Constantine Sandis (University of Hertfordshire)

This week I give you the 25th set of answers in our series of mini-interviews. It is a real pleasure to publish at this point the answers of one of the co-founders of this page, Prof Constantine Sandis from the University of Hertfordshire. (We have also interviewed our other founder Andrei Buckareff earlier.) Prof Sandis has published on most of the core topics of action theory like causal and non-causal explanation, reasons, concepts of action and explanation, co-edited the Philosophy of Action: An Anthology with Jonathan Dancy, and A Companion to the Philosophy of Action with Tim O’Connor, besides several other volumes, and his Character and Causation: Hume’s Philosophy of Action is forthcoming soon. Besides, he hosted Bloomsbury’s Why Philosophy Matters lecture series. Enjoy!

1. How did you first become interested in Philosophy of Action?

In my third year as an undergraduate student I studied Philosophy of Mind with Peter Hacker. After the standard 8 tutorials, he asked me if I would like to have four more tutorials on issues relating mind to action. This was characteristically kind of him, and I proceeded to write essays on action, intention, voluntariness, and reasons. He instructed me to read Anscombe and Davidson, of course. But also Kenny and, most importantly, Alan R. White, whose introduction to criminal law, Grounds of Liability, had a particularly huge influence on me, as had his paper 'What We Believe' when I was writing an earlier essay on belief. So, it was really questions in the philosophy of criminal law about the concepts of voluntariness and the intentional that got me interested in the philosophy of action. H.L.A. Hart writing about Glanville Williams anticipated much of what experimental philosophy would later try to say about intentional action, but he did so from the proverbial armchair. I still believe that philosophy of action has more to learn from criminal law than the other way round, though it is definitely a two-way process.

I later wrote my MA dissertation on the so-called Weakness of Will, before proceeding to wrote a PhD on the explanation of action, with Jonathan Dancy as my supervisor. By then I had become really involved in the debate on the ontology of reasons for action. But it struck me as odd that those involved had almost nothing to say about either action or explanation. So, I tried to argue that one cannot place explanatory constraints on the notion of a reason for action without saying more about the nature of both what it is meant to explain and how it is meant to do so.

While I had read, 'Action, Reasons, and Causes' as an undergraduate I was struck to discover, as a graduate student, just how prevalent the view that all actions are intentional under some description was. This seemed counter-intuitive to me since clearly one may do some things intentionally and others unintentionally. I struggled with this for some years but eventually came to defend the view (in my book The Things We Do and Why Do Them) that while the event of my doing one thing intentionally may be identical to the event of my doing some other thing unintentionally, such events are not themselves intentional under any description.

2. What are you working on at the moment?

I just finished writing a book on Hume's philosophy of action last week and editing another book, which will be out with Routledge in a few months. I now have a number of other projects I'm just returning to, including a collaboration with Microsoft Research on the explanation of AI behaviour. I'm also working on way too many new books. One is a tradebook on mistakes and why we make them, another explores what it takes to understand oneself and others, and a third is a long overdue textbook on the philosophy of action.

But the one that has been taking up most of my recent time attempts to demonstrate the relevance of philosophy of action for normative and practical ethics. I argue that it is quite remarkable how all the main theories of right action say next to nothing about what they take an action to be. My suspicion is that different concepts and conceptions of action render different normative theories more or less plausible, but that the proponents of such theories all assume that they either share a common notion of action or that one can simply plug in one's favourite action theory without this affecting the debate in any way. Take away this assumption, and the whole of normative (and by extension applied) ethics needs to be considered anew.

3. What is your 5-15 sentence account of what an action is?

I think that any such account is bound to fail, so let me try to tell you why in about 10 (or a couple more) sentences. In ordinary language, the word 'action' is ambiguous. It can refer to either what one does or to one's doing (of) it. Some philosophers of action use the words 'doing' and 'deed' (or 'thing done') to disambiguate between the two. These should be treated as technical terms marking important conceptual space, since in ordinary language all of these terms retain the original ambiguity of 'action' to a relatively high degree. So, I think there are at least two different concepts of action, each equally legitimate and important for different reasons.

To complicate matters more, different philosophers have different conceptions of each of these two things. They might thus distinguish them as processes and products, types and tokens, universals or particulars, and so on. And they may have further debates about the ontology of any of these things e.g. whether act tokens are really processes or events. On my view, even these debates are misguided though, because neither doings nor things done are events. We may, of course talk of the event of my doing something, just as we might talk of the process of my doing that very same thing. But the 'of' here marks a relation that is not one of identity. Thus, we can have both processes and events of people acting, neither of which are either doings or things done.

I suppose one might be able to extract the negative view that actions are neither events nor processes from all this. But I'm not too fussed about such pronouncements since we could equally say that in philosophy the term 'action' is used technically, to refer to the 'event' of someone acting (where 'event' is arguably also used technically). In short, I am a conceptual pluralist about actions. The important thing is not to say what actions are or are not, but to not conflate all these different senses and to note that it makes sense to predicate certain things of actions in one sense of the term, but not others. The book I'm working on relating action to normative ethics is in part an exploration of which sense of action is best suited to deontic predicates and which to evaluative ones. It's also interesting, I think, to ask whether the sense of 'action' in which these things can have consequences is the same as that in which they can be done intentionally.

4. In your view, what were the three most important recent developments in philosophy of action?

I shall treat 'recent' to mean post-Wittgensteinian. The first is Anscombe's insight that a single action can have multiple descriptions, each revealing different psychological aspects. Anscombe doesn't distinguish doings from things done, but her examples work much better with the former and I think that this is what she really has in mind.

The second important development, then, is Jennifer Hornsby's distinction between doings and things done. While versions of this distinction existed in the early 20th century, Hornsby places an emphasis on the difference between actions that are repeatable (things done) and ones that are non-repeatable particulars (doings). This is important for both action explanation and action justification.

Finally, I would like to highlight Fred Dretske's distinction between triggering and structuring causes of action. Dretske is mostly praised for his work in epistemology, while his work in the philosophy of action was unfortunately overshadowed by Davidson's popularity. Dretske understands action as the process of perception triggering bodily movements. Strictly speaking, triggering and structuring causes have two distinct explanantia, but Dretske wasn't always very clear about this. To explain an action, is not to look for the cause of a bodily movement but to ask why some other phenomenon triggered the movement in question. This eliminates all sorts of worries about deviant causal chains. If it seems pessimistic of me to highlight 'recent' developments which all occurred in the previous century, let me say that I think it's next to impossible to know without the benefit of hindsight what will prove to be an important development and what is a passing fad. The current obsession with process views of action is an interesting case in point.

5. What direction would you like to see the field go in?

It would be weird, I suppose, for one's answer to this question to be completely unrelated to that given to question 2. Hence, I'd like to see a much closer connection between philosophy of action and normative ethics, and I'd like to see a move away from pure action theory to a more conceptually and historically sensitive approach. It would also be nice to have a more integrated philosophy of action that combines insights from neuro-physiology and social psychology with historical sensitivity and conceptual clarity, not least in relation to understanding AI.

Finally, I'd like to see more philosophers take seriously the sociality of everyday action. One of my graduate students, Katja Behrens, is completing her thesis that is partly about how contemporary accounts of collective action and shared agency are premised on a problematically individualistic or atomistic conception of what it is to act. I'd like to see more work that treats social action as the norm and more isolated actions, such as those of Robinson Crusoe, as the ones in need of special explanation. This would involve returning to the connection between action and rule-following that Peter Winch highlighted in The Idea of a Social Science.

2018 August 4th

Thanks to Prof Sandis!

Check back next weekend on the 11th for our next set of answers.