Kyle Thompson

Ethics isn’t a conversation exclusive to philosophers. There is value, then, in not only understanding how laypeople think about issues in ethics, but also bringing their judgments into dialogue with those of philosophers in order to make sense of agreement, disagreement, and the consequences of each. Experimental philosophers facilitate this dialogue uniquely by capturing laypeople’s judgments and analyzing them in light of philosophical theory. They have done so almost exclusively by using face valid quantitative surveys about philosophically interesting thought experiments. Based on high participant support for this or that response, researchers conclude that a given theory is more or less intuitive, and in some cases that it is or isn’t true. However, such conclusions can only be drawn from quantitative survey data if one assumes those data accurately reflect laypeople’s thinking on the issues of interest, an assumption that can’t be justified by the quantitative data itself since they are potentially opaque. This is the methodological problem of experimental philosophy: experimental philosophers who use quantitative methods alone can’t sufficiently demonstrate their data to be reflective of relevant judgments/intuitions. To better explore people’s judgments on issues in ethics and to determine whether experimental philosophers are or aren’t getting what they take themselves to be getting in surveys, I recreated part of an experiment by Chituc, Henne, Sinnott-Armstrong, and De Brigard on the Kantian principle that “ought implies can”, or that whenever an agent ought to do something then it is the case that she can, and added a layer of qualitative data to it by having participants think aloud while completing the surveys and conduct a follow-up interview. With similar quantitative results to those from Chituc et al., I argue that my qualitative results 1) cast doubt on Chituc et al.’s data being reflective of many people’s judgments, especially on the more interesting questions relating to moral obligation; 2) reveal unique insights into how people think about OIC; 3) generally speak to the value of experimental philosophers using triangulation of methods in better understanding laypeople’s complex and varied thinking about questions important to ethics and beyond. Supplemental documents: Appendices A, B, C, D, and E are surveys I used in my study. Appendices C and D are derived from Chituc et al.’s publication “Blame, Not Ability, Impacts Moral ‘Ought’ Judgments for Impossible Actions: Toward an Empirical Refutation of ‘Ought’ Implies ‘Can.’” Appendix F is the recruitment flyer I distributed for my study. Appendix H is the quantitative data from my study.

Experimental philosophers rely almost exclusively on quantitative surveys that potentially misrepresent participants’ multifarious judgments. To assess the efficacy of qualitative methods in experimental philosophy and reveal limitations with quantitative surveys, a study was conducted on the Kantian principle that ‘ought implies can’, which limits moral obligation to actions that agents can do. Specifically, the think aloud method and a follow-up interview were employed in a modified version of a prominent experiment that recorded participants’ judgments of ability, blame, and obligation using quantitative surveys. The modified version produced quantitative results similar to the original experiment along with qualitative data that reveal that the surveys fundamentally misrepresented participants’ judgments. The qualitative transcripts from 40 participants are analyzed to show that ‘ought implies can’ judgments are complex and multifarious, that ‘ought implies can’ judgments are misrepresented by quantitative survey questions, and that the majority of participants uphold or preserve ‘ought implies can.’ The results suggest that experimental philosophers can more accurately capture judgments by using qualitative methods, and that studies which rely on quantitative surveys possibly misrepresent participants’ judgments.