Abstract
How do reasons combine? How is it that several reasons taken together can have a combined weight which exceeds the weight of any one alone? I propose an answer in mereological terms: reasons combine by composing a further, complex reason of which they are parts. Their combined weight is the weight of their combination. I develop a mereological framework, and use this to investigate some structural views about reasons. Two of these views I call “Atomism” and “Wholism”. Atomism is the view that atomic reasons are fundamental: all reasons reduce to atomic reasons. Wholism is the view that whole reasons are fundamental. I argue for Wholism, and against Atomism. I also consider whether reasons might be “context-sensitive”.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
Why “Wholism”? Mainly to save confusion with another view which is often called “Holism”. More on Holism and its relation to Wholism below, Sect. 3.2.
This needn’t be the case. It might be that all the reasons taken together have no combined force, because they cancel each other out, in which case their fusion will have no force and hence be no reason at all.
This terminology comes from Dancy (2004, p. 16). One might alternatively say “all things considered” (for overall), and “pro tanto” (for contributory).
For example, T. M. Scanlon states his his famous “buck-passing theory of value” like this: “being good, or valuable, is not a property that itself provides a reason to respond to a thing in certain ways. Rather, to be good or valuable is to have other properties that constitute such reasons” (Scanlon 1998, p. 97).
At least this is so when you know sufficiently well what the alternatives are like. If you’re more in the dark, things may be more complicated. It may be that what you ought to do depends not on which properties the alternatives in fact have, but on which you believe them to have. We needn’t resolve such complications here. So let us simply set aside such cases of ignorance.
For a ratio scale, as opposed to merely a cardinal scale, we also need to fix the zero-point. We do this, as stated above, by letting zero be the weight of a non-reason, a property that neither favours nor disfavours.
An automorphism of \(\langle P,\preceq \rangle \) is a permutation \(\pi \) of P such that for all \(p,q \in P,\,p \preceq q\) iff \(\pi (p) \preceq \pi (q)\). A permutation of P is a bijective mapping from P into itself.
Strictly, Atomism says only something weaker: the weight of \(p+q\) reduces to the weights of some atomic reasons, but not necessarily to p and q in particular. It seems natural, however, to think that if a property’s weight reduces to the weights of any atomic reasons, then it will reduce to the weights of the atomic properties of which it is composed.
Because \(\langle P,\preceq \rangle \) is classical, as defined above, it follows that there exists exactly one property q satisfying these two conditions.
This is trivially true, because distinct constant weighing functions cannot be either atom-equivalent or whole-equivalent.
What I mean here is simply that overall reasons trump other reasons. Strictly speaking, other facts—e.g., facts about what you can do—may help to determine what you ought to do. More on this below (Sect. 6.2).
This is not quite right. Below (Sect. 6.2) I suggest another practical need served by reasons: they are useful in explaining not only why we ought to do certain things, but also why we are rational to regret our inability to do certain things. The peatiness of a particular brand of whisky may help to explain why I should regret my inability to drink this whisky if, say, the pub I am in does not have any. What I need to say, then, is that rational regret is also fundamentally a matter of whole reasons. If I have overall more reason to drink a brand of whisky that is available, I should not regret the unavailability of the inferior brand. I’m happy to say this.
Peter Vranas argues in the opposite direction, from the premise that one can only have reason to do what one can do, to the conclusion that one ought to do only what one can do (Vranas 2007). However, he does not argue for the premise, except to say that it has “an almost tautological ring” (Vranas 2007, p. 173). His idea seems to be that, if a reason supported something that could not be done, it could not properly be classed as a “reason for action”. But this seems wrong. Going to the moon, for example, is an action, even if not an action that you can perform.
References
Armstrong, D. M. (1997). A world of states of affairs. Cambridge: Cambridge University Press.
Dancy, J. (2000). The particularist’s progress. In B. Hooker & M. O. Little (Eds.), Moral particularism (pp. 130–157). Oxford: Oxford University Press.
Dancy, J. (2004). Ethics without principles. Oxford: Oxford University Press.
McKay, T. (2006). Plural predication. Oxford: Oxford University Press.
McKeever, S., & Ridge, M. (2006). Principled ethics: Generalism as a regulative ideal. Oxford: Oxford University Press.
Ridge, M. (2007). Anti-reductionism and supervenience. Journal of Moral Philosophy, 4(3), 330–348.
Scanlon, T. M. (1998). What we owe to each other. Cambridge, MA: Harvard University Press.
Schroeder, M. (2007). Slaves of the passions. Oxford: Oxford University Press.
Streumer, B. (2007). Reasons and impossibility. Philosophical Studies, 136, 351–384.
Vranas, P. B. M. (2007). I ought, therefore i can. Philosophical Studies, 136(2), 167–216.
Acknowledgments
Earlier versions of this paper were presented at the annual conferences of the Scots Philosophical Club (2007) and the British Society for Ethical Theory (2008), and at a workshop on Reasons and Rational Choice at the London School of Economics (2011). I would like to thank the audiences for their helpful comments, and especially Mike Ridge and Guy Fletcher.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Brown, C. The composition of reasons. Synthese 191, 779–800 (2014). https://doi.org/10.1007/s11229-013-0299-8
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11229-013-0299-8