The Content and Acquisition of Lexical--论文代写范文精选

根据最近的思想，在学习的本能上,动物行为学是一种知觉学习的神经基础,涉及内在约束和环境之间的复杂性。这支持最近的激进思想，收购概念需要天生的机制,但不要求本身是与生俱来的。

Abstract
This thesis aims to develop a psychologically plausible account of concepts by integrating key insights from philosophy (on the metaphysical basis for concept possession) and psychology (on the mechanisms underlying concept acquisition). I adopt an approach known as informational atomism, developed by Jerry Fodor. Informational atomism is the conjunction of two theses: (i) informational semantics, according to which conceptual content is constituted exhaustively by nomological mind–world relations; and (ii) conceptual atomism, according to which (lexical) concepts have no internal structure. 

I argue that informational semantics needs to be supplemented by allowing content-constitutive rules of inference (“meaning postulates”). This is because the content of one important class of concepts, the logical terms, is not plausibly informational. And since, it is argued, no principled distinction can be drawn between logical concepts and the rest, the problem that this raises is a general one. An immediate difficulty is that Quine’s classic arguments against the analytic/synthetic distinction suggest that there can be no principled basis for distinguishing content-constitutive rules from the rest. I show that this concern can be overcome by taking a psychological approach: there is a fact of the matter as to whether or not a particular inference is governed by a mentally-represented inference rule, albeit one that analytic philosophy does not have the resources to determine. 

I then consider the implications of this approach for concept acquisition. One mechanism underlying concept acquisition is the development of perceptual detectors for the objects that we encounter. I investigate how this might work, by drawing on recent ideas in ethology on ‘learning instincts’, and recent insights into the neurological basis for perceptual learning. What emerges is a view of concept acquisition as involving a complex interplay between innate constraints and environmental input. This supports Fodor’s recent move away from radical concept nativism: concept acquisition requires innate mechanisms, but does not require that concepts themselves be innate.

Introduction
Concepts may be thought of as coming in two kinds: lexical and phrasal. Lexical concepts are (approximately) those corresponding to monomorphemic lexical items.1 Phrasal concepts are all the rest (that is, those corresponding to morphologically complex lexical items and phrases). Classical theories of concepts also come in two kinds: empiricist and nativist. Empiricists and nativists agree about the nature of phrasal concepts, but they disagree on the nature of lexical concepts. For both empiricists and nativists, phrasal concepts are typically complex (they have internal structure just as the phrases corresponding to them do).2 Nativists, following the analogy of the lexical/phrasal distinction, consider lexical concepts to be typically atomic (they have no internal structure, just as the corresponding lexical items do not).3 Empiricists, on the other hand, consider that in addition to phrasal concepts, the majority of lexical concepts are also complex. For empiricists, lexical concepts are typically definable and have a complex structure corresponding to their phrasal definition (so the concept corresponding to the word ‘bachelor’ would be UNMARRIED MAN). The small set of lexical concepts that cannot be defined have no internal structure and are said to be primitive. The empiricists generally consider the primitive concepts to be exclusively sensory (apart from a small number of logical connectives); nativists clearly have to allow for other kinds of primitive concept, since for them nearly all lexical concepts are primitive.

Next consider concept acquisition. It has been common ground between classical empiricists and nativists that learning a concept is an inductive process. Essentially, learning the concept XYZ involves formulating and testing hypotheses of the form ‘x is XYZ iff x is …’, where ‘…’ is a specification of what it is to be an XYZ. While this is reasonable for complex or definable concepts (for example, ‘x is a BACHELOR iff x is an UNMARRIED MAN’), it clearly cannot account for the acquisition of primitive or undefinable concepts (since we would already have to have the concept we were trying to acquire in order to frame the hypothesis: consider ‘x is RED iff x is RED’). It follows, for both empiricists and nativists, that primitive concepts must be innate— or at least unlearned in the sense described above. The difference, of course, is in the nature and number of primitive concepts. 

The empiricist claim is that a small number of sensory concepts (plus a few logical connectives) are innate. Nativists, on the other hand, seem to be forced to adopt a rather radical position: that virtually all lexical concepts are innate. This position is known as ‘radical concept nativism’. Nativists do not claim that experience plays no role in concept acquisition; but they see concepts as being triggered by experience in a brute-causal way (rather than learned from experience in a rational-inductive way). In an influential 1981 paper, “The present status of the innateness controversy”, Jerry Fodor presented arguments against the classical empiricist position on concepts. He pointed out that the available evidence strongly suggested that the majority of lexical concepts have no internal structure. This was the reason, in his view, for the failure of analytic philosophy. It also explained why it had proved almost impossible to come up with plausible examples of definitions (let alone, as Empiricism would require, definitions of non-sensory terms in a purely sensory vocabulary).

It is important to be clear that in arguing for meaning postulates as mentallyrepresented content-constitutive rules of inference, I am not claiming that all mental inference is governed by meaning postulates—in particular, reflective thinking often requires applying rules that we do not have an intuitive grasp of (modus tollens, say). Also, I am not claiming that all concepts have meaning postulates attached—again, reflective thinking often involves concepts that are not deployed in intuitive reasoning (theoretical or religious concepts, say). In chapter 4 we looked more closely at the distinction between intuitive and reflective beliefs, and the corresponding distinction between intuitive and reflective concepts, as developed by Dan Sperber. Intuitive concepts include concepts for objects which can be identified by our perceptual processes, as well as additional (generally more abstract) concepts required for the representation of meaning postulates. Reflective concepts occur in metarepresentational contexts and are thereby insulated from our intuitive thought processes. 

The thesis went on to consider what the view of conceptual content set out above could tell us about concept acquisition. Notoriously, Fodor has in the past endorsed the radically nativist position that all lexical concepts are innate. The informational atomism that he has developed, based on an externalist approach to content, allows him to move away from this radical position. Rather than requiring that all lexical concepts be innate, informational atomism requires only that certain mechanisms are innate—those mechanisms, that is, which are needed to ensure that we reliably get locked to the property of Xness in the presence of stereotypical Xs. In chapters 4 and 5, I looked at one way of cashing out this notion of ‘reliably getting locked to the property of Xness’. One obvious mechanism (but not the only one) that would ensure this is an inbuilt disposition to acquire detectors for the objects that we encounter. 

The acquisition of detectors could also explain how a 226 reflective concept (acquired, say, via communication) could become intuitive. I investigated in some detail how the acquisition of perceptual detectors might work.42 In doing so, I drew on recent work in ethology on ‘learning instincts’, and recent work on the neurological basis for perceptual learning. What emerged was an understanding of how one kind of mechanism of the type suggested by Fodor might work. We saw that perceptual detectors involve a complex interplay between innate constraints and environmental input. Thus, in some limited cases (SNAKE, say), it may be that we do possess innate detectors, and hence have the corresponding concepts innately. In most cases (animal kind concepts at the basic level, say), the cues that are diagnostic for kind discrimination are inbuilt and only the values for these cues must be learned. In other more exceptional cases (such as occurs in the development of perceptual expertise), acquiring detectors involves learning both the cues and the cue values. In support of Fodor’s position, then, the acquisition of perceptual detectors requires innate mechanisms (such as those responsible for directing attention to cues), but does not require that concepts themselves be innate. (A question that we raised, but did not discuss in detail, was whether meaning postulates should all be seen as innate, and if not what process might underlie their acquisition.)

What, then, are the prospects for informational atomism as a theory of concepts? Jerry Fodor once commented that inferential role semantics “having once got a reasonable story about ‘and’, took it for granted that ‘tree’ would submit to much the same treatment. But it doesn’t.”43 It is reasonable in my view to make a symmetrical comment concerning informational atomism. Fodor, after all, has a 42 I also noted that some detectors would operate at the conceptual rather than perceptual level. This would be true, for example, of modality-independent detectors that operated by integrating the outputs of modality-specific perceptual detectors. Some process of this kind must underlie our ability to take both face and voice cues into consideration when identifying a familiar person, say. 43 Personal communication, 2000. 227 pretty good story to tell about ‘tree’, but informational atomism requires that this story extend also to ‘and’. In my view it doesn’t. The reasons why it doesn’t, and what form a possible solution could take, have been the primary preoccupation of this thesis.（essay代写)

51Due网站原创范文除特殊说明外一切图文著作权归51Due所有；未经51Due官方授权谢绝任何用途转载或刊发于媒体。如发生侵犯著作权现象，51Due保留一切法律追诉权。（essay代写)
更多essay代写范文欢迎访问我们主页 www.51due.com 当然有essay代写需求可以和我们24小时在线客服 QQ:800020041 联系交流。-X（essay代写)