Journal of Philosophy; available online at au/lib/sta /slezak_peter/Slezak_Devitt_2007.pdf>
Smith B. C.. Lepore E., Smith B. C.. ‘What I Know When I Know a Language’, The Oxford
Handbook to the Philosophy of Language, 2006OxfordOxford University Press(pg. 941-82)
Google Scholar
Smith B. C.. ‘The Nature of “Linguistic Reality”: A Response to Devitt's Dodging’
Smith N.. ‘Foreword’ to N. Chomsky, 2000, New Horizons in the Study of Language and
Mind, 2000CambridgeCambridge University Press
Google Scholar
Stanley J.. ‘Context and Logical Form’, Linguistics and Philosophy, 2000, vol. 23 (pg. 391-434)
Google Scholar
Crossref
Tversky A., Fox H.. ‘Ambiguity Aversion and Comparative Ignorance’, Quarterly Journal of
Economics, 1995, vol. 10 (pg. 585-603)
Wason P. C.. Foss B. M.. ‘Reasoning’, New Horizons in Psychology, 1966Harmondsworth,
UKPenguin(pg. 135-51)
Google Scholar
Wimmer H., Perner J.. ‘Beliefs about Beliefs: Representation and Constraining Function of
Wrong Beliefs in Young Children's Understanding of
Deception’, Cognition, 1983, vol. 13 (pg. 103-28)
Google Scholar
Crossref
PubMed
.
09.03.2020
Linguistic Intuitions | The British Journal for the Philosophy of Science | Oxford Academic
https://academic.oup.com/bjps/article/61/1/123/1451363
42/45
1 Though I disagree with his view, Devitt is the first philosopher since Katz (
[1981]
) to tackle
the topic of linguistic intuition directly and is responsible for bringing to light many
interesting issues concerning speakers’ intuitive judgements about their language.
2 C-command relations are part of the explanation of a wide range of grammatical
phenomena.
3 For a famous investigation of these issues, see (Chomsky
[1981]
). The principle was
classically stated in terms of the notion of a domain (see
Chomsky [1987]
, p. 185). The
domain of a constituent X is the part of the structure that immediately contains X, roughly
‘the smallest phrase containing it’. So X will c-command constituents in its domain.
Principle A of classical Government and Binding Theory states that a reflexive must be
bound by an antecedent in its domain. Principle A is now stated in terms of locality (see
Chomsky [1995]
). A reflexive must be bound by a local antecedent and cannot be bound by
elements that are too ‘far away’ structurally. Conversely, pronouns cannot be bound by
elements that are too ‘close by’. The notion of locality is a highly theoretical one: there is a
whole branch of grammatical theory that focuses of the lower and upper bounds on the
distance that may separate two elements that enter into a grammatical relation.
4 This simplistic bracketing is in fact totally inadequate with respect to grammatical theory,
in terms of (a) its exocentric labelling, (b) principles of verbal theta assignment, (c) the
absence of tense, and (d) the absence of the copying that pervades even the simplest
linguistic structures. A more realistic structure would be: (1b) [TP I [T′ Φ [vP <I>[v v+<shave>
[VP shave myself]]]]] (see
Collins [2007]
for further elaboration and discussion of issues (a)–
(d)).
5 It has a coherent, independent semantic interpretation, can be replaced by constituents,
can occur in object position and can be cle ed as in ‘It was the man I saw that you wanted
to meet’.
6 More generally, the status of the binding principles as part of the language faculty is still
undecided. Chomsky has argued that binding may fall outside the narrow language faculty
because it requires principles that apply to global properties of the derivation (see
Safir
[2004]
).
7 As we shall see when I present Devitt's model in Section 4, the linguist's concern with
marginal intuitions shared by all native speakers, linguist, and non-linguist, does not sit well
with the significance that he a ords the distinction between the intuitions of the theorist
09.03.2020
Linguistic Intuitions | The British Journal for the Philosophy of Science | Oxford Academic
https://academic.oup.com/bjps/article/61/1/123/1451363
43/45
and the intuitions of ordinary speakers.
8 This is why Devitt (
[2006a]
) is wrong to suggest that we should be relying on intuitions
about grammaticality. Ordinary speakers have no intuitions about the notion in the sense
relevant to a scientific theory of language. (See
Radford [1988]
, p. 13, for a similar point.)
9 Though only a little is known about how language is realized in the brain, some aspects of
language may be relatively localized. There are two areas of the brain's le hemisphere that
appear to be particularly implicated in our linguistic abilities: Broca's area and Wernicke's
area. We know that these areas are important because sometimes they are a ected by
some pathology or trauma, and particular linguistic deficits arise. For example, people with
trauma to Broca's area can have problems with complex grammatical constructions
though their pronunciation may be normal.
10 Chomsky rejects Quine's comparison because he thinks that grammars are natural
objects. On Chomsky's view, they are psychologically, and ultimately biologically, realized.
He thinks there is something real in the mind/brain, a particular procedure for assigning
information about sound, structure, and meaning to expressions. The choice of a theory
will then be no more arbitrary than that in any other empirical inquiry.
11 See, for example,
Wimmer and Perner's ([1983]
) work on ‘Sally-Anne’/false belief tests.
12 See (
Tversky and Fox [1995]
) for the employment of subjects’ intuitions to investigate the
psychology of risk, prediction, and probability. See (
Wason [1966]
) for use of subjects’
intuitive judgements to investigate our reasoning and (
Cosmides and Tooby [2005]
) for the
use of subjects’ intuitive responses to argue that we have a specialized reasoning module
for cheater-detection.
13 See (
Knobe [2005]
) who describes how by systematically varying aspects of presented
vignettes, researchers seek to determine which factors are influencing subjects’ intuitions
and explore moral cognition.
14
Fiengo ([2003]
) makes a distinction between linguistic ‘intuition’ as that access we have to
the structures of sentences that involves no conscious reasoning and linguistic
‘judgement’ as our evaluation of that to which we have the immediate access. I model
linguistic judgements as involving a straightforward report on what we are aware of
rather than evaluating the intuited properties. Of course, it might be that informants do
both. But it seems to me that there are core cases where the speaker intuits some
09.03.2020
Linguistic Intuitions | The British Journal for the Philosophy of Science | Oxford Academic
https://academic.oup.com/bjps/article/61/1/123/1451363
44/45
properties and reports on what they intuit without further evaluating the intuited
properties as such.
15 This would involve, as a component, an account of the role of the system/s of competence
within linguistic performance. For a recent discussion of the di iculties involved, see
(
Neeleman and van de Koot [forthcoming])
.
16 Devitt says of the visual system: ‘Its task is to deliver information to the central processor
of what is seen … information that is indeed the immediate and main basis for judging
what is seen.’ (
[2006a]
, p. 112). It may be inaccurate to say that the visual module delivers
an impression of what is seen to the central processor. Depending on how ‘what is seen’ is
to be understood, it may be that the visual module delivers something shallower than
that.
17 The idea that the operations of a cognitive system are mandatory is o en associated with
the idea of a Fodorian module (
Fodor [1983]
). Though the parser may behave like a
module,
Collins ([2004]
) argues persuasively that it is wrong to think of the competence
system itself as a module in this sense.
18 Devitt (
[2006a]
, Chapters 8–10) argues that these grammatical rules are psychologically
real to the extent that they are rules of a more general language of thought. I cannot
assess this interesting claim here.
19 Notice how this conflicts with the point about marginal data in Section 2.1.
20 One could easily expand this point about optional ‘that’ by considering ‘that-trace’ e ects.
Compare: ‘Who did Bill say (that) Mary loves?’ and ‘Who did Bill say ( that) loves Bob?’.
21 Generally, Devitt holds the view that evidence for linguistic theories might come from
anywhere and we cannot circumscribe the evidence a priori. But the issue in hand
concerns the relative importance of the intuitions evidence in our current state of
knowledge.
22 A typical written corpus might be The Times or The Wall Street Journal. Spoken corpuses
are collected in acquisition studies, sometimes capturing everything a child hears and
utters over the course of months.
© The Author (2009). Published by Oxford University Press on behalf of British Society for the Philosophy of
Science. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org
*
09.03.2020
Linguistic Intuitions | The British Journal for the Philosophy of Science | Oxford Academic
https://academic.oup.com/bjps/article/61/1/123/1451363
45/45
Oxford University Press
Do'stlaringiz bilan baham: |