Objective and Subjective: Sentences without Performatives Author(s): Dwight Bolinger Reviewed work(s): Source: Linguistic Inquiry, Vol.

4, No. 3 (Summer, 1973), pp. 414-417 Published by: The MIT Press Stable URL: http://www.jstor.org/stable/4177781 . Accessed: 09/04/2012 01:50
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

The MIT Press is collaborating with JSTOR to digitize, preserve and extend access to Linguistic Inquiry.

http://www.jstor.org

414

SQUIBS

AND

DISCUSSION

(32) a. *Mesto, byt' v kotorom ja ne xocu place to be in which(loc.) I not want odinokim sejcas, nedaleko ot nas. lonely(inst.) now not far from us b. Mesto, byt' odinokim v kotorom ja ne xocu sejcas, nedaleko ot nas. 'The place which I don't want to be lonely in now is not far from us.' In all the above examples, the instrumental complement of the verb must prepose with it. What seems to distinguish these sentences from the rest is the fact that the instrumentals do not denote independently existing entities but rather attributes of previously mentioned ones.2 Thus in (29), trusom 'coward' denotes the same entity as celovek 'man' and
in (32) odinokim'lonely' refers toja 'I'.

While the sentences presented here exhibit substantial regularities, drawing any detailed conclusions from them is probably unwarranted. They do show that various semantic factors play a role in limiting the usually quite free word order of Russian. Semantic case, the independent existence of an NP's referent, and perhaps the closeness with which the verb implies instrumental NPs are some of these factors. However, the dubious status of these sentences demands caution. Although all informants, especially Soviets, had clear and coinciding intuitions about their "grammaticality", all emphasized that no Russian would ever use any of them. The regularities these sentences exhibit may well be quite indirect reflections of some inner, and as yet undiscovered, semantic constraints on Russian word order. What is needed is more direct evidence of these constraints. Reference Ross, J. R. (I967) Constraints on Variablesin Syntax, unpublished Doctoral dissertation, MIT, Cambridge, Mass. [Distributed by Indiana University Linguistics Club, Indiana University, Bloomington.]

OBJECTIVE

AND SENTENCES

SUBJECTIVE:

WITHOUT PERFORMATIVES

Dwight Bolinger, HarvardUniversity

A nice question about performatives is how one can express the idea that something asserted is in the nature of things, above and beyond the assertion. When a speaker is there truly an underlying persays john is a preacher, formative such that I say toyou thatJohn is a preacher is an equivalent sentence ? Interchanges like (I)
2 I am indebted

to Wayles Browne for this insight.

SQUIBS

AND

DISCUSSION

(i)

"John is a preacher." "Says you." "No, I don't say it; he is it."

suggest that as far as the speaker is concerned, if a performative is present in a sentence of this type it is apt to be-perhaps has to be explicit. Of course explicitness comes in degrees. Shall we call the exclamatory intonation of Jfohnis a preacher! an explicit performative? It is physically present, and it shows the speaker's assertive intent. What of the modal in John must be a preacher? These and others like them are given overtly and would seem to qualify. But there is evidence of a wide range of sentences which behave as if a performative were present and yet in which none is present, at the surface. Their most interesting quality is that they seem to show the speaker's awareness that sometimes what he says depends on his view of things. His knowledge of when he is being objective and when he is being subjective is revealed in the form of his sentence. The need of an explicit performative in some cases can be shown where a context has been rigged to depend on an inference or judgment, which then has to be brought out. In the following, the "objective" sentences, the ones without performatives, are unacceptable:
(2)

a. *His music was symphonic to impress (to have impressed) that audience. b. I claim his music was symphonic to have impressed that audience. c. His music had to be symphonic to impress that audience. d. His music was symphonic, by God, to have impressed that audience. e. His music was obviously symphonic, to have impressed that audience. f. His music was symphonic, to have impressed that audience! 1 g. *George was a lawyer, to have convinced them. h. I claim George was a lawyer, to have convinced them.

Besides this intonation, which might be drawn as phon mu sic His was sym ic to ...

4I6

SQUIBS

AND

DISCUSSION

i. *The house was brick, for them to have wanted to buy it. j. I claim the house was brick, for them to have wanted to buy it. k. *You wrestle, to have bruised him so badly. 1. I claim you wrestle, to have bruised him so badly. Nevertheless there are matching sentences with no explicit performative, which are acceptable: (3) a. His music was brilliant to impress (to have impressed) that audience. b. George was a master, to have convinced them. c. The house was sturdy, for them to have wanted to buy it. d. You stink, to have bruised him so badly.

Adding an explicit performative may improve things somewhat, but it is not essential. What do they have in common that makes it unnecessary? All of them contain degree words, which is to say that all are intensifiable; and by the same token all are value judgments, and hence
subjective. To call a person a fool, or good, or tall, or talented, or to say that he fudged, or goofed, or flattered, or glared, is not precisely to say what he is or does, but to add a subjective judgment to the objective fact. The underlying judgment amounts to an unexpressed performative, and suffices to make a sentence acceptable that otherwise would not be unless a performative were added explicitly. As a tentative generalization, any sentence not containing an explicit performative or a degree expression
There are others implying some kind of repetition, e.g. phon sym ic, was sic mu His His mu sic was sym phon The first might be used in a situation where I'm tellingyou again or In spite of all you say, I insist would be appropriate. The second suggests something like How can you maintain otherwise?Both are intonational performatives. ic, to... to...

SQUIBS

AND

DISCUSSION

does not contain any performative at all, from the speaker's standpoint. There is no underlying performative in John is There is one in John is an idiot. a preacher.

NOTE ON "IDENTITY CONSTITUENTS" 1

OF

Peter Eisenberg, Fre4e UniversitdtBerlin

Fodor (I970, 429) noticed that (i) is a grammatical sentence although the proform do so refers to the transitive verb melt which is not contained in the structure underlying (I): (i) Floyd melted the glass though it surprised me that it would do so.

The transformation which introduces do so can apply in spite of the syntactic differences, apparently because melt (transitive) and melt (intransitive) are phonologically identical. From this observation Fodor (1970, 437) concludes "that the widely held view that such [pro] forms enter into surface structure only as a result of deletion under identity ... may be false". By presenting some data from German I want to show that there seem to be good reasons for refining the notion "Identity of Constituents". More precisely: I will argue that syntactic rules and in particular some rules for deletion under identity cannot refer only to structural, lexical, and referential, but also to phonological, identity of elements. Once this is accepted as a possible condition on transformations it would not be necessary to follow Fodor's conclusion, but rather to use the notion "identity" in a more differentiated way. This would mean that sentences like (i) could be handled in a natural way within a transformational grammar. One of the cases where each one of two identical elements from coordinate structures can be deleted is the case of conjoined subordinate clauses in German. So from (2a) we can derive (2b) and (2c).
(2)

a.

b. c.

weil Hans Bier 'because Hans milk' weil Hans Bier weil Hans Bier

trinkt und Franz Milch trinkt drinks beer and Franz drinks und Franz Milch trinkt trinkt und Franz Milch

Now consider (3): (3) a. weil ich Bier trinke und du Milch trinkst b. *weil ich Bier und du Milch trinkst c. weil ich Bier trinke und du Milch
1 I am gratefulto BarbaraH. Parteefor severalsuggestions and for bringingto my attentionexample (i i).