Verificationism and the self-defeating argument

Millie asked:

Who was it that first pointed out that the verification principle fails to stand up to its own criterion?

Answer by Graham Hackett

One of the most famous statements about verificationism is that it is self defeating under its own terms. A statement can only be verified, (according to verificationism), if it is either an analytic statement (self-evident) or shown to be true by scientific proof. However, we can argue that the statement of the verification principle is neither analytic, nor supported by scientific proof. Therefore it is self defeating. We need to be careful in just baldly stating this as an anti-verificationist argument, since Ayer and Carnap, and other empiricists attempted changes in definition in response to the self-defeating argument. We would need to check whether all definitions have the same weakness. 

In not answering your exact question about the origin of the self-defeating argument (SDA), I find that it it is mentioned many times by critics, but none of them admit to its first formulation. My own view was that it was kicked around by the Vienna school for some time, asserted, refuted and re-asserted. The SDA  may well have even been first formulated by a verificationist! The nearest I can get to a meaningful answer is to suggest that a reading of the argument between Carnap and Putnam would be revealing. But I don’t think that the SDA can be traced to any one source.

As an observation on SDA’s generally, they can be found in other areas of philosophy. For example, the foundationalist argument for knowledge, is (by some accounts) based on the idea that knowledge is based upon arguments, which are themselves based on further arguments, and so on, until we reach an argument which is final and needs no further support. Foundationalism, it is argued, is self defeating because it has no such  basic foundation. SDA’s have sometimes been described as themselves being self–defeating because the argument always regresses to something which cannot be supported by any further evidence, and which has to be held to be true by stipulation. Wittgenstein in his latter days may have held such a view. SDA’s can be very clever, but they are often wearisome.

Rather than rely on the SDA criticism of verificationism, a more powerful criticism is that it is based on a simplistic view of how science is conducted. Scientific procedure is not always based upon proceeding from one set of facts to another set of facts. Sometimes it proceeds by making assumptions or stipulations. Einstein based his success on making the assumption that the speed of light is fixed, and that nothing can exceed it. It has never been proved in the analytic sense, but the assumption has so far yielded correct calculations and forecasts. Quarks, bosons and other particles were long assumed to exist, even though no direct evidence existed for them. (That evidence was found long after this assumption was made.) It is very difficult to see how verificationism could cope with this approach to science. Karl Popper also remarked that his “falsifiable” approach to the conduct of science had killed verificationism. Even if Popper’s blow was not fatal, the description of science contained in Thomas Kuhn’s Structure of Scientific Revolutions was.

Also much more powerful than the SDA, is the serious attack on verificationism made by Quine in his paper “Two Dogmas of Empiricism”. Are there any really self-evident propositions? Further, he questioned the whole notion of truth as consisting of atomistic self-evident or scientifically-proven propositions. Truth was much more a question of a mutually supportive network of propositions.

This argument is not as clever as the SDA, but it is more rewarding.

One thought on “Verificationism and the self-defeating argument

  1. Somewhere around page 114 in the appendix of LTL is an explanation of how Ayer navigates certain kinds of inferences and moves science makes that might not be analytic or sensical … such moves are permissible just in case they 1) necessary and 2) ultimately lead to empirical propositions.

    I think Hume’s is/ought problem is what undermines logical positivism.

    If its norm cannot underwrite itself in applying itself and a person can see this, then that person has a defeater to believing anything warranted by that norm (adapted from Plantinga’s EAAN).

Leave a comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.