The blog post today is by Sophie Keeling on her recent paper "Responding to Second-Order Reasons" (Philosophy and Phenomenological Research, 2024). She is a 'Ramon y Cajal' fellow in philosophy at UNED, Madrid and a member of the Metis research group.
 |
| Sophie Keeling |
Philosophers love talking about reasons. Often this takes place at the first-order level: reasons for attending a party include the fact that it would be fun or that it would make your friends happy if you went. Reasons for believing that it will rain tomorrow include the weather forecast or the fact that it’s rainy season where you are. But arguably, some reasons are second-order. These are reasons to respond to a particular reason or not to respond to it.
For example, I might promise my therapist not to do things simply because it will please others and to only think about myself. And in the epistemic case, however good your reasoning in fact was, the fact that you carried it out under a shortness of oxygen or while drunk would give you reason to exclude otherwise good reasons from consideration. Of special interest to readers of this blog is this idea that supposed evidence about your reasoning capacities could function in this way, as reasons to put aside the evidence that you’ve assessed with them. This could include others telling you that you’re always over-pessimistic or -optimistic, or clinical diagnoses.
The way in which we in fact respond to reasons is hotly debated, albeit without agreement. But little has been said about responding to second-order reasons. And if this isn’t in fact possible, then this gives us reason (no pun intended!) to doubt their existence. Whiting (2017) argues that reasons can’t play their important guiding role if we can’t actually respond directly on the basis of them. Instead, Whiting argues that second-order reasons are in fact just common garden first-order reasons to perform certain actions such as those involved in focusing your attention so as to make it the case that you respond to one reason over another, e.g. trying reeeaaaally hard to push thoughts about your friends out of your mind in the hope that they have less effect on you. And I guess it does sound pretty weird – we’re familiar with believing, acting, desiring for reasons – but responding to a reason for a reason?!
Ever one to embrace a challenge, in this paper I argue that it does in fact make sense to talk about responding to a reason to respond to a reason, and start to sketch out what an account of responding to second-order reasons would look like. I term this the second-order basing relation. There’s various desiderata that would need to be met if we were to formulate an account, such as saying how the response relation would be special and somehow different from what goes on at the first-order.
As my starting point, I adapt a hybrid account of responding to reasons. At the first order, this is to say that the reason causes the response and also that one is disposed to believe that it’s a good reason for it (this is to distinguish responding to reasons from other kinds of causes, such as biases or just talking about c-fibres firing). Likewise, the hybrid account of the second-order basing relation would have two components.
First, the second-order reason must cause the first-order reason to play its causal role or block it from having an effect. This is like how flicking a switch can cause various other electrical processes to happen or can block them from occurring. And for this to be a special kind of relation as opposed to just part of the first order reason, we must be disposed to form a special kind of belief.
While what is important for responding to a first order reason is believing that it’s a good reason for the response, the important thing for responding to second-order reasons is believing that the first-order reason is an appropriate kind of reason. ‘Kinds’ of reason could include promissory reasons, self-interested reasons, reasons processed by unreliable mechanisms, and so on – although work remains to be done on this.
And while I use a hybrid account as a toy model, this is so that one can adapt it more easily to suit one’s tastes; one might in the end just appeal to one of the conditions, for example. In short, appeal to appropriate kinds of reason will be the magic key to constructing our account of responding to second-order reasons however we do it in the end.
While currently underutilised, the hope is that now we can make sense of what it would be to respond to second-order reasons, we can put the framework to good use throughout philosophy, such as for understanding the way in which higher-order evidence has its effect.