Mainstream 6

Critique of Heterophenomenology

Christian Beenfeldt

Journal of Consciousness Studies, 15, No. 8, 2008, pp. 5-34

It is encouraging to read a frontal attack on a central part of Dennett’s reasoning by a writer that is affiliated and peer group reviewed. So often it is left to independent researchers to try and argue the contrary. Dennett has dominated much of consciousness thinking over the last two decades, with many writers simply accepting him as the correct standard, and criticising any deviation. It could be argued that this influence has contributed to the apparent stagnation of consciousness thinking in the present decade, with an orthodoxy that lacks explanatory value, but is effective in pouring ridicule on any attempts to get closer to the problem of consciousness. The fact that this criticism of Dennett comes from a doctoral student perhaps encourages a flicker of hope that a new generation of consciousness researchers will be less in thrall to Dennett.

Beenfeldt’s paper is an attack on Dennett’s concept of heterophenomenology. Dennett starts by saying that he will take first-person experience seriously, ‘or as seriously as it can be taken’, but really this is a method of discounting the value of first-person experience. The heterophenomenology approach is proposed as an exclusively third-person approach to consciousness. In fact, Dennett regards accounts of first-person experience, even when derived from laboratory experiments, in much the same way as we regard fictional accounts of behaviour found in novels. They might give some insight into human behaviour, but they are not true accounts of specific events. As a further image, Dennett invites researchers to treat first person accounts from human laboratory subjects, as if they were accounts of primitive religious ideas given to anthropologists, by members of a rain forest tribe. The experimenters, like the anthropologists, are expected to take a detached and sceptical view of the subjects reports, not absolutely discounting them, but using their superior science, to view them as basically improbable. In practise, the only data Dennett is prepared to take seriously is physiological or behavioural data, while first person reports are viewed as fairy tales.

Beenfeldt asks why Dennett should have reached such a conclusion. He pulls out three arguments from Dennett’s writings. (1.) First person reports are subject to errors and self-delusion. (2.) Heterophenomenology represents a neutral stance. (3.) Heterophenomenology is in line with the usual practise of science.

(1.) With regard to his first proposal, Dennett’s main claim is that subjects are sometimes wrong when they make statements about their experiences and mental processes. He feels that the problem is not so much that they deliberately lie, as that they confabulate, making up likely sounding stories, guessing, theorising or speculating, but presenting this false material as fact. This is the major reason for treating first-person experience as fiction. In particular, Dennett sees only two alternatives, either to treat first-person experience as pure fiction, as he suggests, or to except first-person reports as ‘infallible pontifical proclamations’.

Beenfeldt accepts that statements about experiences can often be incorrect. However, he does not think that there is a simple choice between regarding all such reports as fiction, or else accepting them as infallible. He suggests that the trustworthiness of first-hand experiences may fall into a number of intermediate categories.

He offers four main categories. First, memories of the fine details of events in the distant past might be rightly regarded as of dubious reliability. Secondly, there are outline or general memories of events in the distant past. These are seen as more reliable than detailed memories of the past, and are accorded a rating of moderately reliable. Very detailed short-term memories might also occupy this middle ground. However, general or outline memories of very recent events, for instance the metal bar dropped on my toe hurt, or the book I was reading a minute ago was red, might be thought to have quite a high degree of reliability. It might be hard to convince even scientifically educated people that the hurt toe or the redness of the book fell into the same category as a conversation with Mr. Darcy in Pride and Prejudice. Beenfeldt further points out that in legal proceedings, there is also a sliding scale as to how reliable witness testimony is considered to be, not an all or nothing divide between fantasy and infallibility.

While it is certainly true that many people do have an irritating habit of fabricating altered versions of past events, it seems an extreme position to say that this is true of everybody’s reports of all experiences. It is worth keeping in mind that some of the psychological experiments that are given prominence in the literature tend to refer to very unusual cases. For instance, the example of split-brain patients is sometimes used. Patients with serious epileptic conditions sometimes have an operation to cut the connection between the two hemispheres of the brain. Surprisingly, they can function fairly normal after their operation. However, in experiments which deliberately ascertain that information is only presented to one field of vision, they can make a response without the other half of the brain realising why the response is being made. In this highly unusual circumstance, which is only ever likely to arise under artificial conditions, it is shown that subjects will invent or confabulate a rationalisation of their behaviour. Some writers have gone from this somewhat bizarre laboratory experiment, to implying that such confabulation is an invariable procedure. Not only is this contrary to experience, but it would be a highly unadaptive procedure. A species which fantasised about its experiences, rather than retaining useful reports, would be on a royal high road to extinction.

(2.) Dennett argues that heterophenomenology is a neutral tool for approaching first person experience. Beenfeldt finds this less than plausible given that Dennett’s own writings take the third-person point of view as axiomatic. They also describe the third-person point of view as scientific, with the implication that the first-person must be invalid. The stated reason for developing heterophenomenology in the first place was distrust of the first-person point of view. First-person material is described as fiction, while third person material is the starting point of valid enquiry. Dennett may be right, but it is excessive to describe this attitude as neutral towards the first-person material, when he already decided that it is nonsense before he even started.