«It is a familiar philosophical idea, one that undoubtedly contains much truth, that experience makes an essential contribution to contents of ...»
The Relationship of Experience to Thought
University of Pittsburgh
It is a familiar philosophical idea, one that undoubtedly contains much truth, that experience
makes an essential contribution to contents of thoughts—that without a proper relationship to
experience, thought is empty. To isolate the truth contained in this idea we need to become
clearer on contents of thoughts and on the precise relationship of experience to these contents.
These are the goals of the present essay. The goals are large and distant, and I shall be satisfied if I manage to take a few steps in the right direction.
We need to address a preliminary question before we embark on our project: how should we conceive of thoughts? I suggest we follow the lead of Plato and Sellars and model thoughts as inner speech (or inner writing or inner typing—it does not matter which). Let us take thoughts to have logical structure analogous to that of out-loud sayings, and let us use the same logical vocabulary for thinkings and sayings. So, we can classify certain thinkings and sayings as suppositions; we can call their constituents terms; we can speak of these constituents as expressing concepts and as denoting particulars or universals; and so on. We can move back and forth between sayings and thinkings, and between constituents of sayings and constituents of thinking. Points made for one will hold, mutatis mutandis, for the other.
It may be objected that the Plato-Sellars model is too intellectual, that it denies thought to mute animals. Response: May be so, but my present project is intellectual. I am concerned to understand the logic of empirical inquiry, an inquiry that is discursive. I want to understand such things as what sorts of challenges are proper, what demands for justification are legitimate, and what connections to experience are necessary for content. For this logical inquiry, the Plato- Sellars model of thinking is, prima facie, a good starting point. By adopting the model, I am not denying the possibility of a broader conception of thought, one not so closely tied to language.
Such a conception may well be desirable, even necessary, for certain inquiries. The present inquiry The Relationship of Experience to Thought — Page 2 provides, however, no immediate reason to seek such a conception. We should not fritter away our limited resources chasing a new model of thought when an older one will serve us just as well.
We can always revisit the old model if we encounter problems and we suspect the model to be a contributing factor.
A. Conceptual Criticism
1. Men often, John Locke tells us, “set their Thoughts more on Words than Things” and “speak several Words, no otherwise than Parrots do” (Essay, III.ii.7). This parroting of meaningless words is, if not excusable, at least understandable. Words being superficially all similar, it is easy to suppose that they must be meaningful simply because we use them in familiar constructions in familiar ways.1 Superficial impressions can be misleading, however. Just as we critically assess assertions and beliefs for truth and rationality, similarly we can (and should) critically examine words and concepts for meaningfulness and rational legitimacy. This critical inquiry can be expected to issue injunctions. Locke’s own injunction was this: “A Man should take care to use no word without a signification, no Name without an Idea for which he makes it stand” (Essay, III.xi.8).
2. David Hume follows Locke in his own pursuit of the critical inquiry, and he offers us a crisp
When we entertain, therefore, any suspicion, that a philosophical term is employed without any meaning or idea (as is but too frequent), we need but enquire, from what impression is that supposed idea derived? And if it be impossible to assign any, this will serve to confirm our suspicion. (Enquiry, §2) 1 A terminological note. Philosophers associate a variety of items with words and concepts: reference, signification, sense, character, idea, and so on. Some of these associations I shall acknowledge and explain below. I reserve ‘meaning’ and ‘content’ to talk generally, and sometimes intedeterminately, about these different associations.
The Relationship of Experience to Thought — Page 3 Hume’s maxim rests on a particular conception of thought and experience. “All the materials of thinking,” Hume tells us, “are derived either from our outward or inward sentiment,” from what Hume calls ‘impressions’. Thoughts and ideas are “copies of our impressions”; they are “less forcible and lively” (Enquiry, §2). Hence, if we cannot find an original impression from which the supposed meaning of a term is derived then we have reason to suspect that the term lacks any meaning.
Hume’s conception of experience and thought is undoubtedly erroneous. Experience does not acquaint us with, nor even necessarily presents us with, impressions. “Less forcible and lively” copies of impressions may exist, but these are neither thoughts nor the primary concern of our thoughts. Still, Hume is correct in the general shape in which he casts his critical maxim.
Experience is a foundation of meaning; thoughts and words gain content, in part at least, through their relationship to experience; and a lack of proper relationship is ground for conceptual criticism. The question before us is this: what are the mechanisms through which experience endows language and thought with meaning?
3. Bertrand Russell’s views on experience and thought underwent large shifts through his long philosophical career. Nevertheless, he remained firm on a principle that links the two. In his 1912 book, Problems of Philosophy, Russell formulated the principle thus: “every proposition which we can understand must be composed wholly of constituents with which we are acquainted” (p.
58). Almost a half-century later, Russell reaffirmed his principle, but with some interesting
changes in formulation:
I have maintained a principle, which still seems to me completely valid, to the effect that, if we understand what a sentence means, it must be composed entirely of words denoting things with which we are acquainted or definable in terms of such words. It is perhaps necessary to place some limitation upon this principle as regards logical words—e.g. or, not, some, all.2 The new formulation speaks of sentences whereas the earlier one spoke of propositions. This
change is inessential in light of the background assumption (which I shall accept) that we understand a sentence iff we understand the proposition the sentence expresses.3 The new formulation is a little incautious, however. It neglects the possibility that the sentence may contain context-sensitive words such as ‘I’ and ‘you’. But the problem is easily fixed: we should understand the principle not to be about sentences but to be about particular uses of sentences.
The other change in formulation is in the qualification, tentatively put forward in the extract above, concerning logical words. In 1912, Russell was happy to countenance acquaintance with the meanings of logical words and to regard this acquaintance as a source of substantive knowledge. Later, under the influence of Ludwig Wittgenstein, he abandoned this idea. He came to believe that “logic consists wholly of tautologies.”4 If we bracket the issue of our knowledge of logic, then we can formulate Russell’s principle as follows. Russell would have accepted this formulation in 1912, I think, and also a half-century later.
Russell’s Principle. We understand what a sentence means (on an occasion of use) only if all the words composing it either are logical words or are words denoting things with which we are acquainted or are definable in terms of such words.5 Russell offered in 1912 an argument for his principle that too, I think, he would have accepted a
It is scarcely conceivable that we can make a judgement or entertain a supposition without knowing what it is that we are judging or supposing about. We must attach some meaning to the words we use, if we are to speak significantly and not utter mere noise; and the meaning we attach to our words must be something with which we are acquainted.6
Consider a word that is neither logical nor definable in terms of other words. If we use it meaningfully, if we are not uttering “mere noise,” we must attach a meaning to it. This meaning, Russell tells us, must be “something with which we are acquainted.” Russell had a foundationalist theory of meaning. According to this theory, certain words are basic—namely, some logical words and words that denote things with which we are acquainted. The meanings of the rest of the words, and of expressions, can be reduced to the meanings of basic words. A meaningful sentence, in particular, denotes a proposition which, apart from logical items, is “composed wholly of constituents with which we are acquainted.” Note also that Russell had a monistic theory of meaning. Russell recognized only one semantical relation: denotation. Meaning at the basic level, according to Russell, is to be identified with denotation; and understanding meaning is to be identified with acquaintance with the denotation. A basic non-logical word that denotes nothing is meaningless; it is “mere noise.”
4. Russell’s Principle is in the right form to serve as a foundation for conceptual criticism. We can use it to show that an expression is meaningless if we can show that one of the non-logical words composing it neither denotes anything with which we are acquainted nor is definable in terms of such words. The crucial question for applying the principle is this: what precisely are the items with which we are acquainted? Now, Russell held, early and late, that experience acquaints us with some items. The Russell of 1912 also held that we have a Platonic, non-experiential acquaintance with universals. The scope of this non-experiential acquaintance is plainly crucial for applications of the principle. So, what is its scope? How do we tell whether we have acquaintance with a purported universal or not? Russell’s answer here blunts completely the critical force of his principle: “among universals, there seems to be no principle by which we can decide which can be known by acquaintance.”7 It follows that no principled criticism is possible that an expression is meaningless. In a debate over the meaningfulness of a term, it would always be open for the defenders of the term to invoke Platonic acquaintance with a universal, and thus bring the debate to an impasse.
5. The scope of Russellian acquaintance was sharply narrowed in the theories of the early Logical
Positivists. Through definitions, Rudolf Carnap tells us, “every word of the language is reduced to other words and finally to the words which occur in the so-called ‘observation sentences’ and ‘protocol sentences’. It is through this reduction that the word acquires its meaning.”8 We have here a transformation of Russell’s Principle into a powerful critical tool. Basic non-logical words to which all other words are reducible extend no further than those that occur in observation sentences. The meanings of these words we learn through experience; in effect, experience acquaints us with their denotations.
Now, Carnap acknowledges that the scope of observation sentences is debatable.
Nevertheless, logical analysis of meaning, he argues, “pronounces the verdict of meaninglessness on any alleged knowledge that pretends to reach above or behind experience” (76). What falls in
this realm of meaninglessness? Carnap’s response is bold:
In the domain of metaphysics, including all philosophy of value and normative theory, logical analysis yields the negative result that the alleged statements in this domain are entirely meaningless. Therewith a radical elimination of metaphysics is attained. (61) Russell’s Principle is, thus, deployed to excise a large body of our discourse as meaningless. I should stress that the discourse excised is held to lack only descriptive or cognitive meaning.
Carnap allows that it may possess some other kind of meaning, say, expressive meaning. It may, in Carnap’s words, “serve for the expression of the general attitude of a person towards life” (78). But, even here, Carnap is scathing about metaphysics: “metaphysicians are musicians without musical ability” (80).
6. Conceptual criticism reached its zenith in the early Positivists. Its subsequent decline was steep and swift. The critique that delivered the verdict of meaninglessness for metaphysics was applicable also to science, an enterprise the Positivists revered. Scientific terms, like metaphysical ones, could not be reduced to observation terms; if metaphysics was meaningless, so also was science.