Is It Rhetoric, Or Is It Nonsense?

KNOWLEDGE AND PERSUASION IN ECONOMICS
Donald N. McCloskey
Cambridge University Press. xviii + 445 pgs.

This is a most peculiar book. As a glance at McCloskey's enormous bibliography suffices to reveal, our author appears to have read everything. And he isn't faking. As the text shows, McCloskey can refer learnedly to Wallace Stevens's poetry, James Gibson's psychology of perception, the Gettier problem, the strong program in the philosophy of science, and Saussure's linguistics, all this aside from mastery of the literature of most branches of economics. Is there anything McCloskey does not know?

And yet the book leaves me with an uneasy feeling, preventing me from giving it a standard review. I take the most important part of a review to be a summary of a book's main arguments. Here precisely my difficulty arises. McCloskey obviously thinks rhetoric of vital significance and wishes economists to study it closely. Yet beyond this he appears singularly elusive and I cannot attribute to him any further substantive theses at all. He brings to mind the literary criticism of Sir Arthur Quiller-Couch and Lascelles Abercrombie. Like them, he writes pleasantly and knows a great deal; but after reading them, one is puzzled: what, if anything, have they said?

Perhaps my failure to understand McCloskey stems from my own deficiencies: I learn from him that I have "difficulty understanding books outside [my]....circle" (p. 314). [See endnote] Well, I have certainly have had great difficulty understanding him! No doubt uncharitably, I am inclined to think my incompetence an insufficient explanation. Much of McCloskey's elusiveness stems from his constant efforts to make philosophical points while not having the least idea what he is talking about. He has a tin ear for philosophical argument, and his book is a veritable handbook of fallacies.

As Macaulay's schoolboy would know, McCloskey is a leading opponent of positivism, and his interest in methodology developed in part from arguments with some of his rigidly "scientific" colleagues at the University of Chicago. What sort of points did he make against them? He tells us that he challenged George Stigler, who supported "behaviorist theories of voting, in which people are said to vote according to their pocketbooks" (p. 14).

Against him, McCloskey raised the point that voting appears irrational, since a single voter has almost no chance of affecting the outcome. "The voter therefore appeared to have shown by entering the voting booth that he was nuts (by the economistic definition of nuttiness) and it would be strange if he voted according to his pocketbook with strict rationality after he closed the certain" (p. 14). In reply, Stigler was "abusively positivistic," only the observable implications of the theory mattered, he said.

Here, for once, the abusive positivist was perfectly correct. Unless the behavioral theorists contended that people vote their pocketbooks because it is rational to do so, McCloskey's invocation of the irrationality of voting is not to the point. (It's irrelevant anyway, but never mind.) And McCloskey also went astray in an argument with another Chicago positivist, Gary Baker. Against Baker, who claimed that one execution appears to deter seven murders, McCloskey "remarked that an execution was not the same as murder. . . . Execution . . . .is an elevation of the State to life-and-death power, whereas a murder is an individual's act. The two are not morally comparable" (p. 15). Assuming that Baker was arguing in defense of the death penalty, his point was presumably that the penalty's deterrent effect gives reason to support it. How does this argument depend on the assumption that murder and execution are the same? Imprisonment is also not "the same as" murder. So what? If McCloskey means that deterrence does not settle the death penalty issue, he has a much better argument; but he has utterly failed to express himself in a connected way.

McCloskey rightly criticizes what he calls "the chocolate ice cream theory, namely, that opinions about morality are mere preferences, like an uncriticizable preference for chocolate ice cream" (p. 96). But he deploys a bad argument against it. Emotivism, the view that evaluative judgments are nothing but expressions of emotion, "is of course self-contradictory; the sneer at evaluation applies to the evaluation of evaluation" (p. 97). But of course the statements that comprise the emotivist theory are not themselves evaluations; McCloskey has not grasped the elementary distinction between an evaluative statement and a statement about evaluation.

Although McCloskey has here attempted an argument from self- contradiction, he normally adopts a quite critical attitude to what he calls "the Philosopher's Friend, the rhetorical device of catching someone being committed to X at the very moment of arguing against X" (p. 200). He notes that philosophers often use this type of argument against relativism, but he is not impressed. "Even the admirable Hilary Putnam relies on it. `What relativist really thinks,' he asks indignantly `that relativism is only true- for-my subculture?'. . . .the answer, Professor Putnam, is all of them, and consistently" (p. 200).

McCloskey fails to see the point behind Putnam's question; a relativist who confines himself to saying that his group accepted relativism poses no objection at all to absolute truth. "Some people believe the earth is flat" leaves entirely untouched the issue of the earth's shape. And why must anti-relativists "confront, however, another tu quoque; that you, oh philosopher, are in turn arguing rhetorically. Gotcha yourself" (p. 200). Why should anyone be reluctant to acknowledge that he is arguing rhetorically, when McCloskey has characterized rhetoric so that it embraces all forms of communication?

For some reason unknown to me, McCloskey returns again and again to the theory of knowledge. He is interested in Gettier cases, a central topic of contemporary epistemology. In a Gettier case, one has a justified belief in a true proposition but doesn't know the proposition. McCloskey calls Gettier cases "an unresolved technical problem with the problem of justifying true belief " (p. 190), which is just wrong. The cases appear to show, rather, that justified true belief is insufficient for knowledge, as McCloskey himself recognizes later in his paragraph. Our rhetorician proceeds to offer an "economic version" of a Gettier example but muffs it, since his case involves an unjustified true belief. McCloskey has yet another argument to direct against critics of relativism. Relying on the French sociologist of science Bruno Latour, McCloskey tells us that "there does not exist a safe metalinguistic level." Unfortunately, he tells us neither what a "safe metalinguistic level" is, why he thinks anti-relativist arguments depend on one, nor how he knows one doesn't exist.

After so much fuss and feathers denying relativism the reader is startled to find this: "Rules of argument, even something as fundamental as the law of excluded middle (which is rationally set aside in some forms of logic and mathematics) are instituted by rhetorical agreement. That a statement must be either true or false and not both or neither is something we accept because it is agreed to be useful in certain classes of disputes between people. . . . it is not written in the stars" (p. 241).

McCloskey is on perfectly solid ground in asserting that some forms of logic and mathematics set aside the law of excluded middle; intuitionist logic is a leading example. But how does it follow that whether we accept the law is conventional? And why does he jumble together the law of excluded middle with the law of non-contradiction? (Perhaps McCloskey will reply by citing the work of Routley and Priest on para-inconsistent logical systems). Is the adoption of the law of non-contradiction supposed to be conventional, too? Whatever McCloskey is doing, it isn't philosophy.

NOTE

1. McCloskey gives as an instance of my misunderstanding that I think that he accepts the positivist criterion of meaning. But what I imagined I had suggested was that McCloskey thought that positivism was the only "literalist" alternative to his advocacy of metaphor. In doing so, as it then seemed to me, he remained locked in positivist categories. If McCloskey took me to be claiming that he himself accepted positivist views, I must have expressed myself very badly. Similarly, the remark about metaphor which he is kind enough to quote (p. 42), I intended to represent his views, not my own. I ought to have clearly indicated that the line of thought I suggested was a conjecture about what lay behind his anti- positivism. I have now learned my lesson and will not again attempt to attribute a coherent position to him.