Wednesday, April 13, 2011

Logical Positivism

I encountered a couple internet logical positivists recently. It's unusual that someone will know what "logical positivism" is and still adhere to it, as some of its primary adherents have all but pronounced it dead. It refers to a movement in the 1920s and 30s which emphasized syntax and empiricism. In any case, these particular logical positivists were emphasizing one of the primary tenets of logical positivism: metaphysics is semantically meaningless. It was interesting how quickly these logical positivists were able to get those to whom they were replying to assume with them that some of the more foundational concepts of logical positivism were true, and given that I've been reviewing logical positivism in my philosophy class over the past few weeks, I thought I'd share some thoughts on it so that one can anticipate these subtle tactics and effectively counter them. As the more insightful critiques of a given philosophy usually come from those who used to believe it, for those who wish to skip the text and just listen to the problems inherent in logical positivism, this interview of A. J. Ayer (begin at 6:28) gives a nice, short summary, although Ayer seems a little reluctant to say logical positivism is a wholly unsatisfying philosophy.


Syntax, semantics, and pragmatics all pertain to linguistics. Syntax refers to the rules or structure of a theory of language, semantics refers to the meaning of words, and pragmatics designates that field of study whereby one utilizes the usefulness of language. A development in logical positivism that allowed men in the Vienna Circle (the original logical positivists) such as Carnap to bridge what was thought to be a metaphysical gap between syntax and semantics/pragmatics was the following postulation: “a sentence ‘S’ is true if and only if S.” Meta-language, whose subjects are sentences (“S”), became connected to object-language, whose subjects are real world referents (S). Example: "The snow is white" is true if and only if it is really the case the snow is white.

Problems: an immediately perceivable problem with logical positivism here is that it is basically susceptible to all the problems of empiricism. Examples: the reliability of sensation must be unreasonably presupposed; "snow" and "white" are ostensibly, arbitrarily defined such that knowledge of communication between persons is, as Augustine anticipated some 1600 years prior to logical positivism, impossible - that is to say, empiricism is solipsistic; etc. But as a matter of fact, logical positivism was radically empirical, as shall be seen in the below.

Verificationism and the Verification Criterion of Meaningfulness

Verificationism is the position that a proposition must be able to be determined to be true or false in order to be meaningful. So, for example, propositions in the future tense are apparently not meaningful. The verificationism of logical positivism and its extreme emphasis on empiricism led to the verification criterion of meaningfulness, the principle to which logical positivists (such as the ones I met recently) refer when asked why they believed metaphysics is nonsensical. The verification criterion of meaningfulness refers to the following: propositions which are not analytic (i.e. tautologous) must be empirically verifiable or else they are meaningless. Such being the case, metaphysics would by definition be meaningless.

Problem: the most obvious problem - one which caused Ayer to step back from logical positivism - was that the verification criterion of meaningfulness is a proposition which cannot itself be verified on its own grounds. It's a self-defeating principle. This is, in my estimation, the crushing blow to logical positivism. Popper was another individual who saw this flaw and attempted to replace verificationism with falsificationism as the demarcation criterion of science. This has its own problems, but at least Popper didn't reject metaphysics outright. Even with these problems, one final subject related to logical positivism merits note.


Ayer, consistent with verificationism and Positivism, taught that ethical posits are neither true nor false. “Ought” and “should” statements are rather emotive. This is the ethical theory known as emotivism. Stevenson, another logical positivist, contributed to this theory by explaining that emotivism does not preclude a means of ethical dialogue. By distinguishing between beliefs (eg. “this actor is in the movie”) and attributes (eg. “I prefer tennis to movies”), Stevenson made significant strides in an attempt to resolve disagreements of this nature. He noted that increasing one another’s awareness of facts can only stimulate beneficial conversation, although he realized differences in attributes are irresolvable.

Feigl, on the other hand, thought emotivists missed the point. Forwarding the concepts of validation and vindication, analogous to internal and external questions, Feigl believed he came the closest to resolving ethical disputes. Validation refers to the idea that if two disputants share the same belief in a general ethical principle, this shared belief can function as a means by which they can seek to resolve differences on a specific point. And even in there are no shared principles, according to vindication, the persons can nevertheless compare their respective systems in hopes one will persuade the other. It is only when disagreement on this level persists that one can claim there is no means of resolution on Positivistic grounds.

Problem: I would agree that each of these contributions takes Positivism to its logical end. But that's precisely the problem. One can describe ethical beliefs but not account for them. Ethical discourse is possible, although ultimately, it may certainly be the case conflicts are irresolvable. In fact, it seems to me the emotivists frame the issue in a different manner than does Feigl, but they don’t differ significantly in that both agree insofar as ethical statements are unjustifiable and ethical dialogue can only go so far.

Many logical positivists bite the bullet and accept moral nihilism, in which case one can always appeal to one of the plethora of intrinsically insoluble difficulties, some of which are mentioned above, but given that one will usually hear arguments from incredulity from logical positivists regarding the idea metaphysical statements can be meaningful, it doesn't hurt to be able to throw incredulity back in their face when you ask if they can seriously believe there is nothing objectively objectionable about rape, murder, et. al.


Andrew said...

The least known, but wholly efficacious, problem for Verificationism is that it offends basic principles of first-order logic (or at least those which are not constructive-like).

Ryan said...


Andrew said...

Essentially, if a proposition fails to meet the criterion of meaning viz. if it is unverifiable, then it is senseless. If it is senseless then there is no interpretation which is capable of assigning a value of T or F to it. Granting such, if any proposition which fails to meet the criterion of meaning is a part of a compound sentence then that sentence is also senseless and, accordingly, bereft of a value of either T or F. This means that a plurality of basic classical first-order principles will have counter-examples.