Logical Positivism

Logical positivism in its purest, simplest, most extreme form didn't last very long. In this course. we shall call the fallback positions logical empiricism.

Logical positivism is a form of empiricism, and, like most empiricists, the logical positivists thought that scientific knowledge is of the same sort as all knowledge, it is just particularly good and organized knowledge. Thus, today, I won't mention science any more, just knowledge.

The logical positivists were enthralled by Frege's formal deductive system: They tended to think that we now completely understand deductive reasoning in a way we don't understand anything else, and so we build out from that.

Empiricism is false: There are some things we don't learn from experience. Example: analytic truths. "Bachelors are unmarried males" is true, and known to be true, the logical positivists thought, not as a result of experience, but as a result of conventions we adopt. Conventions are codified in rules we adopt. Frege provided a complete system of rules for deductive reasoning adequate for almost everything including mathematical proof. Therefore, the positivists concluded, logic and mathematics are analytic and true by convention.

Everything else is empirical. (In other words, as it is important to note for later, all truths are either analytic and, when known, know by convention, or synthetic and known through experience.

The logical positivists had a great way to explain away knowledge of logic and mathematics as not empirical: it is merely conventional. But that meant in addition that in acquiring empirical knowledge, they were already entitled to use all of mathematics, which makes it look like it will be possible to give an empiricist account of science, including their central example, physics.

Theory of language. The testability (or verifiability) theory of meaning: The meaning of a sentence is what it takes to tend to show that it is true or false, that is, what experience is relevant to its truth or falsehood.

Every word has a definition (convention, analytic) that shows how it contributes to the experiences related to sentences in which it occurs.

A sentence that cannot be tested against experience is meaningless.

That is what they said. What they meant was

A sentence that cannot be tested against experience is either true by convention and hence uninformative about the world or meaningless.

How experience enters into language. There are observational terms ( words or phrases directly connected with experience) and nonobservational (theoretical) terms. The positivists insisted that every theoretical term must be connected by a correspondence rule to observational terms.

That theoretical sentence might be verified by the following observation:
The top of the red-looking stuff in the thermometer is on a line next to which appear the following symbols: "104."

That is just a sketch—I'd have to say what the observational content of "thermometer" is to complete it. Such incompleteness is characteristic of the logical positivists' examples, which indicates a problem.

What did the logical positivists dismiss as meaningless? Ethics, theology, idealism, Hegel, Marxism, astrology, …. The list is quite long.

The positivists and empiricists were interested in the logical structure of knowledge and of science. They wanted to do for knowledge what they took Frege to have done for logic. They weren't usually interested in particular facts or scientific theories, but the rules by which we produce well confirmed facts and good theories. Thus, they wanted to devise a formalized logic of confirmation, one that could be used as a criterion for what scientific theories we should accept given the observations we have made, and so forth—just as Frege had devised a formalized logic thta could be used as a criterion for the validity of deductive reasoning. None of this has anything to do with describing or knowing anything about science as actually practiced. It is entirely normative, not descriptive.

Of course, a norm that no one had ever conformed to would not be a useful norm, and, since we don't know a good set of norms for confirmation, we may want to look at what people actually do to get ideas about what the norms might be, but that is a secondary interest.

The observational-theoretical distinction is the basis of the logical positivists' claim to be empiricists: every synthetic sentence is closely related (in the simplest version, equivalent) to sentences in which there are only observational terms. Such sentences can be assessed empirically, that is confirmed (or not) on the basis of observation in accordance with the logic of confirmation.

-- ShaughanLavine - 07 Sep 2005 - 05 Sep 2007