Susan Haack

Philosophy of Logics:
Logic and thought

3 Logic and thought

Kant’s confidence in the unrevisability of Aristotelian logic rested on the idea that logical principles represent ʻthe forms of thought’, that we can’t but think in accordance with them: an idea that raises a host of intriguing questions about just what logic has to do with ‘the way we think’.

Although at one time it was quite usual to suppose that the principles of logic are ‘the laws of thought’ (see Boole 1854), Frege’s vigorous critique was so influential that there has been rather little support, of late, for ‘psychologism’ in any shape or form. However, Frege’s arguments against psychologism are, I suspect, less conclusive, and at least some form of psychologism more plausible, than it is nowadays fashionable to suppose. A full-scale re-assessment of psychologism would require, however, a fuller and more sophisticated account of the nature of thought than I am able to offer; so what follows can be sketchy at best.

One can begin by distinguishing—the distinction is pretty crude, but nevertheless may be serviceable as a starting-point—three kinds of position:

(i) logic is descriptive of mental processes (it describes how we do, or perhaps how we must, think)

(ii) logic is prescriptive of mental processes (it prescribes how we should think)

(iii) logic has nothing to do with mental processes

One might call these strong psychologism, weak psychologism and anti-psychologism, respectively. Examples: Kant held something like (i); Peirce a version of (ii); Frege, (iii). In what follows I shall present some arguments for a form of weak psychologism rather close to that adopted by Peirce (1930-58, 3, 161ff.): that logic is normative with respect to reasoning. I shall then go on to point out some advantages of weak psychologism as against anti-psychologism, on the one hand, and strong psychologism, on the other.

Logic is primarily concerned with arguments: how, then, can it relate to the mental processes which constitute reasoning? I’ll tackle this question in two stages, offering, first, a Platonist answer, and then a nominalist version of that answer; the reason for this strategy is that the connection between logic and thought is thrown into sharper relief by the Platonist account, but I think that it is better, though less simply, explained in the nominalist version.

The Platonist answer: Logic is concerned with the (in)validity of arguments, with the connection between premises and conclusion; logical relations are relations, such as entailment or incompatibility, between propositions. Reasoning is a (certain sort of) mental process, such as, coming to believe that q on the strength of one’s belief that p (inferring q from p), or, coming to recognise that if p were the case, then q would be the case; and to believe that p, or to wonder whether, or what if, p, is to stand in a certain relation to a proposition. Hence, logic is normative with respect to reasoning in this sense: that if, e.g., one infers q from p, then, if the argument from p to q is valid, the inference is safe, in that it is guaranteed not to result in one’s holding a false belief on the basis of a true one.

The nominalist version: that s believes that p, or wonders whether, or what if, p, can be analysed, ultimately, in terms of a complicated relation between s and the sentence ‘p’; and Platonist talk of belief in or entertainment of a proposition is to be regarded as a convenient shorthand for this complicated analysis. Logic is concerned with the validity of arguments, which, however, are to be conceived (ch 2 §1) as stretches of discourse/strings of sentences; and Platonist talk of logical relations between propositions is, again, to be regarded as a convenient shorthand (specifically, for quite complicated qualifications about what sentences are to be regarded as inter-substitutable, ch. 6 §4). Once again, it follows that logic is normative, in the sense explained above, with respect to reasoning.

The nominalist version of weak psychologism is, I think, preferable to the Platonist, for reasons which will emerge from a consideration of an argument of Frege’s against psychologism.

Frege’s objections to psychologism are quite complex, and I shall only consider the argument which is most relevant to the position I have defended} This argument runs as follows. Logic has nothing to do with mental processes; for logic is objective and public, whereas the mental, according to Frege, is subjective and private. This is why Frege is so concerned to stress (see especially Frege 1918; and cf. p. 61n above) that the sense of a sentence is not an idea (a mental entity), but a thought (Gedanke: an abstract object, a proposition). Since ideas are mental, they are, Frege argues, essentially private; you can no more have my idea than you can have my headache. [….]

[pp. 240-1: refutation of Frege’s position]

Logic, I suggested, is prescriptive of reasoning in the limited sense that inference in accordance with logical principles is safe. (Of course, safety needn't be an overriding consideration; one might, quite rationally, prefer fruitful but risky to safe but relatively uninteresting procedures; cf. de Bono’s championship, e.g. in 1969, of ‘lateral thinking’.) It is important, however, that on the weak psychologistic view, though logic is applicable to reasoning, the validity of an argument consists in its truth-preserving character; it is in no sense a psychological property. Consequently, weak psychologism avoids the main difficulty of strong psychologism, the problem of accounting for logical error: for, since people surely do, from time to time, argue invalidly, how could the validity of an argument consist in its conformity to the way we think? This isn’t to say that strong psychologism is flatly incompatible with logical mistakes; but that the two can be reconciled only by means of some explanation of such mistakes as the result of some irregularity or malfunction of our reasoning powers. (According to Kant, logical mistakes are the result of the unnoticed influence of sensibility on judgment.) Nevertheless, its much readier reconcilability with fallibilism speaks, I believe, in favour of weak over strong psychologism.

There are, inevitably, many intriguing questions this leaves unanswered: for instance, what, exactly, distinguishes logical from psychological study of reasoning? (It can't be, as is sometimes supposed, that psychology, unlike logic, is never normative, nor even that it is never normative with respect to truth; consider, for instance, psychological studies of the conditions of reliable/illusory perception.) What consequences would psychologism about logic have for questions about the relations between epistemology and psychology? What has logic to tell us about rationality? What would the consequences for psychologism be (in view, especially, of Chomsky’s claim that certain grammatical structures are innate) of the conjecture that logical form can be identified with grammatical form?

It’s good to know (to borrow a phrase of Davidson's) we shan’t run out of work!

SOURCE: Haack, Susan. Philosophy of Logics (Cambridge: Cambridge University Press, 1978), Chapter 12: Some metaphysical and epistemological questions about logic; 3: Logic and thought pp. 238-240, 241-242 [conclusion of main body of book.]

“What is the Relationship Between Logic and Reality?”
by R. Dumain

Susan Haack — An Introductory Guide

Philosophy of Paraconsistency & Associated Logics (Web Guide)

Argumentation & Controversies: Selected Bibliography