Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

The MIT Press: The MIT Press Is Collaborating With JSTOR To Digitize, Preserve and Extend Access To Linguistic Inquiry

Download as pdf or txt
Download as pdf or txt
You are on page 1of 5

A Note on Marking Transparency and Opacity Author(s): Edward L. Keenan and Karen H.

Ebert Reviewed work(s): Source: Linguistic Inquiry, Vol. 4, No. 3 (Summer, 1973), pp. 421-424 Published by: The MIT Press Stable URL: http://www.jstor.org/stable/4177783 . Accessed: 09/04/2012 01:51
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

SQUIBS

AND

DISCUSSION

NOTE ON MARKING
TRANSPARENCY OPACITY AND

semantically

EdwardL. Keenan, King'sCollege Cambridge KarenH. Ebert, Sprache InstitutfiirDeutsche

It was argued in Keenan (1972) that sentences like (i) are having transparent (T) and ambiguous, opaque(0) readings according as the noun phrase the man who wonis used referentially or not. (i) Mary was surprised that the man who won was drunk.

On the T-reading, (i) says that Mary is directly surprised about some particular individual, specifically that she is surprised that he was drunk. The fact that the individual in question is the man who won is not part of the fact surprising to Mary. In other words, on the T-reading of (i), the way the individual that Mary is surprised about is identified is not relevant to Mary's surprise. Thus if it is known that John is the man who won, then (2), on its natural reading, could as well have been used to report Mary's surprise.
(2)

Mary was surprised that John was drunk.

On the 0-reading of (i), on the other hand, the information in the identifying expression themanwhowonis an essential part of the fact which is surprising to Mary. Indeed she may have no direct awareness of who the individual x 'ho actually won was. Rather, Mary is surprised that whoever won was drunk. The primary reading of (3) is the 0-reading. (3) The mere fact that the winner was drunk surprised Mary.

However, it is difficult to find constructions which unequivocally select for the 0-reading. By contrast, constructions such as (4) force the T-reading. (4) Mary was surprised that the man who won won.

In Keenan (I972) the ambiguityin (i) is accounted for by transformationally deriving it from distinct underlying structures. In English these underlying differences are lost on the way to surface. But there are languages which, at least in certain contexts, formally distinguish the two readings in surface. The purpose of this squib is to present these facts, thereby further supporting the claim that the semantic distinction in (i) is a grammatical reality, not merely a "philosophical" one. Specifically we shall present languages having distinct definite articles which function, in sentences like (i), to select either the T- or 0- readings.

422

SQUIBS

AND

DISCUSSION

First, in Fering, a North Frisian dialect of the islands of Fohr and Amrum, we find two definite articles whose forms are: masc.sg. fem.sg. neut.sg. pl. a A-article: a at at don D-article: di det det Although the use of the two articles is a complex matter (see Ebert I971 for a more thorough analysis) we can say that in general the D-article indicates that some contextual information is necessary for the hearer to identify the individual(s) or object(s) mentioned. The D-article may point anaphorically to a previously identified individual, as in (5), or cataphorically to a following relative clause that specifies the referent, as in (6). an hingst keeft. Di (5) Oki hee jister Oki has yesterday a horse bought. The hingst leept uiib stuuven. horse goes on paddock 'Oki bought a horse yesterday. The horse is in the paddock.' (6) Det moolk wat uub boosel staiint as sor. the milk that on table stands is sour 'The milk which is standing on the table is sour.' These functions of the D-article enable it to be used in sentences like (i) with both T- and 0- readings. Thus (7) is ambiguous in the same way as (i). (7) John wonert ham, dat di maan wat woon John wondered him that the man who won bisooben wiar. drunk was 'John was surprised that the man who won was drunk.' The A-article, on the other hand, is used to refer to objects whose identity can be taken as known. Thus the A-article is used with generics (the Danish people= A with objects of common knowledge deensken, *Don deensken), (the sun = a san, *di san) and with objects which can be identified by their known relation to other objects or events mentioned in the sentence or given in the speech situation, as thedoorin (8). (8) Ik fersoocht iin iiiin't hius tiigiingen, man I tried into the house to go but a door wiar ifsl66den. the door was locked 'I tried to enter the house but the door was locked.'

SQUIBS

AND

DISCUSSION

The nondeictic nature of the A-articles forces anly NP they determine to be understood opaquely in sentences like (i). Thus only the 0- reading of (g) is possible. (g) John wonert ham, dat a maan wat woon John wondered him that the man who won
bis6bDen wiar.

drunk was 'John was surprised that the man who won was drunk.' Consequently we predict, correctly, that if the A-article is used in contexts that would otherwise force a T-reading the resulting sentence is ungrammatical.
(io) *John wonert

ham, dat a maan wat woon John wondered him that the man who won woon. won 'John was surprisedthat the man who won won.'

Of course (io) is well formed if the A-article is replaced by the D-article.


(I I)

John wonert ham, dat di maan wat woon woon.

It appears, then, that to correctly account for the ungrammaticality of (io) and limited readings of (9), article selection in Fering must be sensitive to the opacitytransparency distinction. Malagasy also has two definite articles. The one, ny, has a full range of functions: it introduces restrictive relative clauses, generics, objects of common knowledge, objects of inalienable possession; and it is the normal anaphoric article. Thus (I2) has both T- and 0- readings.
(I2)

Rakoto fa mamo ny mpandresy. Gaga surprised Rakoto that drunk the winner 'Rakoto was surprised that the winner was drunk.'

However, Malagasy possessesa second definite article, ilay, whose use is narrowly restricted to objects that the hearer has specifically identified prior to the utterance. This reidentifying function need not be anaphoric in the sense that the specific identifying experience on the part of the hearer need not have been occasioned by an explicit reference to the object in the preceding discourse. The object may be something that both speaker and hearer observed together. On the other hand the function of ilay is not deictic either, in the sense that it cannot be used to pick out an object in the visual field of the speaker and

424

SQUIBS

AND

DISCUSSION

hearer. For this a demonstrative adjective must be used. If the referent of an ilay-determined NP is present in the visual field of both speaker and hearer it must still have been identified by the hearer prior to its mention in the utterance. Consequently if ilay-determined NPs occur in contexts like (i) they can only be understood transparently. So (I3) has no 0-reading but only a T-reading.
(I 3)

Rakoto fa mamo ilay mpandresy. Gaga surprised Rakoto that drunk the winner 'Rakoto was surprised that the winner was drunk.'

Conclusion: article choice in various languages can determine whether sentences like (i), ambiguous in English, are to be understood opaquely (Fering) or transparently (Malagasy).

References Ebert, K. H. (I 97i) Referenz, Sprechsituation unddie bestimmten Artikel in einem nordfriesischen Dialekt, Nordfriisk Instituut, Bredstedt. Keenan, E. L. (I 972) "On Semantically Based Grammar," Linguistic Inquiry III, 413-46I.

You might also like