فلسفه‌ی زبان موریس: سه؛ راسل

در این فصل به راسل و نظریه‌ی توصیفات معینِ او پرداخته می‌شود. مرجع اصلی این فصل این مقاله از راسل است:

‘On Denoting’

 

قسمت یک: مقدمه

As we saw, Frege’s committed to these two claims:

(F3) Ordinary proper names and definite descriptions are singular terms;

(F4) Ordinary proper names and definite descriptions all have Sense.

And the crucial things about the notion of a ‘singular term’ used in (F3) are these:

(ST1) The business of a singular term is to refer to an object;

(ST2) A sentence containing a singular term has no truth-value if there is no object corresponding to that singular term.

Russell was generally suspicious of the notion of Sense. The dramatic proposal of Russell’s great paper of 1905, ‘On Denoting’, is that definite descriptions do not in fact refer to objects. That is, Russell denies (F3), at least for the case of definite descriptions. And this, he thinks, will enable him to deny (F4), and avoid appealing to the notion of Sense altogether.

 

قسمت دوم: مشکلات

Russell treats Frege’s notion of Sense as a theoretical notion designed to deal with certain problems. His claim is that these problems are dealt with better by his alternative theory, which denies both (F3) and (F4). The problems are these:

(P1) How identity statements can be both true and informative;

(P2) How there can be a difficulty in swapping within epistemic contexts words which have the same ordinary reference;

(P3) How something meaningful can be said using ordinary proper names and definite descriptions which refer to no existing objects.

Russell’s theory is designed to solve all of these problems, and two more besides:

(P4) How the Law of Excluded Middle applies to sentences including such phrases as ‘the present King of France’ (It is now customary to distinguish between the Principle of Bivalence – which says that every meaningful sentence has exactly one of the two truth-values – and the Law of Excluded Middle – according to which every instance of the schema ‘p or not-p’ is true. Russell did not distinguish between the two, and it is arguable that his real concern was closer to what we now as the Principle of Bivalence.);

(P5) How there can be true denials of the (apparent) form ‘N does not exist’.

Problems (P4) and (P5) cannot be solved by introducing the notion of Sense, if (ST2) is still maintained.  And it’s clear that Frege himself accepted (ST2). But could we, in fact, modify (ST2)? What if we distinguish between existent and no-existent objects, and allowed that singular terms might refer to non-existent objects? (This view is associated with Alexius Meinong.) Russell himself was impatient with any suggestion that there might be objects which don’t really exit. In such theories, he says, ‘there is a failure of that feeling for reality which ought to be preserved even in the most abstract studies.’

 

قسمت سوم: راه حل راسل به صورتی موجز

At the core of Russell’s solution to the problems he sets himself is a revolutionary approach to the structure of language: superficial similarities between types of sentence should not be taken as evidence for thinking the sentences really work in the same way. Traditionally definite descriptions had been assimilated to the class of proper names. Russell begins by assimilating them instead to a quite different class of expressions: what he calls ‘denoting phrases’. The examples are these: ‘a man’, ‘some man’, ‘any man’, ‘every man’, ‘all men’. All these phrases involve quantifiers. A quantifier is an expression which specifies some quantity of a given group. Russell is concerned with this sentence:

(5) The present King of France is bald.

According to him, (5) means this:

(5r)There is exactly one object which is now King of France, and that object is bald.

   Russell sometimes expresses this in slightly more complicated way, as being equivalent to the following combination of sentences:

(5r*) (1) There is at least one object which is now King of France;

          (2) There is at most one object which is now King of France; and

          (3) Whatever is now King of France is bald.

On his analysis, the phrase ‘the present King of France’ is not a name of the present King of France. And the phrase has no meaning on its own, although it’s meaningful in the context of a sentence.

 قسمت چهارم: استراسون و توصیفات معین

Does Russell give a correct account of the meaning of sentences involving definite descriptions? In his famous paper, ‘On Referring’, P. F. Strawson argues that he does not. Strawson claims that certain crucial semantic terms are not properly applied to sentences, words, and phrases (expressions) at all: they are only properly applied to uses of expressions. He uses the distinction between the kinds of thing that can properly be said of linguistic expressions, on the one hand, and the kinds of things that can properly be said of uses of expressions, on the other, to remove some of the motivation for Russell’s theory. In particular, we cannot insist that either a sentence or its negation must be true if it’s meaningful. It is not even obvious that we should insist that if a sentence is meaningful, any use either of it or of its negation must be true.

Strawson’s overall claim can be put like this. There are certain central uses of definite descriptions which are referring uses. If a use of a definite description is a referring use, then the person who uses the description does not assert the unique existence of an object which satisfies the description, as Russell thought. Instead, the unique existence of an object which satisfies the descriptions is presupposed.

 

قسمت پنجم: دیدگاه دانلان

So far we have a fairly simple picture. Frege thought of definite descriptions, in all uses, as singular terms, and hence as referring expressions. Russell claims that they were never referring expressions. Strawson adopts something close to Frege’s view for some uses. In ‘Reference and Definite Descriptions’, Keith Donnellan complicates this simple picture. He claims that there are both referential and non-referential uses of definite descriptions: in this he disagrees with both Frege and Russell. But he claims that the referential uses do not rest on the presupposition that there is exactly one thing which satisfies the description: in this he disagrees with Frege and Strawson. Moreover, whether a use is referential does not depend on the general form of the sentence in which it occurs, or on what we would ordinarily say if we heard it, but on the intentions of the speaker who uses it.

According to Donnellan there are two uses of definite descriptions: attributive use according to which the definite description is essential to what I want to say, and referential use which is not essential. In the latter case I simply want to say something about that person, and using the description is just one of a number of possible ways of making it clear which person is involved.

We might be tempted to accept Donnellan’s distinction, and give a Russellian account of the attributive uses and a Strawsonian account of the referential uses. But this is exactly what Donnellan does not want to do: he thinks that Russell’ account is wrong for the attributive uses, and Strawson’s is wrong for the referential uses.

 

قسمت ششم: دفاعیات راسل

The Russellian will attempt to meet the criticisms brought by Strawson and Donnellan by making two distinctions:

(1)    Between what is strictly true or false and what is helpful or unhelpful in a conversational context;

(2)    Between what is strictly and literally said in a use of a sentence, and what a speaker means in uttering it.

The difference between the Russellian view and its critics are manifestations of a deeper difference between their approaches to the philosophy of language. On the one hand, we can think of human languages as like very complicated machines: the task of the philosopher is then to understand how they tick. We might call this the mechanical conception of language. If we take this view, we will think that the operation of human language is governed by laws which are similar in status to the laws of physics. Our inclinations will be to look for the uniform explanations of variety of phenomena, with particular variations being due to variations in local circumstances. The Russellian approach to definite descriptions broadly fits this model.

Strawson and Donnellan, however, take a different kind of view. They do not suppose that there is a system which can in any way be compared to a machine whose operation we need to understand. They expect to find nothing deeper than the complications of everyday life: their concern is just to be true to those complications. There is no virtue for them in uniformity of explanation, just as such: they are reluctant to generalize too quickly; they expect there to be exceptions to every rule. For them, in the end, the philosophy of language is not concerned with understanding how language ticks, but with what people are doing when they speak.

 

قسمت هفتم: منابع بیشتر

1-      ‘On Referring’, P. F. Strawson

2-      ‘Reference and Definite Descriptions’, K. Donnellan

3-      ‘Descriptions’, S. Neale

4-      ‘Russell’, M. Sainsbury

5-      ‘Introduction to Mathematical Philosophy’, ch.16, B. Russell

6-      ‘The Philosophy of Logical Atomism’, B. Russell