Is reasoning responding to reasons?

Click to increase image sizeClick to decrease image size

Acknowledgments

I am grateful to John Broome, John Hyman, Julian Fink and Constantine Sandis for very helpful comments on earlier versions of this paper.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes on contributor

Franziska Poprawe received an MA in Philosophy and Economics from the University of Bayreuth (Germany) in 2016. She is currently completing her DPhil in Philosophy at the University of Oxford, and holds a research position at the University of Zürich.

Notes

1 The three views are mostly regarded as mutually exclusive, with the exception that McHugh and Way present their view of reasoning as an activity being guided by the aim of getting fitting attitudes as a kind of rule-following view. Further, a few authors assume that when you follow a rule in reasoning, you thereby count as responding to (the premises as) reasons (for the conclusion), thus committing to both, the Rule-Following View and the Reasons View (especially Pettit [2007, 2016], Boghossian [2014]; and Grice [2001]).

2 Kauppinen (2018) and McHugh and Way (2018a) also make this observation. Among those who assume or in passing commit to some version of the Reasons View are Pollock (1987, 481–482), Grice (2001, ch.1), Korsgaard (2009, 30), Scanlon (1998, 55–56); Audi (2004, 119, 126), Hacker (2007, 3, 203), Schroeder (2007, 26–28); Alvarez (2010); Horty (2012, Introduction), Kur Sylvan (2016, 3.3), and Shah (2006, 486). Raz (2011, 2015) and Kauppinen (2018) also offer detailed defences of the Reasons View, focussing on practical reasoning.

3 Unless otherwise indicated, page numbers from now on refer to Dancy’s (2018) “Practical Shape”.

4 This is not to say that it is possible to reason to a response R on the basis of considerations that play no role in favouring R in the eyes of the agent. As said, many assume (and I agree) that it is essential to reasoning that the agent somehow sees his response as being favoured or supported by the premises, whatever this attitude of seeing exactly amounts to. But Dancy's focalist account makes no mention of the possibility that while the response may not actually be favoured by the considerations adduced, it may be favoured in the eyes of the agent.

5 McHugh and Way (2018b) argue that reasoning is a goodness-fixing kind, not a correctness-fixing kind. They argue that reasoning has an aim, namely the aim of getting fitting attitudes. However, the metaphorical claim that reasoning has an aim is neither essential to their account of the nature of reasoning (according to which reasoning is a rule-following activity), nor is it essential to their account of good reasoning. On this, see also Broome (2015, 201–202), who gives an example that “shows that the point of reasoning is to acquire attitudes that stand in an appropriate relationship to each other, not merely attitudes that are correct.”

6 Dancy seems to use “reasons” and “favourers” interchangeably (e.g. 38, 45, 90, 141).

7 I here say that we cannot accept the quoted passage (from page 34) as giving a sufficient condition for correct reasoning, reading the first “when” as a “whenever”. This seems a plausible reading, though, of course, it is at odds with Dancy's overall focalist approach.

8 Kauppinen (2018) defends this view; McHugh and Way (2018a) criticise a version of it.

9 Jackson and Pargetter (1986) and Broome (2001) argue this way, see Kiesewetter (2015) for a discussion.

10 For simplicity, I restrict my discussion to instrumental reasons for taking necessary means. More generally, we could define an instrumental reason to Ψ as the fact that there is a positive probability, conditional on one's Ψ-ing, that one's Ψ-ing or some part of one's Ψ-ing helps to bring about that one φ's (Kolodny 2018). Though I accept this proposal, I doubt Kolodny's claim that if you have some pro-tanto reason for φ-ing, and Ψ-ing helps you to φ, then you have some reason for Ψ-ing. Instrumental reasons are conditional on the end being permissible, that is, they are conditional on non-instrumental reasons for the end, in a similar fashion in which prudential reasons are arguably conditional on moral reasons on Kant's view (Bader 2015). Korsgaard (1997) makes a similar point in her argument against the Humean claim that all reasons are instrumental.

Additional information

Funding

This work was supported by Hanfling Scholarship, University of Oxford and Studienstiftung des Deutschen Volkes. and The project “The Structure and Development of Understanding Actions and Reasons”.