So ... How are we to understand appeals to intuitions in philosophy? Whose intuitions are being appealed to when philosophers state that some claim or position is intuitive?
Proponents of “Experimental Philosophy”, a burgeoning movement in (mainly) English speaking philosophy, interpret at least some of these claims about intuitions to be hypotheses about the pre-theoretic judgements of non-philosophers. Further, philosophers who identify with Experimental Philosophy, or X-phi, typically hold that assessing claims about “what is intuitively plausible” or whether “we intuitively judge X to be a Y” is best done not simply through traditional armchair reflection but by going out into the world and conducting systematic experimental studies to determine just what and how the “folk” think (Knobe, 2007 : 96). Here we will provide an overview of some of the important trends in X-phi and the nature of the challenges that they raise for appeals to intuition in philosophy.
Experimental Philosophy is not a deeply unified research project, a fact that is sometimes missed by its critics (Nadelhoffer & Nahmias, 2007 : 124). It is rather a loose collection of different projects with diverse aims and approaches that find what unification they have through their broad acceptance of the claim that empirical data are important for philosophical work. Further, and importantly, philosophers and psychologists who identify their work with X-phi are actively involved in running experiments and analysing the resulting data themselves. This latter point about experimental engagement serves to distinguish Experimental Philosophy from other kinds of philosophy that may be deeply informed by the natural and social sciences (Nadelhoffer & Nahmias, 2007 : 124). Although work in X-phi is rather diverse in it's ends as well as in its objects of study – there has been experimental work carried out on, among other things, ethics, causation, and free-will – there has been a fairly constant methodological thread running through most of the past and current work. This is the heavy use of surveys and polling (Cullen, 2009 : 1-4). There is no principled commitment to any particular methodology though, any and all experimental techniques are available for use by Experimental Philosophers. For Example, Joshua Greene's celebrated work in moral psychology and meta-ethics (Greene, 2002) makes heavy use of neural imaging data from his fMRI studies and Tamler Sommers has recently (2010 : 210) suggested that Experimental Philosophers could get philosophical mileage from designing behavioural experiments that might shed light on, inter alia, how subjects' explicit moral judgements might correspond, or fail to correspond, with behavioural responses.
The empirical work on the cluster of questions surrounding free will, determinism, and moral responsibility has been conducted almost exclusively through surveys ( for examples see Nahmias, Morris, Nadelhoffer, & Turner, 2004, 2006; Knobe & Nichols 2007; Feltz & Cokely, 2008; ). We now turn our attention to this work.
The seminal studies of Nahmias, Morris, Nadelhoffer, and Turner (NMNT) provide a useful entry point into the field (Nahmias, Morris, Nadelhoffer, & Turner, 2004,2006).
In their work NMNT provide a number of examples where philosophers such as Robert Kane, Galen Stawson, and Thomas Pink have claimed that the folk are “natural incompatibilists”. By making the claim that their positions are in accord with common sense incompatibilists but the burden of proof on compatibilists to show both why our common-sense intuitions are wrong as well as why we would have misleading intuitions in the first place ( Nahmias, Morris, Nadelhoffer, & Turner, 2006 : 564). NMNT then argue that philosophers who make claims that their theories are somehow supported by, or are in accord with, our everyday intuitions are making claims that can and should be tested empirically.
In order to test these claims about people's natural compatibilism, NMNT ran a series of experiments that were designed to “directly probe” (Sommers 2010 : 201) the folk's intuitions on these matters. In one of these experiments participants were presented with a vignette which described a world in which a Laplacian version of Determinism holds. It the vignette they described a world in which scientists have built a supercomputer that is able to accurately predict events that will happen in the future. These scientists then use this computer to determine, twenty years before he is even born, that a man named Jeremy Hall will rob a particular bank at a particular time. As predicted, Jeremy Hall robbed the predicted bank at the predicted time. Subjects were then asked a series of questions meant to interrogate their intuitions about Jeremy's freedom and responsibility. When they were asked whether they thought that Jeremy acted of their own free will, 76% of the participants answered that he did. Further, when questioned directly about whether Jeremy was “morally blameworthy for robbing the bank”, 83% of the participants responded that he was indeed responsible (Nahmias, Morris, Nadelhoffer, & Turner, 2006 : 567-568).
NMNT interpret these responses as being in favour of compatibilism. While they are careful to stress that they do not think that their results are any bearing whether or not any particular account of free-will is true, NMNT argue that their results potentially undermine the claims of “naturally incompatibilism” and argue that the burden of proof for motivating their theories of free-will (at least in the case of the more “metaphysically demanding” libertarian accounts) should therefore fall on the shoulders of the incompatibilists.
NMNT's experimental work is an attempt to get at the actual content of the folk's intuitions and concepts. As such it falls into the broad category of Experimental Analysis, or what Kauppinen calls “positive experimentalism” (Kaupinnen, 2006). Experimental Analysis comprises of work intended to explore, through empirical methods, the contents of the folk's intuitions and concepts and then determine the relevance of the results for current and future work in philosophy (Nadelhoffer & Nahmias, 2007 : 126). The philosophical relevance of these data might be, as it is with NMNT's work, determining burden of proof through to providing the raw data for revisionist theories of free-will or ethics. Whatever the project, if it makes reference to common-sense concepts or intuitions, Experimental Analysts think that it is somewhat premature to come to philosophical conclusions without first determining the content of these concepts and intuitions empirically (Nadelhoffer & Nahmias, 2007 : 126).
Experimental Descriptivists, while sharing an interest in determining the content of folk intuitions, are additionally interested in understanding the psychological mechanisms and processes through which these intuitions are generated ( Nadelhoffer & Nahimais, 2007: 127). Knowledge about these psychological mechanisms are then used in the service of philosophical argumentation to support, or attack, “first-order” philosophical theories (Nadelhoffer & Nahimais, 2007: 127). Consider recent work in the free-will debate by Knobe and Nichols (2007). They presented participants in their experiment with a scenario describing a universe (universe A) in which
“everything that happens is completely caused by whatever happened before it … so whatever happned at the beginning of the universe caused whatever happened next, and so on right up until the present. For example, one day John decided to have French fries for lunch. Like everything else, this decision was completely caused by what happened before it. So, if everything in this universe was exactly the same up until John made his decision, then it had to happen that John would decide to have French fries” (Knobe & Nichols, 2007).
All participants were assigned to one of two groups. The first group was asked whether, in universe A, “it is possible for a person to be morally responsible for their actions?” to which 86% of the group responded in the negative (Knobe & Nichols, 2007). The second group was presented with a further vignette describing a man in universe A who falls in love with his secretary and who kills his family by burning his house down while they were trapped inside in order to be with her. When members of this second group were asked to judge whether the man was morally responsible for his actions to which 72% of the group responded in the affirmative (Knobe and Nichols 2007). The experimental method used in this study follows that of the NMNT study described above, however, the aim was not just to test a hypothesis about the content of folk intuitions but rather to test a theory about the way in which those intuitions are generated. Knobe and Nichols hypothesised that compatibilist intuitions might be triggered by “affect-laden” scenarios and, indeed, this hypothesis seems to be borne out in their results – the scenario in which the man kills his family was designed to affect the participant's emotions which then drew an overwhelmingly compatibilist response. Knobe and Nichols then try and account for their results by advancing a theory in which compatibilist intuitions reflect a kind of “performance error” (Appiah, 2008 : 102) and therefore, contrary to NMNT's conclusions, shouldn't count in compatibilism's favour.
One notable feature of the experiments already discussed is that the responses aren't unanimous. For the 72% of participants who gave compatiblist answers in Knobe and Nichols' “high affect” scenario there was a substantial dissenting minority of participants who didn't. While Knobe and Nichols' affective performance error theory might be able to account for the general trend towards compatibilism we still stand in need of an explanation of why we see such a high number of responses for incompatibilism. Explaining these kinds of results has opened up a line of research that attempts to uncover correlations between differences in individuals and their intuitions. For example, in a 2009 study Feltz and Cokely set out to investigate whether and how differences in peoples' personalities were related to their intuitions concerning determinism and attributions of moral responsibility. Before presenting participants with surveys similar to the experiments already discussed, participants were administered a brief version of the Big-Five personality test. The Big-Five model of personality identifies five broad traits (Extroversion, Neuroticism, Agreeableness, Conscientiousness, and Openness to experience) that appear as common factors across a range of different personality scales and that seem to be stable across different cultures (de Bruin, 2005 : 159). Feltz and Cokely found that measures of extroversion of personality were a reliable predictor of compatibilist responses to their survey questions (Feltz & Cokely, 2009). Among the possible interpretations that they suggest that their data support is the notion that there might not be a homogeneous set of folk intuitions at all, and further, that the long-standing debate about free-will that we see in professional philosophy could in part be due to differences in philosophers' personalities triggering different intuitions about free-will and moral responsibility (Feltz & Cokely, 2009 : 7).
Considerations along these lines, experimental results that show that intuitions my vary along socio-economic and cultural lines (Nadelhoffer & Nahmias, 2007 : 124) are sometimes thought to be reason enough to be suspicious of the use of intuitions in philosophical argumentation in general. This is the broad claim of the Experimental Restrictionists who are in some respects the most radical of the proponents of Experimental Philosophy. The central claim of Experimental Restrictionism is that empirical evidence showing the diversity and unreliability of intuitions should undermine philosophers' confidence in both appeals to the intuitions of the folk and to the more “discriminating” intuitions of philosophers.
Appiah, K. A. (2008). Experiments in Ethics. Harvard University Press : Cambridge, Massachusetts.
Cullen, S. (2009). "Survey-Driven Romanticism" Rev.Phil.Psych
de Bruin, G. P. (2005). "Personality Assessment" in An introduction to Psychological Assessment in the South African context Foxcroft, C. & Roodt, G. (eds) Oxford University Press Southern Africa : Cape Town.
Feltz, A. Cokely, E., T., (2009). "Do judgments about freedom and responsibility depend on who you are? Personality differences in intuitions about compatibilism and incompatibilism." Consciousness and Cognition 18: 324-350.
Greene, J. (2003). "From neural 'is' to moral 'ought' : what are the moral implications of neuroscientific moral psychology?' Nature Reviews : Neuroscience Vol. 4 : 847 – 850.
Kauppinen, A. (2007). "The Rise and Fall of Experimental Philosophy", Philosophical Explorations Vol. 10, No. 2: 95-118.
Knobe, J. (2007). "Experimental Philosophy and Philosophical Significance", Philosophical Explorations 10(2): 95-118.
Nadelhoffer, T., Kvaran, T., & Nahmias, E. (2009) "Temperament and intuition : A commentary on Feltz and Cokely" Consciousness and Cognition
Nahmias, E., Morris, E. G., Nadelhoffer, T. & Turner, J (2004). "The Phenomenology of Free Will" Journal of Consciousness Studies 11(7-8): 162-179.
Nahmias, E., Morris, E. G., Nadelhoffer, T. & Turner, J (2006). "Is Incompatibilism Intuitive?" Philosophy and Phenomenological Research Vol. LXXIII : 28-53.
Nadelhoffer, T., Nahmias, E. (2007). "The Past and Future of Experimental Philosophy" Philosophical Explorations. Vol. 10, No 2.
Sommers, T. (2010). "Experimental Philosophy and Free Will" Philosophy Compass. 5/2 : 199-212.