“Can anyone live as a committed behaviourist without separating their philosophy from lived experience? I doubt it. B.F. Skinner’s life embodied a fenced-off polarity between a scientific life of disciplined objectivity and an inner life of lush and conflictful subjectivity.” (Baars ‘The Double life of B.F. Skinner p. 15)
Radical behaviourism is typically attacked in cognitive science in two different ways. (1) It is argued that Behaviourists don’t postulate enough innate apparatus to account for various different aspects of cognition. Chomsky is largely responsible for this type of criticism of Skinner; his ‘Review Skinner’s of Verbal Behaviour’ is accepted by everyone in cog-science as a refutation of behaviourism. People who have never read ‘Verbal Behaviour’ or even Kenneth Macquordale excellent reply to Chomsky’s review are certain that Chomsky refuted Skinner. People also bizarrely accept Pinker’s claim that Behaviourists are committed to a blank slate ideology despite the fact that all of the evidence goes against this in belief of Pinker’s. (2) The issue of whether behaviourism can account for conscious experience. Philosophers and scientists have criticised Skinner on this issue for years. In his 2003 paper neuroscientist Bernard Baars attacked B.F. Skinner on the issue of consciousness he argued that Skinner’s life was at odds with his behaviourist philosophy. He told a psychoanalytic story where Skinner who failed at becoming a writer ended up swinging in the opposite direction and publically attacking subjective experience and literature. Baars tells a story of emotional turmoil which is revealed in three key moments. (1) Skinner as a child wanted to be an artist but had that dream crushed. (2) Skinner wrote a short story where the main character Elsa gave up her freedom to accept an unwanted fate. (3) Skinner’s ‘Walden Two’ has two protagonists who he admits represented him (though he didn’t realise this when writing the book) which shows an unconscious struggle between two different sides of him the artist craving freedom and the objective conformist behaviourist. He thinks this shows an unconscious unresolved conflict was bubbling under the surface of Skinner.
His essay is extremely speculative, but is both interesting and entertaining. Nonetheless I disagreed with a lot of it. I have some issues with his characterisation of the philosophical reaction to Skinner’s project. Baars notes that Russell, Ryle, and Wittgenstein endorsed Skinner’s behaviourism. While Ryle and Wittgenstein did admire the work of Skinner they were not radical behaviourists Ryle’s logical behaviourism was an entirely distinct research programme from Skinners. Likewise, Russell greatly admired Skinner but he was critical of aspects of Skinner’s behaviourism (See, for example, his Analysis of Mind and Inquiries into Meaning and Truth). Baars fails to mention Quine whose account of language acquisition is very similar to Skinners. More than any philosopher Quine tried to develop Skinner’s behaviourist account so it is a strange oversight that Baars doesn’t mention Quine and instead focuses on philosophers whose views are much less congenial to Skinner’s research programme.
Baars claimed that even today many philosophers are crypto-behaviourists, though he gives little evidence to support this claim aside from briefly citing the work of philosopher George Rey and Wilkes. But Baars merely says he disagrees with their views on consciousness and doesn’t specify why the views are wrong or why they indicate a crypto behaviourism. There are obviously actual behaviourists working in Applied Behavioural Analysis but there is nothing crypto about them, and since Baars doesn’t engage with any of this scientific research he is in no position to criticise it.
As part of his psychoanalytic interpretation of Skinner Baars notes that Skinner would regularly walk home reading Beaudelaire; and argues that this is somehow inconsistent with Skinner’s experimental behaviourism. I really see no inconsistency here. Saying that x is not an appropriate subject for scientific analysis does not mean that one must ignore it in daily life. There is always a gap between our daily pragmatic interactions with the world and our best scientific theory. Is an engineer who understands the theory of relativity and who uses Newtonian calculations for convenience guilty of some radical incoherency? In our practical engagement with the world we don’t act like perfect logicians we do things for amusement, for convenience etc but when doing our science we obviously try to be more rigorous. The fact that Skinner like all scientists didn’t apply his theory to every aspect of his life is not a valid criticism of him; it merely shows that he is a human. I have discussed this fact in more detail in an earlier blog ‘Theoretical Research and Practical Behaviour’.
Baars repeatedly argues that Skinner believed that consciousness did not exist. While it is true that Skinner was somewhat inconsistent about this issue (See Dennett: Skinner Skinned for an analysis), the central thrust of Skinner’s views was that consciousness was not a proper subject of scientific study. He was pretty much agnostic about the actual existence of consciousness. Its notable that Baars doesn’t cite actual quotes from Skinner to prove his case instead he goes into a discussion of Watson. This is a strange move considering that Watson and Skinner advocated different kinds of behaviourism. Watson did actually deny the reality of things like mental images despite claiming to have been a vivid mental imager earlier in his career (see Berman and Lyons ‘The First Modern Battle for Consciousness: J.B Watson’s rejection of Mental Images’ (2007)). But since Watson’s views on mental imagery have no bearing on Skinner’s views it is difficult to understand why Baars brings them up; perhaps it is because he can find no evidence from Skinner’s writings on the issue.
When he does quote Skinner he notes that he tried to translate some mentalistic terms into behaviouristic terms. He cites Orwell’s essay about ‘Politics and The English Language’ where Orwell warns about the use of cosy euphemisms which serve to block us from direct access to what is actually going on in the world:
“People are imprisoned for years without trial, or shot in the back of the neck or sent to die of scurvy in Arctic lumber camps: this is called elimination of unreliable elements, such phraseology is needed if one wants to name things without calling up things without calling up mental images of them.” (Orwell: 1946)
Baars speculates that when behaviourists translate the idioms of folk psychology into behavioural language one may end up divorcing oneself from his own lived experience. Personally I have no problem with the operational definitions typically used by Applied Behavioural Analysts. The definitions give a measure of objectivity and making prediction and control of behaviour easier. Baars speculation that these operational definitions may well be true (or false); however he presents absolutely no evidence to support the speculation.
If Baars really believes his hypothesis I think he should try to test it empirically. Perhaps through questionnaires, along with MRI scans, on behaviourist and non-behaviourist psychologists to see whether behaviourists are emotionally different. Or through studies of psychologists who have mental breakdowns to see whether behaviourists are more likely than cognitive scientists, or psychoanalysts to have breakdowns. The fact that Baars doesn’t even suggest any tests like this indicates to me that he is not very serious. That he is using the Orwell claim as an ad-hominem attack on behavioural analysts. Whatever his motives for making his claim there is no reason to take it seriously until Baars provides some evidence to support his wildly speculative (and insulting) claims about behavioural analysts.
While I don’t agree with Baars speculative claims about behaviourists being cut off from their own experiences, I am in full agreement with him that Skinner did hold back our scientific understanding of consciousness. Skinner was not a blank slate theorist, he did not deny that consciousness existed; but he like a lot of behaviourists was too timid a theoriser. Behaviourists emphasised that they allowed any innate apparatus which could be determined experimentally. In practice their theories made much less use of innate apparatus than other psychologists. This is because they thought that postulates were a sign of bad science; we are only justified in saying something exists if we can prove this experimentally. But that is just batty science; a science that uses this methodology will proceed at a snail’s pace. Popper was surely correct that bold postulates that go beyond observation and have not yet been justified experimentally are the life blood of science. There is nothing wrong with these postulates as-long as we treat them critically and try to find ways to verify and falsify the postulates.
This timid methodology didn’t just apply to innate apparatus but to consciousness studies as-well. Behaviourists worried about the subjectivity of consciousness, about the fact that it could not be measured by objective scientific instruments. Watson in his (1913) ‘Psychology as the Behaviourist Views It’ pointed out seemingly interminable debate between the followers of Wundt and the followers Titchner. He correctly noted the various methodological flaws in introspectionist psychology. But he simply gave up too soon. It seemed at the time that we simply lacked the tools to study consciousness objectively. Theorists like Crick, Koch and Baars have shown otherwise. Their success came through bold hypothesis, and not viewing methodological concerns as impossible to overcome. Behaviourists were wrong on the issue of consciousness because of the timid nature of the methodology not because they were blank slate theorists or mentally ill zombies.
 I am not suggesting that we are anywhere near being able to solve the hard problem of consciousness or the relation between personal and subpersonal explanations. I am merely saying that the study of consciousness has made progress that most behaviourists doubted would be possible.