Pragmatic experimental philosophy

Abstract

This paper considers three package deals combining views in philosophy of mind, meta-philosophy, and experimental philosophy. The most familiar of these packages gives center-stage to pumping intuitions about fanciful cases, but that package involves problematic commitments both to a controversial descriptivist theory of reference and to intuitions that “negative” experimental philosophers have shown to be suspiciously variable and context-sensitive. In light of these difficulties, it would be good for future-minded experimental philosophers to align themselves with a different package deal. This paper suggests two alternatives. Experimentalists might help fans of “naturalized” approaches discover what natural kinds have been playing an appropriate role in causing us to use concepts as we do. Or, better still, experimentalists might instead help pragmatists and teleo-semanticists discover how our concept usage regularly yields beneficial outcomes, so that we can then craft philosophical analyses that will enable us to yield such beneficial outcomes more consistently. Using free will and explanation as instructive examples, this paper offers concrete guidance and suggestions for how experimental philosophers can pursue new positive projects that will be both pragmatically and philosophically useful.

Keywords::

• Descriptivism
• Experimental Philosophy
• Explanation
• Free Will
• Meta-Philosophy
• Naturalized Semantics
• Reference
• Teleosemantics

Acknowledgements

I am grateful to helpful comments from many people, including Shaun Nichols, Ron Mallon, Joshua Knobe, Tania Lombrozo, and other participants at the 2009 NEH summer institute on experimental philosophy and the 2011 bootcamp on experimental philosophy of free will. I am also grateful for helpful feedback from my colleagues at SMU, and audiences at the Eastern APA (including helpful commentary from Thomas Nadelhoffer), the Southern Society for Philosophy and Psychology, and Florida International University.

Notes

^[1] I think of concepts as mental particulars that play a folder-like information-coordinating role in cognition. In this, I follow a rich tradition in cognitive psychology and the philosophy of cognitive science. For a good introduction to this tradition see Margolis and Laurence (1999), and for a view of concepts very similar to my own see Millikan (1998, 2000). This tradition may be contrasted against an equally rich philosophical tradition that takes concepts to be abstract entities.

^[2] Descriptivist views include Frege (1892/1996), Russell (1996), and Strawson (1996).

^[3] The exodus from descriptivism was spurred in part by classic examples by Burge (1979), Kripke (1972), and Putnam (1996). These include examples where people don't have in mind descriptions sufficient to pick out their referents uniquely (e.g., Kripke's Feynman case, Putnam's Twin Earth), and examples where the descriptions people do have in mind don't even fit their actual referents (e.g., Kripke's Gödel/Schmidt case, Burge's arthritis case).

^[4] Causal/Informational approaches include Boyd (1988), Evans (1973), Fodor (1990), Kripke (1972), and Rupert (1999).

^[5] Teleo/Pragmatic semantic theories include Appiah (1986), Blackburn (2005), Dretske (1988), James (1907), Millikan (1984), Papineau (1987), Ramsey (1931), and Whyte (1990).

^[6] In a moment we'll see that “nay-saying” experimental philosophers have further called into question the reliability of some particular philosophical intuitions by showing that these intuitions vary across people and/or contexts.

^[7] Some contemporary philosophers intend to do just that, including Chalmers (2002, 2004) and Lewis (1984).

^[8] Philosophers have proposed “naturalized” analyses of many philosophical concepts, including knowledge (Kornblith, 2002; Quine, 1996), emotion (Griffiths, 1997), color (e.g., Hilbert, 1987), consciousness (Dennett, 1991), and moral goodness (e.g., Boyd, 1988).

^[9] Other articulations of Pragmatic Conceptual Analysis are given by Craig (1986, 1990), Woodward (2003), and my own Fisher (2006).

^[10] See, e.g., Nichols and Knobe (2008).

^[11] Nay-Sayers include Machery, Mallon, Nichols, and Stich (2004); Mallon, Machery, Nichols, and Stich (2009); Swain, Alexander, and Weinberg (2008); Weinberg (2007); Weinberg, Nichols, and Stich (2001), and Alexander, Mallon, and Weinberg (forthcoming), which is appropriately entitled “Accentuate the negative.”

^[12] Some advocates of intuitive conceptual analysis have indicated openness to having experimentalists gather intuitions, including Jackson (1998) and Lewis (1972). In their “Experimental philosophy manifesto,” Nichols and Knobe (2008) emphasize the continuity of experimental philosophy with traditional philosophical use of intuition. Many experimental philosophers have drawn positive philosophical conclusions from the intuitions they evoke, including Griffitths and Stotz (2008), Nahmias et al. (2006), and Roskies and Nichols (2008).

^[13] As this paper has surely made clear, I don't trust intuitions. However, I have no official views regarding just how surprising and variable our intuitions are, so I can't predict how many more successes Nay-saying experimentalists will end up enjoying. Nay-saying experimental methods can detect inconsistency which is one strong indicator of unreliability, but, as I argued in section 2, we should also doubt the reliability even of consistently held intuitions. Any successes that Nay-Sayers enjoy will provide all the more reason not to trust intuitions, but my own mistrust of intuitions doesn't depend upon nay-saying experimental results.

^[14] Papers that draw a similar distinction between positive and negative programs include Alexander et al. (forthcoming), Alexander and Weinberg (2007), Kauppinen (2007), Nadelhoffer and Nahmias (2007), and Weinberg (2007).

^[15] Unless we've regularly applied a concept in ways that, upon reflection, we find counter-intutive—a real possibility.

^[16] For much discussion of these and related questions, see Nahmias and Murray (2010) and Nichols and Knobe (2007).

^[17] For many purposes, we'll want to construe whatever we think we have reason to pursue as “beneficial.” If we construe “benefits” in this way then Pragmatic Conceptual Analysis will yield explications that will help us more consistently to achieve what we think we have reason to pursue. For certain explanatory purposes, it may be useful instead to construe as “beneficial” whatever is rewarded and/or selected for by the learning and/or evolutionary processes that have shaped the conceptual systems in question. If we construe “benefits” in this way, then Pragmatic Conceptual Analysis yields explications that enable us to give good explanations for the characteristic patterns of behavior and success and failure that these conceptual systems produce. See Fisher (2006; unpublished manuscript, chapter 3).

^[18] Indeed, one might hope to find relevant research on these topics from the various non-philosophers who have studied deterrence. However, in many cases, philosophers will find that the philosophically relevant experimental questions won't have yet been asked. For example, much of the research on deterrence has asked which people are generally deterrable. In contrast, folk usage allows that a given person may produce some actions that are free and others that are unfree. Understanding the usefulness of this practice will require asking which actions are deterrable in a given (sort of) person. When we need data that aren't yet available from other experimenters, philosophers may need to get involved in designing experiments. (We'll return to such division-of-labor questions in the concluding section.)

^[19] It may be worth noting that this enquiry could yield some surprising results. For example, suppose it turns out that offering rewards and punishments does affect the activities of certain sleep-walkers (or phobics, or obsessive-compulsives, or hypnotized people, or drug addicts) in much the same way that it does ordinary “free” actions. If so, then Pragmatic Conceptual Analysis would likely end up counting these activities as “free,” regardless of intuitions to the contrary. Similarly, it might turn out that, surprisingly enough, many of our habitual (or culturally inculcated, or even heroic) activities aren't susceptible to rewards and punishments in the right ways for them to count as “free,” regardless of intuitions to the contrary.

^[20] See note 5.

^[21] This, itself, is an empirical claim, and could be subject to experimental testing.

^[22] Additional evidence for this comes from experiments on illusory correlations, e.g., Wright and Murphy (1984).

^[23] For further evidence, see Sloman (1994).

^[24] This may involve some sort of default presumption that functions are multiply realizable, so that category-members who don't realize the functional effect E the most common way (via C) should instead realize it in some alternative way; whereas mechanistic explanations take an effect to be comparatively unlikely to occur in a category member without its normal cause.

^[25] See also Ahn (1998); Jameson and Gentner (2008).

^[26] This has been evidenced regarding learning in multiple domains, including physics (Chi, Bassok, Lewis, Reimann, & Glaser, 1989), biology (Chi, de Leeuw, Chiu, & Lavancher, 1994), mathematics (Rittle-Johnson, 2006; Wong, Lawson, & Keeves, 2002), folk psychology (Amsterlaw & Wellman, 2006), and even fictional robots (Williams & Lombrozo, forthcoming).

^[27] Some defense and empirical support for this hypothesis is given by Gopnik (1998).

Additional information

Notes on contributors

Justin C. Fisher

Justin C. Fisher is Assistant Professor of Philosophy at Southern Methodist University.