Beliefs and desires: a false theory, like alchemy

Eliminative materialism, eliminativism

Paul Churchland born 1942Our beliefs and desires are a bad theory, which neuroscience will replace.

What it claims

Beliefs and desires (the "propositional attitudes") do not exist: folk psychology and its vocabulary will be eliminated by neuroscience.

Why

Talk of beliefs and desires is an old theory that fails and stagnates; neuroscience will replace it, as it did alchemy.

The same idea through history

Paul Feyerabend1963Feyerabend holds that the language of mental states is a theory that can be replaced by that of neuroscience.

Richard Rorty1965Rorty defends a neighbouring idea: our mental categories might disappear.

Paul Churchland1981Churchland makes it a full thesis: beliefs and desires do not exist, because the psychology that posits them is false.

In short

"Beliefs and desires: a false theory, like alchemy" is an argument by Paul Churchland ("Eliminative Materialism and the Propositional Attitudes", 1981, 1981) on the question "Is the mind the brain?" (answer: the mind comes from matter). Its claim: Beliefs and desires (the "propositional attitudes") do not exist: folk psychology and its vocabulary will be eliminated by neuroscience. According to AskPhilosophy's logic check (Z3 solver), the argument is valid only if one grants a premise the author leaves unstated (P4, P5). AskPhilosophy presents 4 objections, 3 of them with a documented reply, and the original text behind each premise, checked word for word.

"Eliminative Materialism and the Propositional Attitudes", 1981

Valid with an implicit premise

When you say "he took his umbrella because he believed it would rain and wanted to stay dry", you explain his act by a belief and a desire. Paul Churchland asks an uncomfortable question: what if this way of talking were a theory, like alchemy, and a false one? Alchemists spoke of the "spirits" of mercury and sulphur; we speak of beliefs and desires. When neuroscience is complete, he says, these words might vanish as alchemy, phlogiston and "vital spirit" did. This is the most radical version of materialism: the mind as we describe it is not the brain, it simply does not exist.

click a premise to reject it

If you accept every premise, logic forces you to accept the conclusion. So the debate is about the premises.

Checked by a logic solver: the premises cannot all be true while the conclusion is false.
Without the implicit premise: Without P4, the computer finds a case where folk psychology is false and stagnant, where neuroscience surpasses it, and yet it is not replaced but translated: the word "believe" would then roughly name a brain state. Without P5, it finds a case where the theory is replaced but what it posited still exists under another name, like heat, which did not vanish when the "caloric fluid" was abandoned.

The argument is valid with both hidden premises. P4 is the heart of the debate: it says a false theory is replaced rather than "reduced" (translated) into the new one. Defenders of folk psychology hope for the opposite: that "believing" will end up translated into brain terms, as temperature is translated into molecular motion. P5 is a common principle in science, but debatable: we have sometimes kept a word while changing the theory (like "atom"). The argument is also stronger than Huxley's epiphenomenalism: it does not say consciousness fails to act, it says our words for the mind describe badly what goes on.

Analogy. In the Middle Ages, a shiny metal was said to shine because it was "ensouled by mercury", one of the alchemists' "spirits". No one thinks these "spirits" need saving: chemistry changed and they vanished. Churchland asks why it should be any different for beliefs and desires.

Objections

Each objection is a conversation: click to see who replies and where the debate stands.

Lynne Rudder Baker 1987 · attacks the conclusion

Eliminativism contradicts itself. To assert "beliefs do not exist", you must believe it is true and want to tell others. If the thesis is true, no one can believe it or assert it: it destroys itself as it is stated. (Baker describes it as "cognitive suicide".)

Churchland replies

Paul Churchland 1981 · Churchland's anticipated reply (1981), to the same objection before Baker

He knows this "very popular" argument. A reductio is no better than the assumptions it uses: here, a theory of meaning that presupposes the integrity of folk psychology. One can as well reject that theory of meaning. Assuming it begs the question. He shows this with a parallel: in the 17th century, saying "there is no vital spirit" would have seemed self-refuting, since the speaker would have to be dead.

The debate goes on (our summary) Defenders of the objection answer that the parallel limps: one can say "there is no vital spirit" while alive, whereas one cannot say anything without expressing a thought. The debate is about what "asserting" is: an act made of beliefs (Baker), or a brain process described another way (Churchland). We could not read Churchland's answer to Baker (our summary).

Jerry Fodor 1987 · attacks premise 2

If commonsense psychology collapsed, it would be, beyond comparison, the greatest intellectual catastrophe in the history of our species. Nothing except perhaps commonsense physics comes as near our cognitive core. So it must be defended: there is no decisive reason to doubt its explanations.

Churchland replies

Paul Churchland 1981 · Churchland's anticipated reply (1981) to defenders of folk psychology

He admits that folk psychology has some explanatory and predictive success, which makes one want to keep it. But for him it is a "stagnant or degenerating research program", and has been for millennia. Partial success does not show that its categories are the right ones.

The debate goes on (our summary) Fodor had not yet written his book when Churchland published (1981), so the text does not answer Fodor himself, but any defender of folk psychology. Fodor holds that it is literally true and that its basis will be found in the brain (a "language of thought"); Churchland thinks the opposite. Neither has had decisive proof to date (our summary).

Daniel Dennett (the intentional stance) 1991 · attacks premise 2

Even opponents agree that folk psychology allows very good prediction of behaviour. Predicting that someone will duck if you throw a brick at him is easy from the intentional stance; it will always be intractable if you have to trace the photons from brick to eyeball and the neurotransmitters from optic nerve to motor nerve. Beliefs are therefore real "patterns" in behaviour, even if they are not objects filed in the brain.

no reply

We have not yet found a direct reply by Churchland in the texts.

The debate goes on (our summary) Dennett himself places Churchland at the far end of the spectrum: an eliminativism that "denies the reality of beliefs altogether". His "mild" realism is the middle position: beliefs are real as centres of gravity are, without being pieces of brain. We do not have Churchland's answer to this position (our summary).

Objection Churchland anticipates (the functionalists) 1981 · attacks premise 1

Folk psychology is not, strictly speaking, an empirical theory: it is not false, or at least not refutable by empirical considerations, and it cannot be transcended like a defunct empirical theory.

Churchland replies

Paul Churchland 1981 · Churchland's reply (same paper)

He tells how a defender of alchemy could have said that "ensouled by mercury" is a mere functional state, safe from any reduction. This ruse, he says, would work just as well for the phlogiston theory. He sees in it a "smokescreen" protecting error, and asks why the same charade would not be played for folk psychology.

The debate goes on (our summary) The functionalist answers that the analogy proves too much: if any level of description could be eliminated this way, chemistry, biology or economics could be too. For Churchland, the difference is that folk psychology has learned almost nothing in twenty-five centuries (our summary).