CHALLENGING THE PROCESS VIEW OF ACTION

Abstract

There is an ongoing debate in the ontology of action about whether actions are processes, events, relations, or sui generis entities. This paper focuses on the process view, the view that actions are processes. I challenge it in two ways. First, I argue that some actions are not processes because their performance need not be associated with or accompanied by a process. Second, I critically discuss three main arguments that have been advanced to support the process view. My view, the pluralist view, is that process-theorists are right to hold that certain actions are processes, but wrong to maintain that every action is a process: activities are processes, acts are not, and some acts are composed of a process whereas others are not.

Keywords:
Action; Process; Event; Act; Activity

Section 1 - Introduction

On 18 July 2019 Anne Lorimor aged 89 reached Uhuru Peak and became the oldest person to climb Mount Kilimanjaro to its highest summit. Here is a view about what her climbing Mount Kilimanjaro to its highest summit amounts to: there is an act (the climbing of Mount Kilimanjaro to its highest summit), it is made up of a result (her movement to the summit), a boundary or end-point (her being at Uhuru Peak at 3:14 p.m. local time), a process (climbing), and an agent (Anne Lorimor). All four components appear to be essential to her action of climbing the Kilimanjaro to its highest summit. There is no climbing to the summit without a climber. There is no climbing to the summit without a movement to the summit, i.e. without a climber’s occupying a series of continuous locations between where she started her ascent and the summit. There is no climbing to the summit if the climber does not reach it. Finally, there is no climbing to the summit if there is no process of climbing, e.g. if Anne Lorimor comfortably reaches her destination in a helicopter.

This view seems rather simple and natural. Yet, it conflicts with the process view in the ontology of action, the view that actions are processes, as opposed to events, where events are understood as completed occurrents, occurrents whose spatiotemporal properties are fully determinate, and processes are understood as ongoing or dynamic occurrents.¹

On this view, Lorimor’s action is not an event ‘made up’ of a process, but is a process.

Proponents of the process view have justified their view by the explanatory work processes do in action theory. Processes, they argue, explain (i) the dynamic or evolving character of extended actions (actions that take some time to accomplish) (Steward 2012, 2013, 2015; Stout 2018, 2010); (ii) the possibility of acting intentionally (Charles 2018; Thomson 2014; Stout 2018); and (iii) the nature of the causation of action (White 2020; Stout 2010; Hornsby 2012, 2014).

Yet, in this paper, I will try to show how it is still best to think of actions that involve an end-point like Lorimor’s as events as opposed to processes. The arguments I will discuss focus on (i) and (ii). By examining them, I hope to at least make plausible a rival view: the pluralist view. Are all actions processes? The pluralist’s answer is that activities are processes but acts like Lorimor’s are events. I will use the term ‘action’ to refer to both, which means that actions are sometimes processes and sometimes events, which in turn might or might not have processes as constituents.² If that is right, proponents of the process view like Helen Steward and Rowland Stout have been right to give a place to processes in the theory of action, but they have gone too far.

My main task then is twofold. First, I will argue in section 3 that the process view overgeneralises and that we can see how by considering the case of actions which are achievements. Secondly, I respond to three arguments which have been advanced in favour of the process view: the argument from change in section 4 (Steward 2012, 2013, 2015; Stout 1997, 2016), the incompletion argument in section 5 (Wolfson 2012), and the non-observational knowledge argument in section 6 (Charles 2018; Hinshelwood 2022; Stout 2018). I end by outlining some of the advantages of the pluralist view over the process view in section 7. But before, I explain how proponents of the process view understand processes in more detail.

Section 2 - Processes, Properties and Times

The process view is a view about the kind of ontological category to which actions belong. It is opposed to the pluralist view by claiming that actions belong to a single category: that of processes. It is also opposed to another monist view, the event view, according to which actions belong to a single category: that of events. Both the pluralist view and the process view presuppose that there is a distinction between events and processes. And there are ‘as many view of what demarcates processes from events as there are of researchers who have written about it’ (Galton 2018, 41). But the process view is not an account of the event/process distinction. Some of those who have defended an account of that distinction may be pluralists, such as Hornsby who contrasts activities and actions (Hornsby 2012, 2013).

The process view and the pluralist view are compatible with different accounts of what processes are. I will not attempt to give and defend an account of this distinction in this paper. Instead, I will focus on some central features that processes are thought to possess according to those, like Steward or Charles, who support the process view, such as the claim that processes do not primarily possess their properties in relation to time in the same way as events. Let us now turn to the way in which some proponents of the process view understand processes.

Most philosophers who have theorised about processes do not use the term ‘event’ to refer to occurrents in general, but rather to occurrents that exist only once they are completed or finished,³ that is occurrents that have a fixed duration or fully determinate spatiotemporal properties (Hinshelwood 2022, 248; Galton 2008; Galton and Mizoguchi 2009; Haase 2022).⁴ I say ‘only’ because even if we think that processes exist once they are completed and have fully determinate spatiotemporal properties, they already exist when they are incomplete and while they are occurring when their spatiotemporal properties are not fully determinate.⁵

As there are different views of the event/process distinction, there might be different versions of the process view. Some who have theorised about processes think of them as temporal stuff in analogy with the way we think of the stuff that constitutes material objects (Crowther 2011, Hornsby 2012; Mourelatos 1978). Others think of processes as ‘occurrent-continuants’, that is as things that are wholly present at each time at which they are occurring and that progress by enduring and not by acquiring temporal parts (Stout 1997, 2016; Charles 2018). And yet others think of processes as occurrents that have temporal parts but that possess their intrinsic properties temporally-as opposed to events (Steward 2013, 2015). On the latter view, processes persist by having further parts but the addition of temporal parts does not create a new event. It contributes to the growth of the same token process (Steward 2012, 383). But even if the proponents of the process view do not share the same conceptions of processes, they all share the claim that actions belong to a single ontological category: processes.

I wish to set aside the temporal stuff view of processes here, not only because I am ready to grant that there are processes as stuff like walking, swimming or writing, but also because some proponents of the process view admit that some processes are temporal stuff and some are countable individuals to account for the variety of actions (Steward 2012, 382). Some actions, like swimming or talking, progress through time in an indefinite way, are open-ended and have no natural climax or boundaries. This is the category of activities, often contrasted with that of acts. They are naturally associated with processes as temporal stuff. Acts, like writing a letter, swimming the Hellespont or reaching the summit, are performances that are countable, have natural boundaries and are close-ended. Acts are countable individuals, and if they are processes, as the process view claims, there must be countable individual processes. Helen Steward calls elements of this latter category ‘individual processes’ which are occurrents that unfold or progress in the world that are structured ‘in such a way that a certain termination point, product, or ongoing production cycle is the norm’ (2013, 807).⁶ Individual processes progress towards their termination or completion point, at which time they are ‘completed’; whereas the idea of completion does not apply to process-stuff. Individual processes can be completed or incomplete, but a completed process is not an event because events, remember, only exist once they are finished whereas a completed process is already existing before it reaches completion.

Both proponents of the ‘occurrents-continuants’ view and Steward agree on one important claim, namely that processes do not possess their (intrinsic) properties atemporally like events do. But they disagree about whether processes possess their properties temporally like continuants or whether they have a distinctive way of bearing their properties. Let us dwell a bit on these two ideas. What does it mean that events possess their properties atemporally, that is not at times? Take Columbus’ first journey which started on 3 August 1492 and ended on 4 March 1493. Suppose his journey was tumultuous between September and October 1492, but calm between January and February 1493. When we say that the journey had the property of being a tumultuous journey between September and October 1492, we are not, Steward remarks, tying the journey event to a property it had at some time in the past (2015, 116). The event has always (atemporally) the property of being a tumultuous journey between September and October 1492, it had it in August 1493, it has it in 2024 and it will have it in 2025.

This contrasts with the way in which continuants have properties. A continuant, like my brother, had the property of measuring 1.40 meters in 2001. By saying this we locate the time at which my brother had this property, he had it at this time in 2001, not in 2024. Of course, because of this, we can also ascribe atemporally to him the property of measuring-1.40-meters-in-2001. But this atemporal ascription is derivative as it depends on his having had the properties of measuring 1.40 meters at a certain time. According to proponents of the process view then, continuants primarily possess their properties at times, whereas events primarily possess their properties atemporally. I say primarily in the second case as well because Columbus’s journey possesses the property of being in the past at a certain time (e.g. now) and not at others (e.g. in 1491) (see Steward 2015).

Moreover, proponents of the process view argue that events are static, that is they cannot change by gaining and losing properties.⁷ While the height of my brother changed throughout the first decade of the current millennium because he grew, the whole journey of Columbus did not change throughout 1492-1493. The sense in which the journey-event changed is that a part of it was tumultuous (between September and October 1492) and another part of it was calm (between January and February 1493). But this is mere replacement of parts with certain properties by parts with different properties. As the journey-event did not yet exist in October 1492, it is strictly speaking only a part of it that was tumultuous; similarly for the part that was calm. But then, if all that happened is mere succession of parts with different properties, we cannot truly say that the event itself, the journey, changed, as Steward puts it, ‘any more than an apple change which is redder on one side than on the other’ (2015, 113).⁸

How do processes bear their properties in relation to time? Stout and Charles think that processes do not bear their properties atemporally like events, because they think that processes are continuants. Steward agrees but for a different reason (1997; 2013; 2015). As processes are essentially ongoing occurrents, we can ask how they go on through time, e.g. smoothly, intermittently, steadily, constantly, persistently, sporadically, irregularly, perpetually, incessantly. But, she argues, such properties are neither possessed by processes at times nor atemporally, but between times. If the humming of my computer was persistent between eleven and noon it was not persistent at 11:20. ‘Persistent’ means continuing frequently for a long time without interruption. But if the humming of my computer was first persistent and then intermittent, then the humming does not have these properties primarily atemporally either. From there, Steward argues that to say of the humming of my computer that it was persistent at 11:20 is a way of saying that 11:20 was a moment which falls within a period of time over which the humming was persistent (2015, 121). But saying that it possessed this property at a time (11:20) is derivative and an idealization of the fact that processes primarily possess their intrinsic properties (the kind of properties like being intermittent) between times.

Importantly, that processes have properties between times presupposes that they have temporal extension and temporal parts. But change in ‘between time’ properties is not mere succession of parts with different properties. Processes are then dynamic in a sense in which events are not. And the fact that processes can change in this way suggests that their temporal parts are not essential to their identity. As Steward puts it:

“It only makes them into entities which are potentially rather robust with respect to their temporal parts-they can be conceived of as the very same tokens even shorn of many of their actual constituent temporal parts, and hence as existing in possible worlds where they are interrupted and do not run to completion” (2013, 807).

This means that a token process could be shortened or extended while remaining the same. This also means that processes can be seen as things that can become more or less of something, e.g. they can become more or less aggressive, fast, vigorous. On Steward’s view, this sets processes apart from events. As we now have a better grasp of what processes are according to some proponents of the process view, we can now consider my arguments against the latter.

Section 3 - Acts and Processes

There are (at least) two reasons to doubt the claim that all actions are processes. The first is the following. It is, as we saw, in the nature of processes to go on in time. But going on in time is not in the nature of certain actions. Hence, some actions are not processes.

I won’t discuss the first premise, which my adversary accepts. The second premise can be supported by noticing that to go on in time plausibly presupposes that the occurrent is going on in or over a time period or interval. But the occurrence of certain actions does not presuppose such a time period or interval over which they would go on. This suggests that they do not go on in time. And since it is in the nature of processes to go on in time, these actions are not processes.

The second reason is that there are actions such that it is not even in their nature to be composed of something that goes on in time, a process or an activity. This makes it doubtful that these actions could be identified with processes. Let me explain these two points in turn, starting with distinctions between different kinds of actions.

In the previous section, we distinguished two kinds of action: activities and acts (Vendler 1967; Kenny 1963; Simons 1987; Mourelatos 1978; von Wright 1963). Hyman says that in the case of activities but not of acts, if X is φ-ing at all times between t1 and t2, then X φs between t1 and t2 (2015, 34). Climbing is an activity, like swimming, walking or running. If Lorimor was climbing at all times between noon and one, then she climbed between noon and one. Climbing Mount Kilimanjaro to its highest summit, like swimming the Hellespont or running a 100-metre race, is an act. If Lorimor was climbing Mount Kilimanjaro to its highest summit between noon and one, it does not follow that she climbed to the highest summit between noon and one, because she may not have reached the summit until five. If we now focus on acts, some acts can be described in the progressive by using continuous tenses and some cannot. Those that can are Vendler’s accomplishments: actions that have a climax or bound and that go on for a time but take a definite amount of time (to be finished) like drawing a circle, running a mile or climbing to the highest summit (1967). As they have a climax, if X was φ-ing at t, it means that t is on the time stretch in which X φ-ed (that is in which the climax essential to this act was reached). This is why the notion of accomplishment implies the notion of unique and definite time periods. For instance, if Lorimor was climbing to the highest summit at noon, then noon is on the time stretch in which she climbed to the highest summit, e.g. 16 July at 9:00 to 18 July at 16:00.

Acts that cannot be described in the progressive using continuous tenses are Vendler’s achievements: actions that have a climax or bound and that may take time, but as we will see, that do not go on in time. For instance, the action of reaching Uhuru peak took Lorimor time, say 27 hours, but it is not the case that she was reaching Uhuru peak at any moment of these 27 hours. We can ask at what time she reached it, but not for how long she reached it. The notion of achievement involves definite time instants: if she reached Uhuru peak between three and four, it means that the time at which she reached it is between three and four.

Achievements, like finding one’s keys, voting, winning the race or reaching the top, are crucial for the argument because unlike accomplishments, they do not go on in or over time, which is a distinctive characteristic of processes. Vendler (1967) and Mourelatos (1978) think that achievements are punctual in the sense that they have no parts: they are atomic events. If this is true, achievements only presuppose a single time instant. But going on in time plausibly presupposes that the occurrent goes on in or over a period of time or a time interval. This presupposes more than a single time instant. But then, these actions are not processes. What is more, if going on in time presupposes that the thing that goes on has parts (or more than one), then it is also clear that these actions are not processes.

But even if they are not atomic events and have parts, it does not follow that achievements go on in time.⁹ For an achievement might have parts at t1 and t3, but it does not follow that it occurs throughout t1 and t3 or that it goes on between or at these times. For instance, my act of finding my pair of socks may have two parts: the first at t1 when I found my left sock, and the second at t3 when I found my right sock. But it does not follow from this that my achievement of finding this pair of socks was going on between its two parts or even that it was going on while I found the left sock or while I found the right sock. This holds whether or not these are, in turn, atomic events. This is also what is suggested by the fact that we cannot ask for how long I found my pair of socks.

What is more, we can see that the features that Steward associates with processes are not possessed by achievements. Because achievements like winning the race do not go on in time or progress over time, it makes no sense to ask how the winning of the race went on throughout some period of time. And thus, the temporal modifiers typically applicable to processes do not apply to the act of winning the race, e.g. persistent, continuous, incessant, perpetual, unremitting. And such an act cannot become more or less of something. One can win the race with more or less difficulty, but the act of winning the race cannot become more or less difficult.

It might be replied that things like winning the race, voting, or reaching the summit are only the culmination point of actions as processes, e.g. winning the race is the culmination point of racing, and reaching the summit the one of a climbing process. But this response concedes too much if it does not deny that these culmination points are acts-for it would still mean that some actions are not processes. And my view is compatible with the claim that these achievements are the culmination points of processes. The question is whether proponents of the process view can plausibly deny that achievements are actions?

I take it that the burden of proof falls on the shoulder of those who want to deny that achievements are actions. Perhaps a prima facie reason to think that they are actions is that achievements like reaching the top or winning the race can feature in answers to questions like ‘what deeds did she accomplish in her life?’.

Let us now move to the second reason to doubt the claim that all actions are processes: there are actions such that it is not in their nature to be composed of something that goes on in time, a process or an activity. Here again, a distinction between two different kinds of acts is crucial.

Some actions have as an essential component a corresponding process or activity. Climbing to the highest summit is a case at hand, since it is essential to that act that it is composed of a climbing process. Walking to the store is another. These acts are essentially composed of such a process or activity, which is why they cannot be performed by an agent unless she engages in the type of process specified in the nature of this act.

But this is not true of all acts. The occurrence of certain acts does not presuppose that their agent is or was engaged in a corresponding type of process which is one of their essential component. For instance, most of the time, an act of choosing occurs after a deliberation process (which is itself an action), but not always. Some acts of choosing might be spontaneous-even if their occurrence presupposes that some brain activity or neurological process was taking place. But the point is not confined to the case of acts of choosing. Consenting, promising, or voting are not essentially composed of a verbal activity or a written activity (a process). For, one may vote ‘no’ in a meeting without the need to engage in any process in virtue of a convention that says that those who do not raise their hand vote ‘yes’.

It may be objected that this is an example of an omission or a refrainment, and thus not of an action that is not composed of a process. True, there is indeed an omission in this case (not raising one’s hand). But the action of voting is not identical with that omission. We can act by omitting, as when we kill a plant by not watering it. But it does not follow from this that the act we perform by omitting is an omission. Similarly, if I vote by not raising my hand, it does not follow that my voting is an omission or is identical to my omission of not raising my hand.¹⁰

Now, it might be the case that, in general, a condition for the performance of acts is that some process or activity was underway for some time--whether that be an activity of their agent or of something else, like their brain. For instance, it might be that to vote (in the context of the above example), it must have been the case that, at some point, I was engaged in some political or deliberation process. But the fact that the performance of acts generally depends on engagement in processes or activities does not entail that all acts have as their component some process (even if some acts do). And thus, it does not entail that they are essentially composed of some process. This, I think, should make one doubt that we can always identify acts that are achievements with a process.

In addition, we may also reverse the question and ask whether some action-processes presuppose an act, and whether engaging in activities depends generally on the performance of acts. Moving is an activity, a process, but to start moving does not seem to be one, although to start or to stop a process seem to be acts as von Wright points out (1963, 41-42). One may reply that to start moving presupposes the action of, say, sending out nerve impulses. But if this is an action, it is an act, and if it is constituted by a process-perhaps ‘the travelling of the impulse down the nerves’-this process does not seem to be an activity or process of the agent (but rather of the agent’s nervous system). And thus, it seems we cannot reduce every act or achievement of an agent to an action of the same agent that is a process.

If this is right, not only some actions are not processes, but certain actions are not even made up of a process. And, even if proponents of the process view could meet the challenge posed by this argument, I would maintain that their arguments, which we examine in the next three sections, fail to establish their view.

Section 4 - The argument from change

The first argument for the process view is rather straightforward. As we saw in section 2, processes are dynamic (they can change) whereas events are static (they do not change) because the former do not possess their properties primarily atemporally whereas the latter do. This gives proponents of the process view a first premise:

• P1: Events are static.

I won’t be questioning this premise.

The next step is to argue that actions are dynamic and can change. Here is Steward’s argument. Suppose that we observe Jones butter some toast. Now consider the sentence ‘he is buttering the toast angrily’. If Jones buttering the toast is an event, then there is a buttering of the toast by Jones which is angry.¹¹ But as Steward points out, we are not predicating ‘angry’ of the whole buttering event, because then the claim would be false if the whole buttering turned out to be mainly a calm affair in the end (2012, 378). And the suggestion that we are predicating ‘angry’ of a subevent or a stretch of buttering won’t do either according to Steward. What would be the bound of that subevent? It cannot be the bound of the stretch during which we observed Jones, since what we observed here might well extend so as to encompass ‘the whole of a largely calm buttering’. But neither it can be an ‘artificially’ delimited event, because it wouldn’t be clear what would make this subevent the unique object of our thought, i.e. what we single out when we say of Jones’s buttering that it is angry.

Similarly, consider Jones’s pushing of the door and consider the sentence ‘Jones pushed increasingly firmly at the door’. It seems natural, Steward argues, to think that Jones’s pushing at the door increases steadily in firmness as time goes by. But this cannot be the case, strictly speaking, if his action is an event. Events don’t change: for no phases of Jones’s pushing can change in respect to its degree of firmness, except by consisting itself of phases which have differing degrees of firmness. What is more, it also seems plausible to think that Jones’s pushing of the door could have lasted longer than it actually did (2012, 379), and thus that its parts are not essential to its identity, unlike with events.¹² If Steward is right, then:

• P2: All actions are dynamic (and so not static).

From P1 and P2, we get:

• C: No action is an event.

From here, one reaches the conclusion that actions are processes by inference to the best explanation. If actions are processes, given that processes are dynamic, it explains why actions are dynamic. Some responses have been offered to this argument,¹³ but I will focus on showing why I think the above argument does not give more support to the view that actions are processes than it does to the pluralist view according to which only some actions are processes.

We can accept both the claim that processes are dynamic and change, and the claim that events are static without accepting P2. As we said in section 2, to account for the variety of actions, that is open-ended activities like wandering and closed-ended acts with a climax like killing, the process view needs two kinds of processes: processes as stuff and individual processes.

Assuming that this is right, let us go back to Lorimor’s climbing of Mount Kilimanjaro to its highest summit. Suppose that her action was laborious between t1 and t3 and then steady between t4 and t5. Suppose that we observe Lorimor at t2 and say that her action is laborious. Let us grant that we predicate ‘laborious’ of a process-which is what Steward’s example with ‘buttering the toast angrily’ is supposed to show. The question now is why should we assume that the process we are predicating ‘laborious’ of is the act of climbing Mount Kilimanjaro to its highest summit? This may be a natural suggestion. But another natural suggestion, in line with our initial description of her action at the start of this paper, is that the process that takes this predicate is the process or activity-stuff of climbing in which Lorimor is engaged. What we may observe become more laborious or steadier during her ascent is the activity, the stuff, not the countable act of climbing Mount Kilimanjaro to its highest summit.¹⁴ Similarly, if it is true that Jones pushed increasingly firmly at the door, it is because Jones’s pushing (the activity, the stuff) became increasingly firm, and it was directed at the door. But if Jones pushed the door closed, his act of pushing it closed did not itself increase in firmness, although it is made-up of an increasingly firm pushing activity (stuff). Why should we go, then, with the first suggestion?

Steward seems to assume that at t2 we would be predicating ‘laborious’ of the act of climbing Mount Kilimanjaro to its highest summit. But this already presupposes that this act is not an event. For, on her account, an event cannot change and there is yet no (whole) event of climbing Mount Kilimanjaro to its highest summit--this is what she says in the toast buttering case.¹⁵ But this begs the question against the other natural suggestion, according to which the process of which we are predicating ‘laborious’ is not the act of climbing Mount Kilimanjaro to its highest summit but the activity of climbing. For, on this account, the act of climbing Mount Kilimanjaro to its highest summit is an event.

Nor, of course, can we simply assume that this act is an event and not a process, since it would beg the question against Steward’s view. But this only helps to show that the argument from change is undecisive. What it establishes, if it establishes anything, is that we are predicating ‘laborious’ of a process, not that we are predicating laborious of the act of climbing Mount Kilimanjaro. The crux of the matter is whether the view that all actions are processes is the only one that can account for the fact that we are predicating ‘laborious’ of a process. And the answer here must be no, since a pluralist view can account for that too. It remains to be shown why the view that all actions are processes would give a better account of this datum than the pluralist view.

The upshot is that if we admit processes-as-stuff in our ontology, we may explain the apparent dynamicity of certain actions (as events) by the actual dynamicity of the process- stuff of which these actions are constituted. The first natural suggestion that the action that is changing over time is Lorimor’s act of climbing Mount Kilimanjaro to its highest summit is thus explained away by the fact that it is made up of a process (the climbing activity) which is itself changing. Likewise for any other act such as my run in the park this morning, which may be composed of my running activity and which was increasingly laborious. If this is right, we can reject P2 and accept that only some actions (namely activities) are dynamic.

One may object that the process of climbing is the same thing as the act of climbing Mount Kilimanjaro to its highest summit. This is is not so. The former may come to constitute more than one action (e.g. the action of climbing to the third highest summit and the second highest summit), but not the latter; and the former may have existed without the later, if e.g. Lorimor had stopped before reaching Uhuru peak.

It might also be objected that changes in the process (the climbing activity) constitutive of Lorimor’s broader action is not the best explanation of why we want to say that the latter action is changing over time. That is because it posits an activity (a process) and an event (the act), whereas Steward’s ontology is more parsimonious. It posits only a process, albeit an individual process which reached completion at Uhuru peak. But is her ontology more parsimonious? She accepts events (as I do), individual processes (which I reject) and she has to make room for open-ended processes (like I do), without which we could not make sense of activities like wandering in the woods.

To sum up, the best explanation of why some actions, namely activities, are dynamic may indeed be that they are processes. But this does not force on us the conclusion that every action is a process.

Section 5 - The incompletion argument

On the pluralist view presented at the outset of this paper, Lorimor’s action of climbing Mount Kilimanjaro to its highest summit, which is an event, is partly constituted by a terminus point or boundary, i.e. her being at Uhuru peak. The individuation of such an action parallels the individuation of changes, that are events, because changes are typically of a kind if their terminus point is of a corresponding kind. If Lorimor never reached the summit, then the action she was performing while climbing is not of the same kind than the one whose terminus point is her being at Uhuru peak, no matter how we describe it.

However, proponents of the process view deny that Lorimor’s action is not a process but an event because, they argue, its actual terminus point is not essential to her action. They do so for two reasons- I discuss the second, which concerns practical knowledge, in §6.

The first reason, given by Wolfson (2012, 327), is that it is not true that if the agent did not φ, then he was not φ-ing. For instance, if Lorimor was climbing Mount Kilimanjaro, that does not imply that she did climb Mount Kilimanjaro, e.g. she might have been crushed by a rock on her way up. This is a key feature of the progressive known in linguistics as the imperfective paradox. The linguistic datum here is that there is a lack of entailment from some sentences with progressive aspect to their perfective counterparts (Szabó 2008, 511). The objection is that the pluralist view fails to recognize this, and as consequence, it entails a radical revision of ordinary practices regarding how we describe actions in progress. For instance, Wolfson surmises that nobody would be inclined to say that if the fire department puts out a fire, this is a sign that prior to the intervention the house was not burning down. He adds that we could not replace ‘x is φ-ing’ with ‘x will φ if not stopped’ because what grounds the assessment that x will φ if not stopped is that x is φ-ing. Hence, he concludes, if we were to regulate our speech so as to deny that x was not φ-ing if x did not φ, we would have to invent another phrase to talk about what was happening before we knew whether what x was doing was φ-ing or something else (2012, 327).

In response, I shall argue that the pluralist view does not entail a radical revision of our ordinary practice of describing actions (with the progressive). Once we pay attention to the fact that certain verb phrases may describe different actions, we can readily acknowledge the lack of entailment from some progressive sentences to their perfective counterparts. To explain this, I need to make two remarks.

First, as I said, we need to pay attention to the fact that ‘climbing Mount Kilimanjaro’ may describe different actions, one of which may be the act of climbing-Mount- Kilimanjaro-to-its-summit, which I take not to be a process. Other descriptions may describe activities. In a sentence like ‘Lorimor was climbing Mount Kilimanjaro’, the verb phrase may describe the activity she was engaged in, climbing, as taking place on Mount Kilimanjaro (not on Mount Everest). And even if we specify the description, such as ‘climbing Mount Kilimanjaro to its summit’, we may still describe different actions which are activities. For instance, we may describe the activity she was engaged in until she reached the summit-she was on her way to achieve what she later achieved. We may also describe her activity of climbing as being directed towards the summit. Or, we may describe her activity of climbing as being engaged in with the intention to make it to the summit. Activities do not essentially have an endpoint, and the claim my view makes is not about them, but about an act.

Secondly, we need to distinguish a linguistic claim from an ontological claim. My view makes an ontological not a linguistic claim. It says that the act of climbing-Mount- Kilimanjaro-to-its-summit is not a process, but an event, and that this act has an essential endpoint, namely being at Uhuru peak or reaching it. As we just saw, there might be many other actions in the vicinity for which this point does not apply. But the view does not say that the sentence ‘she was climbing Mount Kilimanjaro to its summit’ is not true if the sentence ‘she climbed Mount Kilimanjaro to its summit' is false, which is what the imperfective paradox is about. I do not think that the truth of the first sentence entails the truth of the second one. The question is how should we account for that fact? This is a controversial question and I do not have enough room to do it justice here. But here is what my view entails and it is not the problematic revisionary consequence pointed out by Wolfson.

The view, as we saw, says that ‘climbing Mount Kilimanjaro to its summit’ may describe different actions and that the act of climbing-Mount-Kilimanjaro-to-its-summit is an event that has an essential endpoint. If the perfective sentence ‘Lorimor climbed-Mount-Kilimanjaro-to-its- summit’ designates this act, then its truth entails the existence of a reaching of the summit by Lorimor. Now, consider the progressive sentence ‘Lorimor was climbing- Mount-Kilimanjaro-to-its-summit’. If this sentence does not designate an event, but an activity or a process, then on my view, it simply does not designate the same action as the one designated by its perfective counterpart.¹⁶ In that case, the progressive sentence may be true and designate an activity, while the perfective sentence is false. For the truth-maker of the first sentence can be an activity that exists when the act of climbing-Mount-Kilimanjaro-to-its summit does not (and hence cannot serve as a truth-maker for the second perfective sentence). The entailment from the progressive sentence to the perfective one does not necessarily go through. And this is in line with the imperfective paradox.

Importantly, the view does not have the revisionary consequence that we need to invent a phrase to talk about what Lorimor was doing before we knew whether she would reach the summit. We can and would often give true descriptions of what she was doing as ‘she was climbing Mount Kilimanjaro to its summit’. There are simply several actions in the vicinity that we could refer to, even if we do not ordinarily notice that this is the case, perhaps because of the conversational context.

Let me now go on the offensive against Wolfson’s argument and challenge his claim that we could not replace ‘x is φ-ing’ by ‘x will φ if not stopped’. The reason he gives why we could not do so is that what grounds the assessment that x will φ if not stopped is that x is φ-ing. I disagree. Suppose that we don’t know yet if Lorimor will make it to the top. What would ground our assessment that she will climb Mount Kilimanjaro to its highest summit if not stopped? It would not be the fact that she is climbing it to its highest summit. For we do not know yet if this is what she is doing-that is, if we take ‘climbing to its highest summit’ to refer to the act that has being at the summit as an essential endpoint. Rather, it is the fact that Lorimor is engaged in the activity (climbing) that is characteristic of and will constitute the action she will perform if she succeeds (climbing- Mount-Kilimanjaro-to-its-highest-summit), along perhaps with our knowledge of the circumstances, e.g. that we have no reason to believe that she will be prevented from reaching the top; that she is moving in the direction of the top; that she has the capacities she has; that she intends to reach the top, etc. If so, it seems that the sentence ‘Lorimor is climbing Mount Kilimanjaro to its highest summit’ can, after all, be replaced by ‘Lorimor will climb Mount Kilimanjaro if not stopped’. Here too, there is no need to ‘invent’ another phrase to describe what Lorimor was doing before she eventually reached the summit to describe what she was doing then.

Consequently, the view that an actual endpoint is essential to some actions (like Lorimor’s climbing-Mount- Kilimanjaro-to-its-highest-summit) is not refuted by Wolfson considerations. And such actions, I contend, are events. Let us move to the next argument.

Section 6 - The argument from non-observational knowledge

The next argument in favour of the process view relies on a now widely accepted claim in the philosophy of action, namely that to act intentionally, one needs to know what one is doing.¹⁷ The intuition behind the view is this. Suppose you are asked ‘Why are you φ-ing?’. If you answer ‘I did not know I was φ-ing’, then the φ-ing would not count as an intentional action (at least not under a certain description). Following Anscombe (2000), it has become common to treat the relevant form of knowledge as non-observational knowledge. We may set aside the well-known difficulties with spelling out what this distinctive type of knowledge involves. Roughly, to know without observation what I am doing in acting intentionally is to know this spontaneously and in a way that does not turn on my observational evidence (Setiya 2007, 24).

The following principle embodies the general idea:

Non-observational knowledge (NOK): An agent is doing an action φ of a kind K intentionally if she can know without observation that φ is of kind K.¹⁸

If this is right and we (normally) perform actions intentionally, then we normally know without observation that we are doing an action φ of a kind K while we are doing it.

Now, trouble arises if actions are events because for an action φ to be of a kind K, it must have a terminus point or result of a corresponding kind G. And, if we normally know without observation that the action we are doing is of a certain kind, then it means that we normally know without observation that the action we are engaged in has an actual terminus point which is of a corresponding kind. Clearly, however, while we are doing an action, we normally do not yet know its terminus point because it is located, as it were, in our future. If so, our intentional actions cannot be events. To illustrate, if Lorimor is climbing Mount Kilimanjaro now and the terminus point of this action is the state of being at Uhuru peak, then she does not yet know while she is climbing that she will stand at Uhuru peak in the future.¹⁹ She might be crushed by a boulder during her ascent.²⁰ So, if actions are events, Lorimor can’t know that she is climbing Mount Kilimanjaro, which seems absurd.

The best explanation, then, of the fact that we normally know without observation that we are doing an action of a certain type while we are doing it is that actions are processes. That is because processes are either open-ended if they have no terminus point, or can remain incomplete if they have a terminus point such that reaching it is not essential for it to be of the relevant kind. This view implies that, for instance, whether or not Lorimor reached the summit, her action would be of the same type; it would simply be an incomplete instance of the same action-type.²¹

The argument is supposed to give us a reason to reject the claim that actions are events and thus occurrents (at least partly) individuated by an actual terminus point or boundary on the ground that we would not intentionally perform most of the actions we think we perform intentionally. I think, however, that NOK should be toned down, and that once this is done, we can see that it is compatible with the claim that some actions are events.

Why should we assume that what Lorimor knows while she is climbing is that her action falls under the kind ‘Climbing Mount Kilimanjaro to its highest summit’? Anticipating on this question, Charles argues that what the agent knows while acting intentionally is not merely what she has done so far together with what she intends to do, e.g. she knows how far she has climbed now; what distance remains; that she intends to reach the summit, etc. For, he says, she could know these even when she has stopped climbing Mount Kilimanjaro, provided that she retains the intention to complete the task in the distant future (2018, 36). There’s something more that she knows while she is progressing, namely that she is climbing Mount Kilimanjaro.

However, his response fails to prove the point because it does not show that what Lorimor knows while she is climbing is that her action is of the kind ‘climbing Mount Kilimanjaro to its highest summit’. Charles might reply that what shows this is the fact that she would be able to answer ‘I am climbing Mount Kilimanjaro to its highest summit’ if while climbing, she was asked ‘What are you doing?’. But it could be that all that is meant by ‘I am climbing Mount Kilimanjaro to its highest summit’ is that she is climbing on the surface of Mount Kilimanjaro and that she intends to reach the summit. Similarly, if while I’m breaking eggs, someone asks me ‘What are you doing?’ and I reply ‘I’m baking a cake’, it does not show that what I know is that the action I am now performing is of the kind ‘baking a cake’. All this shows is that I am doing something in order to perform an action that falls under the kind ‘baking a cake’, i.e. that I intend to token this kind of action. The argument is, so far, inconclusive.

If something like NOK is right, then, what we need to know is not so much which actions we are engaged in, but which activities, which constitute our acts, we are engaged in.²² I have no quarrel with the idea that one may know non-observationally what one is doing when we are talking of an activity, an open-ended process. This is because, as soon as the agent has engaged in it, she has succeeded. No endpoint needs to be reached for it to be true that the agent has performed this action. What is more, there is no mystery about the fact that one may know non-observationally what one intends (Paul 2009). If we restrict what we need to be able to know without observation in order to act intentionally to activities and our mental states, the pluralist view is compatible with a non-observational knowledge condition on intentional action.²³

In short, considerations about the non-observational knowledge condition on intentional action do not force us to reject the view that some actions are not processes. On the pluralist view, the actions we normally know we are doing without observation are only activities or actions that are such that as soon as we engage in them, we succeed in doing them, like climbing or walking. But this is not true of acts.

Section 7 - Conclusion

In this paper, I have discussed the view that all actions are processes with an eye on an alternative pluralist view according to which only some actions are processes (activities), while other actions are events (acts). On the pluralist view, acts, like events, are completed occurrents, whereas activities, like processes, are ongoing or dynamic occurrents. I provided some reasons in section 3 to cast doubt on the idea that some actions (achievements) are processes. I argued that what makes it even more doubtful that these actions are processes is the fact that they do not have processes as essential parts. I then examined three arguments in favour of the process view and found them inconclusive.

To finish, let me highlight some of the advantages of the pluralist view over the process view. The pluralist view incorporates insights of two monist views, the process view and the event view, while avoiding some of their problems. Since it holds that some actions (acts) are events, it can readily acknowledge that achievements can be actions. As a result, the pluralist need not try to reduce these acts to processes-a move, which I argued, should be rejected. Moreover, since the pluralist accepts that some actions are processes (activities), she can therefore respect the intuition that actions can change, be dynamic and ongoing without overgeneralising. This fits in well with one of von Wright's key insights, namely that in general, not only do acts depend on activities for their performance, but activities depend on acts to be engaged in: to reach the shop, you have to walk, but to walk, you have to take a step.

Another advantage of the pluralist view is that, contrary to the process view, it is compatible with an account of the event/process distinction in terms of ‘temporal stuff ’ à la Hornsby or Crowther. On the pluralist view, activities have a stuff character, not acts. But the process view must find the distinction elsewhere because in order to account for the variety of actions (activities, accomplishments, achievements), it has to posit processes that do not have a stuff or massy character.

Finally, by insisting that acts, like events, essentially have an endpoint, the pluralist view fits in well with a certain appealing conception of acts present in von Wright (1963). On this conception, an act is the bringing about of a change, and an act is of a certain type, in part, because its result, the intrinsic change of which the act is the causing, is of a certain type.

In the light of all this, I hope that by challenging the process view, I have at least made plausible the alternative I favour: the pluralist view, according to which actions can be events (acts) or processes (activities). More needs to be said, but we can already see that it combines simplicity with some insights of monist views.

Acknowledgments

I owe a special thanks to Vincent Grandjean and Antoine Vuille for their detailed and thoughtful comments on a previous version of this paper, as well as to Simon-Pierre Chevarie-Cossette and Patricia Sánchez. I thank Lilian O’Brien for the discussion of the topic of the paper, which helped me clarify my thoughts. I am also grateful to the two anonymous referees for this journal who provided invaluable feedback.

References

• Alvarez, Maria (2001). “Letting Happen, Omissions and Causation”, in Grazer Philosophische Studien, 61, 63-81.
• Anscombe, G.E.M (2000). Intention, 2nd edition. Cambridge MA: Harvard University Press.
• Baratella, Riccardo (2020). “Are There Occurrent Continuants? A Reply to Stout’s “The Category of Occurrent Continuants””, in Dialectica, 74(3).
• Benovsky, Jiri (2007). “On Presentist Perdurantism”, in Sats - Nordic Journal of Philosophy, 8(2), 79-88.
• Clarke, Randolph (2014). Omissions: Agency, Metaphysics, and Responsibility. Oxford: Oxford University Press.
• Crowther, Thomas (2018). “Processes as Continuants and Processes as Stuff”, in Stout R. (ed.), Process, Action, Experience, Oxford: Oxford University Press , 58-81.
• Crowther, Thomas (2011). “The Matter of Events”, in The Review of Metaphysics, 65(1). 3-39.
• Charles, David (2018). “Processes, Activities, and Actions”, in Stout R. (ed.), Process, Action, Experience, Oxford: Oxford University Press , 20-40.
• Davidson, Donald (1970/2001). “Events as Particulars”, in Essays on Actions and Events, Oxford: Oxford University Press , 163-180.
• Davidson, Donald (1963/2001). “Actions, Reasons, and Causes”, in Essays on Actions and Events , Oxford: Oxford University Press , 3-19.
• Davidson, Donald (1967/2001). “The Logical Form of Action Sentences”, in Essays on Actions and Events , Oxford: Oxford University Press , 105-48.
• Dretske, Fred (1967). “Can Events Move?”, in Mind, 76(304), 479-492.
• Frost, Kim (2019). “A Metaphysics for practical knowledge”, in Canadian Journal of Philosophy, 49(3), 314-40.
• Galton, Antony (2018). “Processes as Patterns of Occurrence”, in Stout R. (ed.), Process, Action, Experience, Oxford: Oxford University Press , 41-57.
• Galton, Antony & Mizoguchi, Riichiro (2009). “The Water Falls but the Waterfall Does Not Fall: New Perspectives on Objects, Processes and events”, in Applied Ontology, 4(2), 71-10.
• Galton, Antony (2008). “Experience and History: Processes and their Relation to Events”, in Journal of Logic and Computation, 18(3), 323-40.
• Gill, Kathleen (1993). “On the Metaphysical Distinction between Processes and Events”, in Canadian Journal of Philosophy , 23(3), 365-84.
• Grandjean, Vincent (2022). The Asymmetric Nature of Time, Springer Nature: https://doi.org/10.1007/978-3- 031-09763-8
  » https://doi.org/10.1007/978-3- 031-09763-8
• Haase, Matthias (2022). “Agency, Events, and Processes”, in L. Ferrero (ed.), The Routledge Handbook of Philosophy of Agency, New York: Routledge, 47-58.
• Hacker, P.M.S (1982). “Events and Objects in Space and Time”, in Mind, 91(361), 1-19.
• Hinshelwood, Alec (2022). “How to Make Do With Events”, in European Journal of Philosophy, 30, 245-58.
• Hornsby, Jennifer (2014). “Actions in Their Circumstances”, in Ford A., Hornsby J. and Stoutland F. (eds.), Essays on Anscombe’s Intention, Cambridge MA: Harvard University Press , 103-27.
• Hornsby, Jennifer (2013). “Basic Activity”, in Proceedings of the Aristotelian Society Supplementary Volumes, 87, 1-18.
• Hornsby, Jennifer (2012). “Actions and Activity”, in Philosophical Issues, 22, 233-45.
• Hyman, John (2015). Action, Knowledge, and Will, Oxford: Oxford University Press .
• Kassel, Gilles (2019). “Processes endure, whereas events occur”, in S. Borgo et al. (eds.), Ontology Make Sense, IOS Press, 177-93.
• Kenny, Antony (1963). Action, Emotion and Will, London: Routledge & Kegan Paul.
• Lombard, Lawrence B. (1999), “On the Alleged Incompatibility of Presentism and Temporal Parts”, in Philosophia, 27, 1, 253-60.
• Mourelatos, Alexander (1978). “Events, States, and Processes”, in Linguistics and Philosophy, 2, 415-34.
• Oderberg, David (2004). “Temporal Parts and the Possibility of Change”, in Philosophy and Phenomenological Research, 69(3), 686-708.
• Paul, Sarah K (2009). “How We Know What We’re Doing”, in Philosophers’ Imprint, 9(11), 1-24.
• Parsons, Terence (1989). “The Progressive in English: Events, States and Processes”, in Linguistics and Philosophy , 12(2), 213-41.
• Setiya, Kieran (2007). Reasons without Rationalism, Princeton: Princeton University Press.
• Simons, Peter (1987). Parts: A Study in Ontology. Oxford: Oxford University Press .
• Steward, Helen (2015). “What is a Continuant?”, in Proceedings of the Aristotelian Society Supplementary Volume, 89, 109-23.
• Steward, Helen (2013). “Processes, Continuants, and Individuals”, in Mind, 122(487), 781-812.
• Steward, Helen (2012). “Actions as Processes”, in Philosophical Perspectives, 26, 372-88.
• Steward, Helen (1997). The Ontology of Mind: Events, Processes, and States, Oxford: Oxford University Press .
• Stout, Rowland (2018). Process, Action, and Experience, Oxford: Oxford University Press .
• Stout, Rowland (2016). “The Category of Occurrent Continuants”, Mind, 124(496), 41-62.
• Stout, Rowland (2010). “What Are You Causing in Acting?”, in J. H Aguilar & A. A. Buckareff (eds.), Causing Human Actions, London: Palgrave Macmillan, 101-113.
• Stout, Rowland (1997). ‘Processes’, in Philosophy, 72, 19-27.
• Szabó, Zoltán Gendler (2008). “Things in Progress”, in Philosophical Perspectives, 22, 499-525.
• Thomson, Michael (2014). “Anscombe’s Intention and Practical Knowledge”, in A. Ford, J. Hornsby & F. Stoutland (eds.), Essays on Anscombe’s Intention. London: Harvard UP, p…..
• Vendler, Zeno (1967). Linguistics in Philosophy, Ithaca and London: Cornell University Press.
• Von Wright, G.H (1963). Norm and Action. London: Routledge and Kegan Paul.
• White, Andrea (2020). “Processes and the Philosophy of Action”, in Philosophical Explorations, 23(2), 112-29.
• Wolsfon, Ben (2012). “Agential Knowledge, Action and Process”, in Theoria, 78, 326-57.

• 1
  I will simply assume that there is an event-process distinction and that it is an ontological one here. But it must be pointed out that this has been challenged, see Gill (1993), Parson (1989) and Haase (2022). Moreover, it has been pointed out to me in discussion by Vincent Grandjean that this way of characterizing the distinction only makes sense if an A-theory of time is true, since according to B-theories of time, any occurrent has determinate spatiotemporal properties. If so, an A-theory of time is to be presupposed in the discussion.
• 2
  I shall not presuppose that actions are necessarily intentional and I will not use the term ‘action’ to mean ‘things done intentionally’ or ‘intentional doings of things’.
• 3
  There is a derivative sense of ‘exist’ in which things composed of non-existent (or non-present) things exist that is found in the literature on perdurantism: an event exists at a time if a temporal part of it exists at that time (Lombard 1999; Grandjean 2022). But this is not the sense involved in this sentence. And, in fact, positing this sense of existence is controversial (Benovsky 2007, 85).
• 4
  Process theorist typically associate this conception of events with Davidson’s work (1963; 1967; 1970).
• 5
  Does a process become an event once its spatiotemporal properties are fully determinate? I am not sure what the authors discussed here would say. I am only committed to the claim that some events can be constituted by processes like my fall on the ground is constituted by the process of falling. Once my fall on the ground occurs, the process of falling may now have determinate spatiotemporal properties. If this means that the fall event is constituted by another event (which used to be a process of falling) then so much worse for the process view. I would be tempted to say that the falling process remains a process even when it comes to constitute an event because I think that further criteria distinguish occurrents that are processes from those that are events, like whether they are essentially bounded and whether they unfold homogeneously.
• 6
  See also Hornsby (2013, 6) who calls them ‘individual activities’.
• 7
  This claim must be qualified because events are dynamic in the sense in which they ‘move’ in time by becoming increasingly past. Thanks to Vincent Grandjean for pointing this out to me.
• 8
  This echoes the three-dimentionalist’s argument against four- dimensionalism according to which the latter cannot countenance real change in their theory, see Oderberg (2004). See Baratella (2020) for a discussion of this point in the context of the distinction between events and processes. For discussions regarding the difference between events and processes in relation to change, see Simons (1987, 135), Dretske (1967), Hacker (1982), Galton and Mizoguchi (2009), Galton (2008), Kassel (2019).
• 9
  Steward is suspicious of the idea of occurrents that have no parts because she takes having of temporal parts to be the distinguishing feature of occurrents, see her (2015).
• 10
  For a discussion and defence of this view, see Clarke (2014) and Alvarez (2001). I thank a reviewer for pressing me on this point.
• 11
  This follows from Mourelatos’s proposal that the nominalized transcription of an event prediction is normally count-quantified, whereas the one of a process predication is mass-quantified (1978).
• 12
  We should note that in this sense of dynamic, it is highly contentious whether events are not dynamic, and thus more doubtful that P1 is true. The matter depends on how modally robust we take events to be and whether mereological essentialism is true of all events. However, I shall set this issue aside as I cannot hope to properly address it here. For a discussion, see Hinshelwood (2022).
• 13
  Hinshelwood offers a response which denies that events are static (2022). Crowther challenges the idea that processes are capable of real change by proposing that we view changes in processes as changes in their underlying substance(s) (2018).
• 14
  This sentence admits multiple readings, some of which make my case less convincing, see section 5.
• 15
  The case of buttering the toast is misleading, because it could be meant as an act (more specifically an accomplishment that is over once the whole toast is buttered) or as an activity.
• 16
  What if, one might ask, it designates an event? There are two possible scenarios. Once the act of climbing-Mount-Kilimanjaro- to-its-summit is achieved, one may use the sentence to designate this event but from an ‘internal’ perspective—as opposed to the perspective of its completion. If the act of climbing-Mount- Kilimanjaro-to-its summit is not achieved, then it designates a different act. But if it refers to an act (albeit a different one), it does entail the existence of something the agent reached, for instance a certain point halfway to the summit. That is because acts have an essential endpoint. This is analogous to the way in which the truth of ‘I was building a house but never built it’ entails’ that there was something x I was building. Even if this x is not a house, since I never built one. For a discussion, see Szabó (2008).
• 17
  This argument is endorsed in one form or another by Charles (2018); Hinshelwood (2022); Frost (2019) and Thomson (2014).
• 18
  The formulation is Hinshelwood’s (2022, 252).
• 19
  It should be noted that Charles takes this argument to show that actions are not identical to their result (2018).
• 20
  Lorimor may believe that she is engaged in an action of climbing Mount Kilimanjaro and that the terminus point of such an action will be the moment at which she stands at Uhuru peak. But she does not know whether there will be one such actual terminus point.
• 21
  This is expressed in Anscombe’s idea that ‘my practical knowledge of what I am doing would be the same even if I failed to do what I wanted’ (2000, 82); see also Thomson (2014, 209).
• 22
  One way of arguing for the opposite would be to maintain that if an agent intends to be doing an action of kind K, then he counts as doing an action of a certain kind K—keeping in mind that there are constraints on what an agent can intend (Paul 2009, 15-16). This is a controversial claim about the individuation of action and I have no room to provide a response here.
• 23
  Some theorists of the knowledge involved in intentional action restrict it to actions we do without doing something else, like Setiya (2007) or activities like Thomson (2014, 209).

• Article info
  CDD: 128.4
• Funding:
  No funding agency.

Publication Dates

• Publication in this collection
  16 Dec 2024
• Date of issue
  2024

History

• Received
  19 Mar 2024
• Reviewed
  24 Sept 2023
• Accepted
  16 Oct 2024