Paper on intuitions

Steve Clarke has a paper forthcoming in Philosophical Papers, scheduled for publication in July 2013, which he thought would be of interest to experimental philosophers. The paper,  'Intuitions as Evidence, Philosophical Expertise and the Developmental Challenge' can be downloaded from here.

Abstract:

Appeals to intuitions as evidence in philosophy are challenged by experimental philosophers and other critics. A common response to experimental philosophical criticisms is to hold that only professional philosophers’ intuitions count as evidence in philosophy. This ‘expert intuitions defence’ is inadequate for two reasons. First, recent studies indicate significant variability in professional philosophers’ intuitions. Second, the academic literature on professional intuitions gives us reasons to doubt that professional philosophers develop truth-apt intuitions. The onus falls on those who mount the expert intuitions defence to meet these objections because it is implicitly being claimed that training and practice caused professional philosophers to acquire reliably accurate intuitions and we are owed an account of how this transformation takes place. A possible response to this situation is to attempt to reform philosophical practice to improve the quality of intuitions. Another possible response, advocated here, is to avoid appeals to intuitions as evidence.

Comments

[Joachim Horvath]

Hi Steve,

thanks for your very interesting and challenging paper. For now, just a couple of quick comments.

You frame the debate in terms of "intuitions as evidence", which is fine, given that almost everyone does it. But I think that ultimately it does not matter for the force of the experimental challenge. For example, even if Williamson is right that our judgments about thought experiment cases are counterfactuals, how does this help when experiments show that our (counterfactual) judgments about e.g. Gettier cases are subject to various forms of bias? So, the more sophisticated form of the challenge still stands: that our assessments of thought experiment cases are sensitive to various irrelevant factors, irrespective of their precise logical form or of the exact mental state that they express (this is a completely unoriginal point that I've just taken from people like Weinberg or Alexander).

The main problem you identify is that there is no "clear, reliable, and timely feedback" to intuitions (I continue with that term now) about thought experiment cases, and so - given the empirical research on expert intuitions - there is no plausible way how philosophical expertise concerning these intuitions could have developed. You use this to set up the following inductive argument against professional philosophers' intuitive expertise (on p. 22):

"Premise 1: In all (of a significant number of) examined domains where accurate professional intuitions have been acquired, clear, reliable and timely feedback is available to enable intuitions to be improved.

Premise 2: Clear, reliable and timely feedback is unavailable to enable philosophers’ intuitions to be improved.

Conclusion: Therefore, it is very unlikely that professional philosophers have developed accurate professional intuitions."

The response that I tried in my earlier paper was, in effect, to challenge Premise 2 and argue that there is a lot of room for appropriate feedback in philosophical training. I think my main idea was that this feedback need not be directed to the intuitions themselves, but can instead bear on the improvement of certain abilities, like the competence in properly interpreting thought experiment cases, that are crucial for making the relevant intuitive judgments. And then I tried to make a case that the empirical literature on expertise does not really challenge that seemingly plausible suggestion.

In addition to that, I'm now inclined to object that the term "intuitions" does not refer to the same psychological kind in Premise 1 and 2, and that your argument therefore is not a good inductive argument. For, it seems to me that the professional intuitions that expertise researchers have studied so far (i.e., the Premise-1-intuitions) are something like intuitive predictions of uncertain outcomes, e.g. the judgments of clinical psychologists about people's academic or job aptitude. Philosophers' intuitions about hypothetical cases (i.e., the Premise-2-intuitions), on the other hand, are not in any sense predictions of uncertain outcomes. In a good thought experiment, all the relevant details are explicitly specified, and if they are not, it's simply not a good thought experiment. In other words, in our judgment about a well-designed thought experiment, there's zero uncertainty with respect to the relevant target property. For this reason, it seems plausible that intuitions about thought experiment cases draw on very different cognitive resources than the intuitions of fire fighters or clinical psychologists. My own view is that conceptual competence plays a crucial role here. Be that as it may, I think that expert intuitions of the kind that are relevant in philosophical thought experimentation have not really been investigated systematically before experimental philosophers began doing this about ten years ago.

The upshot is that we cannot rely on the existing literature on expertise when it comes to the merits of philosophers' intuitions. No, we really need to test these intuitions directly. In your paper, you cite four initial studies that have taken up this important task, and their results are clearly not encouraging for proponents of the expertise defense of philosophers' intuitions. But when it comes to deciding a controversial empirical issue like that, four studies are almost nothing. We are still very far away from being entitled to draw any justified conclusions about the overall quality of professional philosophers' intuitions on such a meager basis. Therefore, I think that the presumption in favor of philosophers' intuitive expertise still stands, for which I tried to argue in my earlier paper - but it stands on an entirely defeasible ground that isn't as solid anymore as it was just a few years ago.

[Steve Clarke]

Dear Joachim,

Thank you for your comments. My thoughts are as follows:

You mention that in your previous paper you offer the suggestion that indirect feedback on abilities that are relevant to intuitive judgments could be sufficient to enable the acquisition of accurate professional intuitions in the course of philosophical training. This would be very nice if it were true, but unfortunately for us philosophers the evidence from other professions tells strongly against this suggestion. Financial planners, psychotherapists and clinical psychologists all receive such indirect feedback on the various abilities that bear on their intuitions but available evidence suggests that in all of these professions accurate intuitions are not (currently) acquired. It looks like accurate professional intuitions are only acquired when experts receive clear, reliable, timely information about when their intuitions are accurate and when these are inaccurate, and some lesser form of feedback won’t do in these other professions. So we have good reason to think that it won’t do in philosophy either.

Regarding your attack on my inductive argument: You point out that that philosophy is different from other professions, and therefore, you suggest we are not entitled to induce from evidence about other professions to a conclusion about professional philosophers. Now just showing that philosophy is different from other professions won’t get you what you want here. The inductive argument I offer involves an induction over evidence about a variety of very diverse professions. Nursing is different from firefighting, which in turn is different from accounting, which, in turn is different from astronomy. Unless you can show that philosophy is a truly unique profession the inductive argument goes through. Judging by what you say about the contrast between intuitive predictions of uncertain outcomes in other professions and thought experiments in philosophy you may indeed think that philosophy is a truly unique profession. However, I’m inclined to think that this contrast is exaggerated. I’m unconvinced by your characterisation of thought experiments in philosophy, which point to an ideal that few if any actual philosophical thought experiments live up to. It’s also worth noting a point that I make early on in the paper, which is that some philosophical appeals to intuitions aren’t made in the context of thought experiments at all. One final point: Thought experiments in philosophy can straightforwardly be recast as predictions of uncertain outcomes. If I conduct a thought experiment, and conclude that p, and if I believe that other professional philosophers share my intuitions, then I will be entitled to predict that other philosophers who also conduct the same thought experiment will also conclude that p.

Anyway, thanks for pushing me to think these issue through.

[Joachim Horvath]

Dear Steve,

sorry for responding so slowly, which is surely much too slow for the usual pace of the blogosphere!

Concerning your first point, I think I would now try to offer a more detailed story (which I haven't worked out yet) about how precisely philosophical thought experiments work and how exactly this supports the claim that improving some of the relevant abilities might also improve our intuitive judgments about the cases in question. This more detailed story then would give rise to a quite specific and empirically testable hypothesis, one that is tailor-made to philosophical thought experimentation. Testing this hypothesis directly would be much more satisfying than all these analogical arguments from financial planners, psychotherapists and the like, which still strike me as importantly different in too many ways.

To your second point: of course it would not follow that philosophy is truly unique, but the judgments we make about hypothetical cases may indeed be psychological special in important ways. Nevertheless, other disciplines, like linguistics or the psychology of concepts, arguably rely on the same kind of judgments.

Your final point strikes me as off the mark. Predicting whether other philosophers share my intuition about some thought experiment is surely a prediction of an uncertain outcome, but it crucially relies on the premise that other philosophers generally share my intuitions. So, it is, first, clearly an inferential and not an intuitive judgment, and it is, second, not an armchair judgment, because no reasonable defender of intuitions would claim that we are entitled to such sociological assumptions on the basis of armchair intuitions.

Our predictable disagreement notwithstanding, thanks a lot for engaging in this debate with me and other "intuition-friends"!