Action as a form of temporal unity: on Anscombe’s Intention

Abstract

The aim of this paper is to display an alternative to the familiar decompositional approach in action theory, one that resists the demand for an explanation of action in non-agential terms, while not simply treating the notion of intentional agency as an unexplained primitive. On this Anscombean alternative, action is not a worldly event with certain psychological causes, but a distinctive form of material process, one that is not simply caused by an exercise of reason but is itself a productive exercise of reason. I argue that to comprehend the proposed alternative requires an account of the temporality of events in general. An event does not simply have a position in time, but is itself temporally structured. With the inner temporality of events in view, the Anscombean conception of action as a specifically self-conscious form of temporal unity is made available for critical reflection.

Keywords:

• Action
• Anscombe
• event
• intention
• practical reason
• process
• time

Notes on Contributor

Douglas Lavin is Lecturer in Philosophy at University College London, UK. He works primarily on ethics, history of ethics and action theory. Published work includes “Must there be basic action?” and “Other wills: the second person in ethics.”

Acknowledgements

I am particularly indebted to discussions with Matthew Boyle, Matthias Haase, Eric Marcus, Ram Neta, and Matthew Silverstein. I benefited as well from comments at Dartmouth and Oxford. I am grateful to the Alexander von Humboldt-Stiftung for generous financial support.

Funding

This work was supported by the Alexander von Humboldt-Stiftung.

Notes

1. Does the fact that S’s being F implies S’s being G, but that S can be G without being F, entail that it must be possible to decompose S’s being F into S’s being G and S’s being H, where H is some nontrivial further condition, not identical to F itself? Answer is: No. For instance, S’s being red implies S’s being colored, and S can be colored without being red, but there is no prospect of analyzing what it is for S to be red into S’s being colored plus some nontrivial further condition. Likewise with lots of cases: being a parent and being an ancestor, arguably; being a horse and being an animal, arguably; knowing something and believing something, arguably (Williamson 2000); perceiving something and having an experience of something, arguably (McDowell 1982). To infer that, if being G is a necessary but not sufficient condition for being F, there must be some nontrivial H which, when conjoined with G, constitutes a necessary and sufficient condition for being F, might be called the fallacy of analysis. For particular application to the analysis of action see (Ford 2011). This section develops an earlier treatment of the decompositional approach (Lavin 2013b).

2. Must we take these appearances seriously? At the very least, there is room for a different view. The verbs ‘wore’ in ‘Jones wore a smile,’ ‘take’ in ‘Smith took a bath,’ and ‘perform’ in ‘I performed an act of moving the matchbox’ do not express real relations even though the surface grammar of these sentences is relational. For further discussion, see Hyman’s (2001) and Sellars (1991) on the logic of ‘looks’.

3. Here is Michael Bratman: ‘We use the notion of intention to characterize both people’s actions and their minds … A theory of intention must address both kinds of phenomena and explain how they are related. A natural approach, the one I will be taking here, is to begin with the state of intending to act … Instead of beginning with the state of intending to act [some theorists] turn immediately to intention as it appears in action … This is, for example, the strategy followed by Elizabeth Anscombe in her ground-breaking monograph, Intention’ (1987, 5). Velleman (2007) approaches the text in a similar spirit.

4. The passage continues, ‘That is to say, where the answers ‘I am going to fetch my camera’, ‘I am fetching my camera’ and ‘in order to fetch my camera’ are interchangeable as answers to the question ‘Why?’ asked when I go upstairs’ (§23). She seems to be thinking that these are clearly all one notion and that we are confused about what that notion is.

5. Bratman again: ‘It is standing in an appropriate relation to such [mental] states that makes an action intentional … We can cite Anscombe, Goldman and … Davidson as, among others, subscribers to this view … Davidson and Goldman insist, while Anscombe emphatically denies, that the appropriate relation is in some sense a causal relation’ (1987, 6).

6. In light of this, we do best to treat her early disparaging remarks (§5) as simply an insistence that we should not take the idea of causality for granted in our inquiry. And, in light of her final account, it would seem that this is because what is at issue in understanding ‘intention’ is laying hold of a certain distinctive form of causality (§48).

7. A number of recent essays are especially concerned to bring out just how utterly different Anscombe’s account is from anything available within the decompositional framework, for example (Hursthouse 2000; Vogler 2001; Ford 2011; Hornsby 2011; Moran and Stone 2011; Stoutland 2011; Thompson 2011).

8. As I read it, the central aim of Anscombe’s opening discussion of the three ways we speak of ‘intention’, as well as of a number of other early remarks, is to show that an account is needed by inducing a condition of Augustinian perplexity. She says that when we are inclined to speak of different senses of a word which is not equivocal, we are ‘in the dark’ about the kind (not simply the content) of concept it represents. And yet, what could be more familiar – the phenomenon of intention is a pervasive and ineliminable part of human life. Moreover, unlike some other pervasive and ineliminable parts of human life, such as cell division, intentional action seems to be a process of which we necessarily have immediate and privileged knowledge. As Anscombe observes, if our concern were simply to determine on any particular occasion whether someone’s saying ‘I am going to such-and-such’ is a prediction or an expression of intention, or if our concern were to determine, when someone is doing such-and-such, whether this is something she is doing intentionally, we could simply ask the subject. The subject who acts is in a special position to tell us what we want to know. And, as we will see, Anscombe’s approach to the general theory of action presupposes this: it is essentially an interrogation of one who does things for reasons. The resulting account is merely the development of the self-consciousness of the agent.

9. A remark on genus and species: To define a certain Fs as ‘members of genus G, with such-and-such specific differences’ need not imply that we can explain what it is to be a G without appeal to an understanding of various concrete species of Gs. The genus might be, and often is, an abstraction from the species, rather than an independently intelligible kind-of-thing-to-be. Furthermore, if X and Y are species of a common genus G, and G’s do A, it need not follow that there is any direct relation between what it is for an X to do A and what it is for a Y to do A. What it is ‘to do A’ might differ essentially from species to species, so that the generic notion of ‘doing A’ is a mere abstraction, not an independently intelligible way to do something.

10. This is not the place to enter into the details of the various ways one might work out a single, undifferentiated conception of the form of event talk. I have used formulations emphasizing the contrast of transitive and intransitive verbs to maintain contact with our earlier discussion of the decompositional theory. Still, to my mind, the account of event representation put in place by Davidson (1966) (which we all know has nothing especially to do with action) and worked out in terrific detail by Parsons (1990) is the most powerful framework within which to develop a decompositional theory of action. The central idea is that ordinary event talk – whether transitive or intransitive (The sun melted the wax, The wax melted), whether of something done intentionally or not (Jones turned on the light, Jones alerted the prowler) – is about a special class of concrete particulars, what Davidson calls ‘events.’ On this analysis, the sentences have the structure of existential quantification over this domain: ‘For some event e, e is such that …’. In this framework, the commitment to the homogeneity of event description shows up most directly in this: the principle of individuation of events is prior to and independent of the truth of any descriptions of someone’s having done something intentionally.

11. The temporality of events and actions will elude us, if we conduct our discussion using only the abstract nouns (e.g. ‘event,’ ‘process,’ ‘happening,’ ‘behavior,’ ‘action’) and event-denoting noun phrases (e.g. ‘Jones’ raising of his arm,’ ‘the matchbox’s moving,’ ‘the movement of the matchbox’) that are the stock-in-trade of much of action theory. The following discussion draws and improves on (Lavin 2013a, 2013b).

12. The claim is not that it is impossible to express a perfective thought in the grammatical present tense. The so-called reportative present does precisely this (e.g. ‘He shoots! He scores!’). The claim is that perfective thought (He shot. He scored.) cannot be analyzed as the past tense of a present tense thought (as of the form ‘It was the case that p’) because there is no present tense thought to do the relevant work (Galton 1984, 1–23).

13. Aristotle (2006), Metaphysics Theta 6, 1048b, 18–35. I was introduced to the philosophical importance of the topic of aspect by Thompson’s (2008, ch.2) and also have been helped especially by Anscombe (1964), Galton (1984), Rödl (2012) and Waterlow (1982).

14. I am presupposing a certain take on this material. First, I am working with the idea that the aspectual contrast is a distinction among ways in which subject and predicate combine to form a complete thought. But if aspect is a form of predication, then it is not to be understood in terms of further material contents whether, say, implicit quantification over temporal intervals (Hamblin 1971), or as implicit quantification over particular events with certain primitive properties (Parsons 1990). Second, I am working with the idea that the metaphysical category of kinesis (event, process) is to be explained through the formal contrast of aspect. I develop and argue for these claims elsewhere (Haase and Lavin Forthcoming).

15. Anscombe asks, ‘Would intentional actions still have the characteristic “intentional,” if there were no such thing as … further intention in acting?’ (2000, §20). She answers: No. Her thought is that the very idea of a general capacity for intentional action contains the idea of a capacity to act on the basis of specifically instrumental thought, or again to act from a further intention or ‘forward-looking motive’.

16. A first step in its defense would be to note (1) how the generic notion of a process proceeding toward some limit admits of a distinction between non-telic processes, in which parts or phases accumulate toward this limit but do not accumulate because they tend toward this limit, and telic processes, in which the accumulation does occur because it tends toward the relevant limit, and (2) how a further distinction can be drawn within the genus of telic processes between non-self-conscious telic processes in which the telic accumulation of parts or phases toward an end does not depend on any apprehension of the relation between parts and whole by the subject, and self-conscious telic processes in which a guiding apprehension of this relation by the subject is implied in the accumulation (Boyle and Lavin 2010).