. In so doing, I will suggest (the beginnings of) an account of the absence in English of various words that resemble wherewithal but are not at all possible. For example, the following are unacceptable: (2) *whatwithal; *whenwithal; *herewithal; *therewithal; *wherewithboth A merge-based account of (2) is, I think, within our reach, in a way that would seem not to be the case within a lexicalist (non-merge) approach to internally complex words. It should be noted that to call wherewithal a word is in effect to take it to be a constituent. But words in standard orthography, although they may well correspond to syntactic constituents in many cases, do not necessarily do so,2 and in fact the analysis to be suggested in what follows leans toward the conclusion that wherewithal is not a syntactic constituent in sentences in which it occurs. 2. The subparts The wherewith subpart of wherewithal recalls the following English words:3 (3) whereby; thereby; whereupon; therefore; wherein These are to one degree or another still part of contemporary English. In earlier English, words of this form were more productive (e.g. whereof, thereof, wherefore), in the manner (as a first approximation) of ———————————— 1 Cf. Baker (1985; 1988), Pollock (1989) and Roberts (2017), among others. Cf. Myers (1987), Koopman and Szabolcsi (2000), Julien (2002). 3 Also whereabouts, which raises somewhat different questions; see Kayne (2014). 2 contemporary Dutch and German.4 As in van Riemsdijk’s (1978) original discussion, we can take, say, therefore to be very close to for that (or more exactly to ‘for that REASON’ with a silent noun corresponding to reason), and in the same way we can take wherewith to be very close to with what, apart from the difference in order between adposition and wh-word, and the difference between the R-pronoun where and the non-R-pronoun what. As for the -al subpart of wherewithal, it seems virtually certain that it is simply all.5 (On the spelling difference, cf. alright, altogether, almost, also, already, all with al- rather than all; also careful, spoonful vs. full.) Taking the -al of wherewithal to equal ordinary all might seem a bit surprising at first, insofar as all in English cannot normally be preceded by the, whereas the -al of wherewithal normally is (though not immediately preceded), as in (1). However, the definite article in (1) recalls that found overtly with whole (as in the whole day), which has a fair amount in common with all. In addition, the fact that the in (1) cooccurs with all (written as -al) recalls even more specifically: (4) He gave it his all. in which all is not in its more usual DP-initial position.6 Returning to the wherewith component of wherewithal, let me take it to be an instance of a relative pronoun followed by an adposition, parallel to: (5) They had a plan whereby they would both leave at the same time. From this perspective, (1) should be thought of as:7 (6) ...the wherewith all to... In contrast to (5), though, there is no visible noun in (6) between the determiner (a, the) and the relative pronoun where. Two possibilities come to mind. First, the apparently missing noun might be all itself. The second possibility, which is the one I will prefer here, has there being a silent noun in (6), much as there is arguably a silent noun present in headless relatives of the sort seen in:8 (7) They ate what was put in front of them. If all itself is not the nominal head of the relative in (6),9 the question arises as to what the status of this all is. Let me take as a clue the existence in earlier English of interrogative wherewithal, as in:10 ———————————— 4 Cf. Kayne (2004, Part III) on a possible link within Germanic to ‘OV’ vs. ‘VO’. For recent discussion of OV/VO, see Cinque (2016). 5 Note that the final vowel of wherewithal (equal to that of all) is not reduced, as opposed to that of refusal, removal. 6 In this example, there is very likely a silent N, too (akin to EFFORT, perhaps), within the DP containing his all. 7 Note the parallelism with: i) They don’t have the money with which to buy that big a house. Some speakers accept : ii) They would like to buy that car, but they don’t have the wherewithal. iii) They don’t have the wherewithal for that car. recalling Collins (2014) on deleted relatives. 8 Cf. Groos and Riemsdijk (1981). 9 Keeping in mind that the term ‘head (of a relative)’ here is not the X-bar sense of that term. 10 This example is from the King James version of the Bible, as given by the Oxford English Dictionary (wherewithal). Archaic withal is found in the OED, too; in some of its uses, it may be accompanied by a silent WHERE. (8) Wherewithall shall wee be clothed? which to my more recent ear recalls McCloskey (2000, 66) on West Ulster English sentences such as: (9) Who all did you give tea to? (10) Where all did you move the books to? If so, then we can think of (8) as: (11) Wherewith all...? with the difference in position of the adposition reflecting the extensive use in earlier English of wherewith, whereof et al. If the -al of earlier English interrogative wherewithal is, as I am suggesting, related to McCloskey’s all in (9) and (10), then it is plausible to think that current English relative wherewithal, as in (1)/(6), also contains an all that is substantially the same as that of West Ulster English. Thus we now have additional grounding for taking the -al of current wherewithal to equal all. As for the constituent question alluded to earlier, i.e. the question whether wherewithal in (1)/(6) is a syntactic constituent, it is logically possible that it is a constituent, but the presence of the adposition with between relative pronoun where and al(l) makes that unlikely. If so, then wherewithal is an example of a word that is not a syntactic constituent. 3. Derivation As for more specific questions of properties and derivation, we can take wherewith(al)+infinitival relative in (1)/(6), repeated here: (12) They don’t have the wherewithal to buy that big a house. to have a lot in common with: (13) They don’t have the money with which to buy that big a house. even though a simpler where-initial infinitival relative is at best marginal in English (as opposed to French):11 (14) ?They need a place where to store their books. Sharply impossible, on the other hand, is the result of adding all to (13): (15) *They don’t have the money with which all to buy that big a house. It may be that the contrast between (12), which allows ‘...wherewith all...’, and (15), which disallows ‘...with which all...’, can be understood as follows. Contrary to appearances, (12) does have a pronounced relative head, namely all, or more likely all+silent noun (N), meaning that (12) would have at an intermediate stage in its derivation: (16) all N wherewith to buy... Wherewith would then move to the left of all (perhaps via remnant movement). This movement would have something in common with the movement of relative clauses from postnominal (post-head) position to prenominal (pre-head) position suggested for languages like Chinese and Japanese in Kayne (1994, sect. 8.3), especially for cases like the recently arrived letter, in which the preposed relative ends up being preceded by the, as well as having something in common with the movement of destruct- past -ion in the derivation of derived nominals like the destruction of the bridge, as proposed in Collins (2006) and Kayne (2008). ———————————— 11 Note that the English restriction found in infinitival relatives: i) *We’re trying to think of someone who to invite. is found in both French and Italian even with finite relatives, with direct object qui, cui; cf. Kayne (1976) and Cinque (1982). The idea would then be that the specificity of wherewith in contemporary English is precisely that it, and it alone, is subject to movement past all, so that (15) is not derivable. Of course (15) is not derivable for a second reason, from this perspective, namely that it would have two heads, both money and all (or all+N, henceforth just all), as also seen in: (17) *They don’t have the money wherewithal to buy that big a house. Removing (the) money from (15) and undoing the movement of with which yields: (18) *They don’t have all with which to buy that big a house. the unacceptability of which is almost certainly linked to that of: (19) *They bought all which was on the top shelf. Relatives headed by all are hardly possible with relative which. In a similar vein, we can note the impossibility in contemporary English of: (20) *They don’t have the money wherewith to buy that big a house. In effect, wherewith is currently possible only in combination with al(l), in a way that recalls the limitation of German relative was (‘what’) to light-headed relatives.12 English relatives headed by all are to some extent possible with that rather than which: (21) ?They bought all that was on the top shelf. In contrast to (12) with wherewithal, though, (21) cannot have the: (22) *They bought the all that was on the top shelf. That the all of wherewithal can be preceded by the within its DP recalls French le tout (on the assumption that French tout is a good match to English all, which seems plausible), as in:13 (23) Le tout a été envoyé à Paris. (‘the all has been sent to Paris’ = ‘the whole thing...’) and also recalls, though the determiner is different, (4), repeated here: (24) He gave it his all. as well as, less directly: (25) They watched his every step. (26) They ate the whole cake. Why universal quantifier-like elements can sometimes be preceded by the and sometimes not is left an open question. 4. al(l) vs. both That the al(l) of wherewithal and the all of (24) are closely related is supported by the fact that neither can be replaced by both: (27) *wherewithboth (28) *He gave it his both. nor with all+numeral: (29) *wherewithalthree (30) *He gave it his all three. nor with whole or half or almost all: (31) *wherewithwhole; *wherewithhalf; *wherewithalmostal (32) *He gave it his whole/*his half/*his almost all. Possible, on the other hand, is: (33) He gave it his whole effort. ———————————— 12 13 Cf. Citko (2004). On le tout vs. tout alone, see Obenauer (1994). but not: (34) *wherewithwholeeffort due to the fact mentioned in the discussion of (20), namely that wherewith is restricted to occurring in combination with bare all (or all + silent N) as relative head. The restriction seen in (27) is also found in West Ulster English, in the sense that the West Ulster English examples (9) and (10), repeated here: (35) Who all did you give tea to? (36) Where all did you move the books to? have no counterparts with both:14 (37) *Who both did you give tea to? (38) *Where both did you move the books to? This gives further support to the idea that the all of wherewithal is related to the all studied by McCloskey (2000). We can now see clearly, I think, that the non-existence of (27) is not an accidental isolated fact about the English lexicon, but is tied to differential syntactic properties of all vs. both seen in (24) vs. (27) and in (35)-(38). This linkage is expressible, though, only if words like wherewithal (and by plausible extension all words containing more than one morpheme) are put together in the syntax (via Merge, both external and internal). There remains, needless to say, the question why all and both diverge in their behavior is these specific ways. The core of an answer may be as follows. Both is more complex than all, just as all three is more complex than all.15 Both is comparable to all two, perhaps with a silent ALL. The appearance of all in (24) and in (35)/(36), and therefore in wherewithal, depends on its (relative) lack of complexity (i.e. on its not being accompanied by a numeral). Consequently, there can be no examples of words like (27) (and perhaps similarly for (31), at least for almost all and for half). 5. Romance In addition to accounting for the absence of certain logically possible words in English, the perspective outlined above can account for the absence of wherewithal itself in any Romance language (as far as I know), in terms of the absence of wherewith (along with whereby, whereof, wherefore) in any Romance language (as far as I know).16 6. Other impossible words ———————————— 14 I am grateful to Jim McCloskey (p.c.) for the judgments. The all in question may have something in common with the ever of free relatives. 16 As for why these are absent from Romance languages, see note 4 on the VO vs. OV question, keeping in mind that English is more ‘OV’ than any Romance language when it comes to compounds of the magazine reader or magazine reading sort - cf. Emonds and Faarlund (2014,20) - though the details of this linkage remain to be worked out. Some 'OV'-ness may also be a necessary condition for 'V DP Prt' order of the pick the book up sort, which seems to be absent from all Romance languages (Andrea Padovan, p.c.), as opposed to 'V Prt DP', which is found in some Romance. More complex than the where-X cases are those with there; it may be, though, that French là-dedans ('there of in') is closer to in there than to therein - cf. McCawley (1988, note 12) and Rizzi (1988). 15 The syntax of (the) wherewithal that I have proposed takes its where component to be a relative pronoun. If a relative clause structure is the only way to reach that sort of complex word, then we may have an account of the following: (39) *They don’t have the herewithal to buy that big a house. (40) *They don’t have the therewithal to buy that big a house. since here and there are not possible in English as relative pronouns. Somewhat similarly, if wherewith in wherewithal is of the whereby type found more robustly in Dutch and German, then we can exclude: (41) *They don’t have the whatwithal to buy that big a house. (42) *They don’t have the whenwithal to buy that big a house. in terms of the general exclusion within Germanic of forms like:17 (43) *whatby, *whenby with non-R-wh-words. 7. Challenges In my English, wherewithal has no plural: (44) *We don’t have the wherewithals to buy that big a house. From the present perspective this may be related to: (45) *We gave it our alls. To judge by a Google search, however, there seem to be speakers for whom sentences like (44) are possible. It may be that the silent N arguably present with wherewithal can be accompanied by plural -s for some speakers, but not for others, though it remains to be understood why I allow plural -s with a silent noun in: (46) the others; two four-year-olds; the extra-wides while disallowing it in (44). A second challenge comes from the fact that some speakers accept: (47) We don’t have the wherewithal with which to do it. Such speakers must be allowing adposition doubling, recalling the existence, in some English, of cases like: (48) the problem to which they’re alluding to The speakers in question must also be allowing, from the present perspective, two wh-phrases (where, which) both of which have ‘all N’ as antecedent, thereby having something in common with:18 (49) Prof. Mary Smith, whose students’ admiration for whom is well-known,... and perhaps also with German:19 (50) lch habe Blumen gebracht. Was für welche? (‘I have flowers brought. What for which? (= ‘What kind?’)’) 8. Conclusion ———————————— 17 In all likelihood, examples like: i) What about were you guys talking? which are acceptable to Bob Frank, are of a different character. 18 Cf. Kayne (1983; to appear). 19 Durrell (2002, 93). In conclusion, then, sentences containing wherewithal have a richer syntax than might at first glance be apparent. This richer syntax provides a handle on the absence of various potential words such as *whatwithal, *whenwithal, *herewithal, *therewithal, *wherewithwhole, *wherewithhalf, *wherewithalmostal, *wherewithboth that an approach associated with a less rich syntax could not provide. References: Baker, M. (1985) “The Mirror Principle and Morphosyntactic Explanation,” Linguistic Inquiry, 16, 373415. Baker, M.C. (1988) Incorporation: A Theory of Grammatical Function Changing, The University of Chicago Press, Chicago. Chomsky, N. (1957) Syntactic Structures, Mouton, The Hague. Chomsky, N. (1970) “Remarks on Nominalization,” in R.A. Jacobs and P.S. Rosenbaum (eds.) Readings in English Transformational Grammar, Ginn, Waltham, 184-221. Cinque, G. (1982) “On the Theory of Relative Clauses and Markedness,” The Linguistic Review, 1, 247294. Cinque, G. (2016) “A microparametric approach to the head-initial/head-final parameter,” to appear in S. Karimi and M. Piattelli Palmarini (eds.) Parameters: what are they? Where are they?, Linguistic Analysis vol.42 . Citko, B. (2004) “On Headed, Headless, and Light-Headed Relatives,” Natural Language and Linguistic Theory, 22, 95-126. Collins, C. (2014) “Relative Clause Deletion,” ms., New York University. Collins, C. (2006) “A Note on Derivational Morphology,” ms., New York University. Durrell, M. (2002) Hammer’s German Grammar and Usage, fourth edition, McGraw-Hill, Chicago. Emonds, J.E. and J.T. Faarlund (2014) English: The Language of the Vikings, Palacký University, Olomouc. Groos, Anneke & Henk van Riemsdijk (1981) “Matching Effects in Free Relatives: A Parameter of Core Grammar,” in A. Belletti, L. Brandi & L. Rizzi (eds.) Theory of Markedness in Generative Grammar. Proceedings of the 1979 GLOW Conference, Scuola Normale Superiore, Pisa, 171-216. Julien, M. (2002) Syntactic Heads and Word Formation, Oxford University Press, New York. Kayne, R.S. (1976) “French Relative ‘que’,” in F. Hensey and M. Luján (eds.) Current Studies in Romance Linguistics, Georgetown University Press, Washington, D.C., 255-299. Kayne, R.S. (1983) “Connectedness,” Linguistic Inquiry, 14, 223-249 (reprinted in Kayne (1984)). Kayne, R.S. (1984) Connectedness and Binary Branching, Foris, Dordrecht. Kayne, R.S. (1994) The Antisymmetry of Syntax, The MIT Press, Cambridge, Mass. Kayne, R.S. (2004) “Here and There” in C. Leclère, E. Laporte, M. Piot & M. Silberztein (eds.) Syntax, Lexis & Lexicon-Grammar. Papers in Honour of Maurice Gross, John Benjamins, Amsterdam, 253273 (reprinted in Kayne (2005)). Kayne, R.S. (2005) Movement and Silence, Oxford University Press, New York. Kayne, R.S. (2008) “Antisymmetry and the Lexicon,” Linguistic Variation Yearbook, 8, 1-31 (also in A.M. di Sciullo and C. Boeckx (eds.) The Biolinguistic Enterprise: New Perspectives on the Evolution and Nature of the Human Language Faculty, Oxford University Press, London, 329-353 (2011)) (reprinted in Kayne 2010). Kayne, R.S. (2010) Comparisons and Contrasts, Oxford University Press, New York. Kayne, R.S. (2014) “Once and Twice”, in Inquiries into Linguistic Theory and Language Acquisition. Papers offered to Adriana Belletti, C. Contemori and L. Dal Pozzo (eds.), Siena, CISCL Press (also in Studies in Chinese Linguistics (2015) 36, 1-20). Kayne, R.S. (to appear) "A Note on Some Even More Unusual Relative Clauses" in L. Bailey and M. Sheehan (eds.) Order and Structure in Syntax, Language Science Press. Koopman, H. and A. Szabolcsi (2000) Verbal Complexes, The MIT Press, Cambridge, Mass. McCawley, J.D. (1998) “Adverbial NPs: Bare or Clad in See-Through Garb?,” Language, 64, 583-590. McCloskey, J. (2000) “Quantifier Float and Wh-Movement in an Irish English,” Linguistic Inquiry, 31, 57-84. Myers, S.P. (1987) Tone and the Structure of Words in Shona, Doctoral dissertation, University of Massachusetts, Amherst, Mass. Obenauer, H.-G. (1994) Aspects de la syntaxe A-barre. Effets d’intervention et mouvements des quantifieurs, Thèse de doctorat d’Etat, Université de Paris VIII. Pollock, J.-Y. (1989) “Verb Movement, Universal Grammar, and the Structure of IP,” Linguistic Inquiry, 20, 365-424. Riemsdijk, H. van (1978) A case study in syntactic markedness, The Peter de Ridder Press, Lisse, Netherlands. Rizzi, L. (1988) “Il sintagma preposizionale,” in L. Renzi (ed.) Grande grammatica italiana di consultazione. Volume I. La frase. I sintagmi nominale e preposizionale, Il Mulino, Bologna, 507531. Roberts, I. (2017) “Verb Movement and Cartography in English and Romance (and elsewhere),” talk given at Yale University.