Scattered Thoughts on The Phenomenology of Free Will.

‘The Phenomenology of Free Will’ here refers not – directly to the experience of being free – but to an article by our very own Eddy Nahmias, Stephen Morris and Thoma Nadelhoffer (as well as Jason Turner), published in a recent issue of The Journal of Consciousness Studies. The experiphils argue that libertarians often appeal to the phenomenology of choice or deliberation to motivate their project. They claim that the experience of choice is active, whereas compatibilists deny this. But, they point out, generally philosophers have done no more than consult their own experiences. There is a danger, they argue, that their experiences – or their reports of their experiences – will be contaminated by their theories. Therefore, they argue, we ought to get our data from laypeople.

First set of comments (I warned you that these would be scattered thoughts). Is it true that the folk have intuitions and experiences untutored by philosophy? Someone – I forget who – wrote something to the effect that today’s common sense is the graveyard of yesterday’s philosophy. Certainly the folk – at least the Western folk – seem to find some philosophical theses intuitive. Dualism (for instance) seems to come naturally to people. It’s a part of most Western religions, and its often implicit in sci-fi movies (Dark City for instance). Paul Bloom claims that we are natural born dualists. I haven’t read his evidence – perhaps he’s done the spadework here – but I’d like to see the hypothesis tested: is dualism cross-cultural (and cross-historical)? In pre-Socratic Greek thought, survival after death seems to be bodily, not spiritual. Now, if turns out that dualism is not universal, I suggest we have some evidence against the claim that untutored intuitions are uncontaminated by philosophy.

Second, suppose its true that the experience of freedom naturally is as compatibilists claim, and that libertarian experiences are theory driven. In that case, the phenomenology can’t be driving the theory at all. Some other consideration must be doing the work. I suspect that however we go on the experience, its not phenomenology that’s really the driving factor – and I’m not convinced that the philosophers quoted in the article, saying that it seems to them that at the moment of choice they feel free to choose otherwise are (or take themselves to be) reporting the phenomenology of choice.

Now turn to the data the experiphils have gathered. This comes from two studies they have conducted, and two earlier studies. One comes from a study of choice, in the introspectionist tradition, the second a much more recent set of studies on the experience of freedom. First, a few words about the design of the studies conducted by the experiphils.

They asked 96 introductory philosophy students (none of whom had yet studied to free will) to think about making a tough decision where they believe they could have done otherwise. They then asked which statement best described what they had in mind when they thought they could have done otherwise
(a) even with everything the same;
(b) only if something had been different
(c) neither.

62% answered (b). I have two problems with the study. First, I think this is asking the students to do philosophy, not to report on phenomenology at all. The question seems equivalent to asking them what is meant by ‘could have done otherwise’. Second, I can’t help wondering whether philosophy students, asked a question by a philosopher, mightn’t be on the lookout for a trick. Suspecting that there are deep issues here, they might answer (b) precisely because it seems unintuitive.

In the second study, the experiphils asked subjects deliberating about a decision to report their thoughts as they crossed their minds, They simply described the features of the objects about which they were deliberating, and made no reference to themselves as causes of their decisions. But this, surely, is neutral between compatibilist and libertarian accounts. Of course when we make decisions we attend to the options. That doesn’t mean there isn’t, also, an experience of choice. Indeed, we know there is such a feeling – or, better, such feelings. These are the feelings of agency and of control which are lost, for example, in delusions of alien control in schizophrenia.

Now the earlier evidence. One set comes from Westcott’s Psychology of Human Freedom. Westcott asked subjects to rate how free they felt in certain situations, in order to see what descriptions apply to situations in which subjects feel free or unfree. He found that situations in which subjects felt free fell under the headings ‘self-direction’, ‘absence of responsibility’, and ‘exercise of a skilled behavior’, whereas those in which they felt unfree were instead characterized in terms of - ‘prevention from without’, ‘diffuse unpleasant affect’, and ‘conflict and indecision’.

It seems that conflict, indecision and close-call decisions are associated with a feeling of ‘unfreedom’, whereas easy and routinized behavior is associated with a feeling of ‘freedom’. West’s 1927 introspectionist study supports this finding: subjects felt less control when confronted with difficult decisions, which were experienced as effortful, and more control when they made easy, confident choices.
The reason I have put freedom in inverted commas in the preceding paragraph is I wonder whether Westcott’s subjects are not actually reporting a feeling of control (or even lack of unpleasant affect – think of locutions like ‘freedom from hunger’). It may be that West’s description, of difficult decisions as feeling less controlled, is more accurate. That’s what we should expect from the neurobiology, in any case. The forward model explanation of the feeling of control (for a quick sketch, see Frith and Hohwy’s article in the same issue of the Journal of Consicousness Studies) explains the feeling as the result of mechanisms which ensure that the experience of movement is not informationally salient for the subject. When the feedback from the movement matches the model, we have the feeling of control, and awareness of the movement is attenuated. It is only when there is a mismatch (or the attenuation fails) that we lose the feeling of control. So routine and easy actions give rise to the feeling of control.
But if that’s right, then the feeling of control is lost in difficult decisions. If libertarians are right that these are the paradigms of free action, then the phenomenology in the routine case is irrelevant. At best, the excursion through the phenomenology seems to leave us where we started: with each camp able to point to experiences which support its view, in just the place we (and they) would expect to find them. I don’t see how this will break the stalemate.

Comments

[Eddy Nahmias]

OK, it's about time I responded to Neil's interesting post about my co-authored paper "The Phenomenology of Free Will." First, let me say I appreciate the detailed reading and commentary. Second, let me say that the main thesis we were advancing is: To the extent that philosophers support their analyses of free will with descriptions of the phenomenology of decision-making and action, they would do well to back up such descriptions with evidence that goes beyond their own introspective reports. Libertarian and compatibilist descriptions of the phenomenology conflict and we suspect that their descriptions are driven largely by their prior commitment to their respective theories. (Another explanation is that there are two types of experiences and each camp picks out one of them to advance their theory. Another explanation is that the experiences are amenable to descriptions in terms of both theories. Another explanation is that different people experience choice and action in different ways and that leads them to develop different theories. And these explanations are not mutually exclusive.)

Now, Neil may be right that there is no such thing as pre-philosophical phenomenology because ordinary folk's intuitions and experiences are permeated with the traces of philosophical theory. First, I am more dubious of the power of philosophy (though religion is a more likely theoretical influence) and also the power of theory to determine phenomenology (Shaun Nichols helpful example: no matter what your theory of teeth or pain, won't a toothache feel roughly the same? Of course, action phenomenology may be more susceptible to theory). Second, the research we suggest would help determine if this claim is true. Third, cross-cultural evidence would be useful to test this claim (and to examine if there are different experiences of choice and action--and intuitions about free will and moral responsibility--across different cultures). Finally, if ordinary folk's phenomenology is in fact determined largely by the philosophical theories they've absorbed, it would be nice to know which theory they've absorbed, compatibilist or libertarian or a bit of both or what?

Regarding the claims Neil makes in the third paragraph, I don't agree. The philosophers we quote and others (especially libertarians) most certainly do suggest that the experiences we discuss (of an unconditional ability to do otherwise, of a self as source of action, and as close calls as paradigms of free will) provide reasons to analyze free will as they do. They think the phenomenology drives, or at least supports, their theory. We conclude libertarians likely have it backwards--though this conclusion is very preliminary given the dearth of evidence on the relevant phenomenology.

Regarding the evidence, we share some of Neil's concerns. For instance, we explicitly say in the paper that our survey of what people mean when they feel they could have done otherwise may be tapping in to their intuitions or theories more than their phenomenology. But we say it's still interesting to hear what they would say about the philosopher's own descriptions (and it's interesting to hear that almost twice as many folk offer the answer that jibes best with compatibilist descriptions than the one that jibes with libertarian descriptions). And I doubt our students were answering the opposite of what they really thought just to be philosophical or something (demand conditions are always a worry, but we try to avoid them with our instructions).

Our protocol analysis study does not deny that there is 'an experience of choice.' The question is whether that experience involves the aspects libertarians say it does and we found no such evidence. The feelings that are lost in schizophrenia, etc., do not, as far as I know, correspond with libertarian descriptions (though there may be some interesting ways to tie them to the 'self-as-source' idea). In any case, our main point is that more work needs to be done, not that our work fills the gap.

I agree that the Westcott and West descriptions may be picking up on feelings of control, but the concept of free will is always tied to the issue of control (even by libertarians, especially agent causationists). I don't see how it supports the libertarian to find out that the choices they point to as paradigms of free will are those that involve agents feeling a diminished sense of control over the outcome of the choice.

Finally, we don't suggest that phenomenological research will "break the stalemate." As with our work on folk intuitions, we are simply offering a type of data that is relevant to the extent that philosophers have posited such data--without evidence--as support for their theories. If we can simply motivate philosophers to say what (if any) role they think phenomenology or intuitions does and should play in their theories and in the debate, we will have done something important.

And finally finally, please don't call me an "experiphil" (not a very euphonic term). I don't like the term "experimental philosophy" (as I've posted before) and we'd rather encourage trained experimentalists to do this work precisely because we are not so trained (though I did talk with Ericsson about protocol analysis). How about "philosopher who thinks empirical evidence is relevant to the philosophical debates and would like such evidence to be obtained and to better understand exactly how it is relevant to the debates" (not very euphonic either).

[tnadelhoffer]

Perhaps it would be better to call those who are averse to using empirical data about folk intuitions in philosophy "empirically averse apriorists," "empirically unsubstantiated intuition-driven conceptual analysts," or "anti-laypersonists" rather than trying to label those who are NOT so averse "experimental philosophers" or "experiphils." Seriously though, I, too, would like to thank Neil for his post.

[Neil]

Thanks Eddy and Thomas. I think our views are not all that far apart. I'm not an "empirically averse apriorist"; I think the more empirical data the better, and that this data should include phenomenological data (both normal and pathological). I'm more sceptical about asking ordinary folk to describe their phenomenology: I think the introspectionist tradition might be on to something here. Is this feeling I have in moving my arm a feeling of control, or of agent causation, or of alternative possibilities, or of self-as-source (actually I suspect it is of several of these things at once). Sorting out these things requires not only paying attention to one's phenomenology, but also, I think, reflecting on a variety of experiences. I've been thinking about the phenomenology of action more or less continuously for the last week or so (I'm writing a paper with Tim Bayne on the topic), and I think I have noticed things that I wouldn't have if I hadn't been reading about anarchic hand (so there's more to it than proprioception... let's see... yes, that seems right) and other kinds of syndromes.

A related worry is this: not only is the phenomenology hard to sort out; it is also somewhat indeterminate (in a way in which the experience of a toothache is not indeterminate, or at least to a greater degree). It may well be that theoretical commitments play a determining role here (in the way in which, as Elster suggests, culture helps determine the precise experience of emotions). I have anti-compatibilist intuitions myself (jostling alongside compatibilist) and I think they're driven by theoretical considerations at least in the first instance, not phenomenology. In fact, I think I was trying to make the point you were making, Eddy: libertarians have it backwards if they think that phenomenology drives their view.

Thanks again for the responses.

[A. Darafshdar]

Dir Sir
I'm looking for the relation between human brain potentialities and human ontology from phenomenology point of view. For example, which part of brain is engaged in thinking on basis of science and which part in heedful circumspection? I'll be very appreciated receiving any advice

[A. Darafshdar]

Is there any relation between neurology and phenomenology?