Processes and the philosophy of action

Abstract

While the concept event has been an important tool in our thinking about causation and action, the concept process has not been appealed to so readily. However, recently, several philosophers have argued that the distinction between events and processes is of much greater metaphysical significance than previously thought. Many of these philosophers have suggested that recognising process as a distinctive ontological category is important for understanding action and agency. One theory of processes which has become popular is a theory I call the “temporal stuff view” of processes. According to the temporal stuff view, processes are the “temporal stuffs” from which events are composed. I propose an alternative to the temporal stuff view. On my theory, processes are a special kind of universal. I suggest that engaging in a process is analogous to instantiating a property, and that events are instances of processes. On this proposal, a substance’s engagement in a process is a special sort of state of affairs (a dynamic state of affairs). I argue that this alternative theory of processes does better than the temporal stuff view when it comes to helping us understand action and agency.

Keywords:

• processes
• universals
• events
• agency
• action
• causation

Acknowledgements

Greatest thanks are owed to Helen Steward who supervised my PhD. Thanks also to the organisers of the 21st Oxford Graduate Philosophy Conference for allowing me to present an early version of this paper. I particularly want to thank Ursula Coope for her constructive comments. I also want to thank two anonymous referees for this journal for their helpful comments on earlier drafts of this paper.

Disclosure statement

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

Notes on contributor

Andrea White completed her PhD at the University of Leeds in 2018. Her research focuses on metaphysics and philosophy of mind, in particular the metaphysics of action and the problem of mental causation.

Notes

1 For example, Kim, Davidson and Bennett. See especially Kim (1976), Davidson (2001), and Bennett (1988). Salmon (1984), by contrast, does take the distinctive features of processes to be important in understanding causation, but even for Salmon processes are merely series of events.

2 For example, Armstrong (1968, 82).

3 See, for example, Stout (1997, 2016), Steward (2013), Galton and Mizoguchi (2009), Crowther (2011), and Hornsby (2012).

4 See also Crowther (2018) for more specific claims about the relations between the notions of process and stuff.

5 Although Hornsby does not explicitly state that temporal stuffs are concrete, as entities which pervade intervals of time and which compose events, it seems plausible that processes on her view are concrete entities and, just like spatial stuffs, are not entities which are instantiated and so cannot be universals.

6 Crowther (2011) makes a similar argument.

7 I am proposing that the quantification “There is some furniture in here,” is a kind of plural quantification: it is a way of quantifying over many individual items at once, and need not involve any ontological commitment to a kind of stuff. One could deny that “There is some furniture in here,” can be analysed in this way. Alternatively, one could argue that “There is some furniture in here,” quantifies over a kind of stuff in addition to quantifying over a collection of discrete individuals. In other words, “There is some furniture in here,” claims that there are individual items of furniture and there is furniture which is a non-countable entity, of which there can be more or less, which is distinct from the individual items of furniture but is nevertheless intimately connected to them (perhaps composed of them). It is not possible to settle this dispute here. However, the fact that it is debatable whether or not “There is some furniture in here,” entails ontological commitment to a kind of stuff (furniture), shows that it is not obvious that mass-quantification necessarily entails ontological commitment to stuffs, and hence that the way we speak about processes isn’t enough to establish that processes are metaphysically similar to stuffs.

8 As in, for example, Lewis’s (1983, 1986) class nominalism.

9 As in, for example, Armstrong’s (1978a, 1978b, 1989) realism.

10 As in, for example, the nominalism of Quine (1948), Devitt (1980) and the resemblance nominalism of Rodríguez Pereyra (2002).

11 One could argue that there are relevant linguistic differences between process predications and mass-nouns like ‘furniture’ which might explain why is it valid to conclude from the grammatical features of nominalised process predications that processes are stuff-life even though a structurally similar inference in the case of furniture is not valid. For example, ‘There is some furniture in here’ is synonymous with ‘There are some bits of furniture in here’, whereas ‘There was running by Roger’ does not mean ‘There were some bits of running by Roger’ just as ‘There is snow on the roof’ does not mean ‘There were some bits of snow on the roof’. However, even if there is such a relevant linguistic difference between mass-nouns like ‘furniture’ and process predications, it is unclear what the relevant linguistic difference between mass-nouns like ‘happiness’ and process predications is. So, just as happiness does not quantify over an entity which is stuff-like, even though it is a mass-noun, it remains possible that nominalised process predications do not quantify over entities which are stuff like.

12 See, for example, Gill (1993).

13 See Davidson (2001, 3–21, 43–63), Bishop (1989, 40–44) and Smith (2012).

14 See Alvarez and Hyman (1998, 222), Davidson (2001, 52) and Hornsby (1980, 101).

15 One way Steward might avoid this objection is by adopting the same strategy Alvarez and Hyman adopt in their (1998) paper. Alvarez and Hyman argue that an action is a causing of an event by an agent, but they deny that the event caused is the agent’s action (224). Steward could similarly deny that the individual process an agent sustains is the agent’s action and instead hold that the agent’s action is their sustaining of this individual process. The trouble with this suggestion is that it is unclear, on this view, what sort of entity the sustaining of an individual process would be. Alvarez and Hyman are clear that the causing of an event is not itself an event. Indeed, one reason there has been so much interest in the nature of processes recently is that it offers some promise of explaining what Alvarez and Hyman’s ‘causings of events’ are. This puzzle would be re-opened if Steward’s view is modified in the way suggested. Furthermore, the resulting metaphysics would be cumbersome including events, individual processes and causings/sustainings of individual processes.

16 In fact, I do not think there are any substances which are not, at some time, engaged in at least one process, hence I do not think there are any substances which have not, at any time, supported any processes or events. But I take this to be a contingent truth. Worlds where substances exist without engaging in any processes at all seem metaphysically possible.

17 Denying that engagement and instancing are elements of being allows me to avoid an objection akin to Bradley’s regress (1893), namely: does a substance S need to engage in engagement in order to engage in a process P, and if it needs to engage in engagement, does it need to engage in engagement* to engage in engagement, and so on? The answer is: no, a substance S does not need to engage in engagement in order to engage in a process P, because engagement is not a process (or an entity of any kind). A similar answer is offered to the question of whether an event needs to be an instance of instancing in order to be an instance of a process. I intend this sort of response to be parallel to responses to Bradley’s regress offered by Armstrong (1989, 109–110) and Lowe (2005, 34–49).

18 The fact that engagement, like Armstrong’s instantiation, is an inter-category relation that holds between particulars and universals means that a temporal stuff theorist would not be able to claim that agents are connected to temporal stuff by engagement unless they also believed that temporal stuffs were universals. In that case it is doubtful that temporal stuffs could be entities which pervade intervals of time and compose events, as universals do not seem to be the right sort of entities to play these roles.