I use the normative theory of intentional omissions to argue that ordinary uses of ' s intentionally' are systematically ambiguous. There are occasions where they might be used to attribute intentional omissions rather than intentional actions. One can thus explain a number of puzzles that have been taken to be puzzles about the concept of intentional action: the Knobe effect, the connection between foresight and intentionality (in legal contexts), the Butler problem, the compatibility of moral luck with intentionality, the requirement of ability to do otherwise, the intentionality of akrasia.
We do some things intentionally. Sometimes we also intentionally fail to do things. It is natural to think that in the former case we are speaking of intentional actions, while in the latter of intentional omissions. Whether or not omissions are actions, it is natural to think that the concept of intentionality is the same.
In this paper, I will sketch how to use the normative theory of intentional omissions to explain a number of puzzles about the concept of intentional action. In §4.1, I show how to explain the so-called Knobe effect (according to which people’s attributions of intentional action depend on normative context). In §4.2, I sketch how this explanation can be extended to account for the connection between foresight and intentionality (in legal contexts). In §4.3, I explain the so-called Butler problem. In §4.4, I show how certain types of moral luck are compatible with intentionality. I also show how those types of cases are the flip-side of Frankfurt-type cases. The account can further explain the intentionality of akrasia (§4.5).
I should make clear that the paper does not pretend to offer a fully worked out argument for any of the claims I make. I believe that the main idea is worth advertising and worth defending, though a proper defense will need to be done on another occasion.
On the normative account of intentional omission ([
Interestingly, the normative account of intentional omission has often been thought to
require only the agent’s awareness (true belief) or knowledge about the conditions of
the omission (see [
I will not argue for the normative account of intentional omissions. There are many
problems that the account needs to sort out. The notion of ability, for example, is so
unclear that one may reject any account that does not spell it out further ([
Let me mention and briefly respond to one problem.1 Suppose that Chris tries to do something that he can and should do but fails to do it. In such a situation, clearly the addition of Chris’ knowledge that he failed what he could and should have done does not amount to making the failure intentional. There are at least two ways to respond. One would be to simply exclude such cases by the addition of a further condition (Oe) that the agent did not try to . There are legal codes (for example, the Polish criminal code) where this is done almost explicitly. Aside from the requirement of knowledge (foreseeability) one requires acquiescence to the foreseen failure.2 There is a legal debate what acquiescence means. On some interpretations, it is some sort of volitional state. On others, it simply means that the agent did not try (or did not try reasonably hard enough) to fulfill the duty.
On the other hand, there are surely cases of intentional omissions to fulfill the duty to where the agent tried to . I tried to file the taxes, I started filling out the forms. I just did not complete or send them out in the end. Perhaps I lied on the couch all day and could not put down a novel I was reading. I knew I would miss the deadline if I don’t put it to a side. It seems to me that in such a case I intentionally omitted to file the taxes after all even though I may have tried to file them. However, suppose that the reason why I did not send out the tax forms in the end is that as I started to calcu1 I want to thank one of the reviewers for pointing it out. 2 “A prohibited act is committed intentionally, if the agent has the intention of committing it, i.e. either wants to commit it or, having foreseen the possibility of committing it, acquiesces to such a possibility.” (Kodeks Karny Art. 9. §1, my translation) late the taxes I owe, I suffered a cardiac arrest and was taken to the hospital. I knew all along that I am missing the deadline, which only aggravated my state. In this case, however, it is arguable that I was unable to fulfill the duty (i.e. that condition (O-b) was not satisfied). One suspicion that a normative theorist may harbor about cases of failed tryings is that such failures indicate that the condition of ability may not have been satisfied after all. This would, however, demand a careful investigation of the concept of ability.
Let me stop here and bag all the problems for the normative account of intentional omissions. I acknowledge that the bag exists. The ideas I put forward depend on an appropriate way of sorting out the problems in the bag. I want to argue, however, that the normative theory of intentional omissions promises to offer results so interesting that they make it worthwhile taking another look at the problems in the bag in the hope of solving them.
Before proceeding it will be useful to emphasize that there is no agreement on the
concept of intentional action [
If we accept (O) and (I) then the intentionality of an omission is fundamentally
different from the intentionality of an action. In intentional action, we execute our own
commitments, the commitments we have undertaken in intending to do something
[
There are many concepts of omissions. Some are ontological: e.g. Clarke [
In fact, I want to show that on the normative theory of intentional omissions, some actions (ings) can be described as the agent’s intentional omissions. Consider a negative duty, a duty not to . It is possible to fail to do what is required by a negative duty. In general, we can apply the conditions of (O) to such a case in the following way. intentionally omits not to just in case: (a) should not , (b) is able not to , (c) s (and so fails not to ), and (d) is aware of (a), (b), (c).
One may have a number of reservations here. Is it at all possible to omit not to do something? If one accepts the normative theory of omissions, there is prima facie no reason why it should not be possible. The theory can be applied to such a case. This is not to say, however, that we go about attributing omissions not to to one another (not explicitly at any rate).
In fact, there are very good reasons why such descriptions will likely not be found in natural languages (not just English). For these descriptions contain a double negation. One of the negations is present in the content of the negative duty while the other is part and parcel of the concept of omission (to omit to is not to after all). Indeed, given the general linguistic pressures to cancel double negations,5 it might be argued that when intentionally omits not to , in fact intentionally s. Here is a simple argument (I have dubbed it Γ) that shows how to do so.
(1) (2) (3) (4) intentionally omits not to . So: what intentionally does not do is not to . So: intentionally does not not . So: intentionally s.
The argument relies on seemingly intuitive principles: (Γ1) If intentionally omits (not) to then intentionally does not (not-). (Γ2) If intentionally does not not- then intentionally s. (Γ2) simply removes the awkward double negation. While (Γ1) appears to be a special case of the claim that if omits to then does not , it is in fact problematic. If one takes the intention condition to be necessary for intentional action (I) while accepting the normative theory of intentional omissions (O), which requires only the satisfaction of the knowledge condition, then one has a reason to deny (Γ1) [44].
If the account is accepted thus far, we reach the conclusion that ascriptions ‘ s
intentionally’ could be systematically ambiguous. I will argue now that this
theoretical possibility generated by the normative theory of intentional omission can actually
cast light on a number of troubling phenomena. Philosophers of action have typically
here is debatable but the conclusion does not follow. I believe that both actions and
omissions are expressions of agency. It does not follow that the concept of intentional action and
the concept of intentional omission are the same. A normativist could even restrict the
concept of omission so that it applies only to actions (by making condition (O-c) agentive).
Given such a restriction, all intentional omissions would be intentional actions, but it would
still not be the case that all intention omissions to were intentional actions of not ing.
This is because the intentional omissions to could be intentional actions under a different
description than that of ‘not ing’ [
4.1
In Knobe’s classic study [
They started the new program. Sure enough, the environment was harmed.” The subjects were supposed to say how strongly they agreed or disagreed with the claim that the chairman intentionally harmed the environment. Most subjects (82%) agreed with it:
(A) The chairman harms the environment intentionally.
Knobe presented an alternative version of the story to other subjects, which differed from the original only in that the word ‘help’ was put in place of ‘harm’ in the three marked places and in that ‘and’ was substituted for ‘but’. Most people (77%) who read the help vignette were inclined to disagree with the claim that the chairman intentionally helped the environment: (~B) It is not the case that the chairman helps the environment intentionally.
The asymmetry in intentional action attributions is surprising because the two
cases seem to be fully symmetrical with regard to the conditions that have been taken to
be relevant in orthodox theories of intentional action (I). The chairman seems to be in
exactly the same state of mind in both cases, the way in which the state of mind
relates to the ensuing events and the ultimate result (the harming/helping the
environment) again does not seem to be different. The most striking difference between the
cases concerns the normative context. Since such features of the normative context
have been largely thought to be irrelevant to the orthodox notion of intentional action,
the results are puzzling, cf. [
However, the orthodox theories of intentional action (I), which require an intention
(or more generally a pro-attitude) toward the performed action can explain why
people are inclined to assert (~B). The chairman neither intends nor wants to help the
environment. Indeed, when people are asked to explain why they assert (~B), they do
cite precisely the fact that the chairman did not intend or did not want to help the
environment [38]. The orthodox theories of intentional action are thus capable of
explaining one side of the puzzle. They fail in explaining why people are inclined to
assert (A). In fact, when people are asked why they hold (A), they tend to cite the fact
that the chairman knew the environment would be harmed [38], cf. [
I have argued [44, 45, 46] that one can accept what I have called the omissions account, which conjoins the orthodox theory of intentional action and the normative theory of intentional omission. The normative theory of omission provides a natural explanation of the harm case in the Knobe effect. If we look for intentional omissions in Knobe’s cases, we will find one in the harm case: the chairman commits an intentional omission not to harm the environment. Presumably, he has a duty not to harm the environment (Oa). It is within his power not to harm the environment (Ob), for it is within his power not to start the program. He does harm the environment (Oc), i.e. fails to do the duty.6 His omission is intentional because he knows all of this (Od).7 Thus: (C)
The chairman intentionally omits (to do his duty) not to harm the environment.
As argued (§3), there are reasons to think that the problematic intentional action attribution (A) is a disguised form of the intentional omission attribution (C). We can use argument form Γ to derive (A) from (C). Since the chairman intentionally omits (to do his duty) not to harm the environment (C), what he intentionally does not do is not to harm the environment. In other words, he intentionally does not not-harm the environment. When we cancel the awkward double negation, we obtain the claim that he intentionally harms the environment (A).
However, it would be erroneous to think that in assenting to the claim that the
chairman intentionally harmed the environment, people attribute an intentional action
to the chairman. The concept of intentionality is different. The intentionality of action
requires an intention or pro-attitude (I), while the intentionality of omissions requires
knowledge (O)8. Indeed, as already mentioned, when people are asked why they claim
that the chairman intentionally harmed the environment they appeal to the fact that the
chairman knew that the environment would be harmed [38]. This is just what we
would expect if the attribution of intentional harming were an intentional omission
rather than an intentional action.
6 Note that one can consistently think that he harms the environment without thinking that he
performs an intentional action of harming the environment. He does something else
intentionally, i.e. he intentionally starts the program, which causes the environment to be harmed.
7 One might worry that the story does not tell us that the chairman knows that he should not
harm the environment (I thank one reviewer for noting this). This point can be granted,
though it is natural to think of such knowledge as part of common knowledge. At any rate, I
claim that those people who hold that the chairman intentionally harmed the environment
(and not all do [38], cf. [
The results reached above can be extended to explain quite generally why foresight
(without intention) is sufficient for intentionality in some cases [
Note too that the connection between intentionality and foresight is likely to be
very common in law. This is because lawyers are mostly concerned with negative
duties designed to prevent harm. Failure to abide by those duties will be intentional as
long as the knowledge condition is satisfied.9 Legal theorists sometimes use the
construct “oblique intention” to cover such cases. If intentional omissions are understood
properly, the motivation for introducing oblique intentions will be removed, cf. [
Butler [
(~E) It is not the case that Brown throws a six intentionally.
Once again this is surprising. Luck has been thought to exclude intentionality ([
When we look at the cases in search for intentional omissions, we will find that we can ascribe an intentional omission to Brown in the killing case: (F)
Brown intentionally omits not to kill Smith.
All of the conditions in (O) are satisfied: (a) Brown has the duty not to kill Smith, (b) it is within her power not to kill Smith (it suffices that she not fire or touch the gun), (c) she does kill Smith (again one need not think that she kills Smith intentionally; she presses the trigger intentionally, which causes the death of Smith), (d) she is aware of all of this.
Once again we can use Γ to transform (F) into (D). Since Brown intentionally
omits not to kill Smith (F), she intentionally does not not-kill Smith. When we cancel
the awkward double negation, we obtain the result that she intentionally kills Smith.
We thus arrive at the claim we are intuitively inclined to give (D). Because there are
9 Indeed, there are a number of alternative philosophical theories of intentional action that
embody the idea that foreseen but unintended consequences of intentional action are brought
about intentionally (cf. [
As Lowe [
Philosophers have sometimes thought that intentional action requires dual powers: we
need to be able to do what we do and we need to be able to do otherwise [
For most ordinary action-types we have dual powers. I am able to raise an arm (I’m reliable in fulfilling the task to raise the arm) as well as I am able not to raise the arm (I’m reliable in fulfilling the task not to raise the arm). But other combinations are possible. Consider someone at early stages of Parkinson’s disease, who cannot control the tremor of his hands. Such a person is no longer reliable in performing many manual tasks, such as pressing a certain button. The person will occasionally hit it but not reliably. In the wanted sense, he is unable to hit the button. However, he may very well be able not to hit the button. (We would think that he is unsuited for the job of a traffic controller, which requires precise button pressing, but that he is perfectly suited for the job of a night guard in an aquarium, which involves the duty not to press buttons that initiate the removal of water from the tanks, for example.)
Intentional action has been thought to require the agent’s ability, skill, reliability to
[
We have seen reasons to think that when we use the form of words ‘ s intentionally’ we may have in mind that performed the intentional action of ing or that committed the intentional omission not to . In cases where the agent has dual powers, depending on whether other conditions of (I) and (O) are fulfilled, the agent may have performed an intentional action or committed an intentional omission or both. If the agent is not able to but does as a matter of luck (Butler-type cases), then as long as he is able not to (and the other conditions of (O) are satisfied), the ascription of intentional ing is an ascription of an intentional omission not to . This shows in a general way how luck is compatible with intentionality.
“ s intentionally” is able to is able not to (Dual powers cases) intentional action intentional omission
(Butler cases) intentional omission (Frankfurt cases) intentional action
The third possibility of ascribing intentional ing involves cases where the agent is
able to but is not able to do otherwise. Such cases have been made famous by
Frankfurt [
If one understands the dual nature of intentionality, one will see that the connection between responsibility and intentionality is doubly secured, so to speak. It is relatively immune to luck (Butler cases) and to the lack of ability to do otherwise (Frankfurt cases). 4.5
The normative account of intentional omission can also answer the question why akratic behavior is intentional. Akrasia can conceptualized in various ways, but one way to capture its core is to think that it involves a failure to do carry out the commitment that one has undertaken whether in intending to do something or in judging that doing something would be the rational thing for one to do. The paradigmatic cases of omissions are cases of failures to do what is required by an external commitment typically placed by another agent. But clearly it is also possible to omit to do what is required by one’s own commitment (the commitment engendered in one’s intending to do something). This is what happens in akrasia.
The normative theory of intentional omission does not aim to answer all questions about akrasia but it can answer why akratic action is intentional without the agent’s having an intention to do what she does. If akratic behavior is covered by the intentionality of omission then all that is required is the agent’s knowledge of what she is required to do (she knows what she intends), of the fact that she is able to do what she is required to do (arguably she could not intend to do it unless she believed that she is able to do it). What is further required is her knowledge that she does not do what she resolved to do. 5
I have argued that the normative theory of intentional omission shows a great deal of promise in explaining puzzling phenomena usually taken to lie within the purview of theories of intentional action.
It might be thought that there is something disconcerting about accepting such
different accounts of intentional action (I) and of intentional omission (O). But this
duality is not as disconcerting as one might initially think [
I have claimed that what appears to be the bastion of individualism [39, 41, 49] in philosophy of action, viz. the concept of intentional action, should be buttressed by the nonindividualist concept of intentional omission. To fully account for our uses of ‘ s intentionally’, we need to invoke not just the concept of intentional action but also the normative – and, at roots, social and interpersonal – concept of intentional omission.
Acknowledgements. I want to thank three anonymous reviewers for many interesting, useful and detailed critical comments. I have tried to accommodate most of them in the paper, though because of the length restrictions I am unable to do full justice to them. The project has been financed by an NCN grant (DEC-2012/05/B/HS1/02949). 35. Mele, A.R., Sverdlik, S.: Intentional Action and Moral Responsibility. Phil. Stud., 265-287 (1996) 36. Nadelhoffer, T.: The Butler Problem Revisited. Analysis, 64, 277-284 (2004) 37. Nadelhoffer, T.: Skill, Luck, Control, and Intentional Action. Phil. Psychol. 18, 341-352 (2005) 38. Nichols, S., Ulatowski, J.: Intuitions and Individual Differences: The Knobe Effect
Revisited. Mind Lang. 22, 346-365 (2007) 39. Paprzycka, K.: Collectivism on the Horizon: A Challenge to Pettit’s Critique of
Collectivism. Australas. J. Philos. 76, 165-181 (1998) 40. Paprzycka, K.: Normative Expectations, Intentions, and Beliefs. SJP 37, 629-652 (1999) 41. Paprzycka, K.: False Consciousness of Intentional Psychology. Phil. Psychol. 15, 271-295 (2002) 42. Paprzycka, K.: Social Anatomy of Action. Towards a Responsibility-Based Account of
Action. Ph.E. dissertation, Univ. of Pittsburgh (1997) 43. Paprzycka, K.: Social Re-Construction of Agency. In: Galavotti, M.C., Dieks E., Gonzalez, W.J., Hartman, S., Uebel, T., Weber, M. (eds.) New Directions in the Philosophy of Science, pp. 323-338. Springer, Dordrecht (2014) 44. Paprzycka, K.: The Omissions Account of the Knobe Effect and the Asymmetry
Challenge. Mind Lang (forthcoming). https://uw.academia.edu/KatarzynaPaprzycka 45. Paprzycka, K.: Intention, Knowledge, and Disregard for Norms: The Omissions Account and Holton's Account of Asymmetrical Intentionality Attributions. https://uw.academia.edu/KatarzynaPaprzycka 46. Paprzycka, K.: The Omissions Account of the Epistemic Side-Effect Effect.
https://uw.academia.edu/KatarzynaPaprzycka 47. Peels, R.: What is Ignorance? Philosophia 38, 57–67 (2010) 48. Peels, R.: What Kind of Ignorance Excuses? Phil. Q. 64, 478-496 (2014) 49. Schmid, H.B.: Plural Action. Springer, Dordrecht (2009) 50. Smith, P.G.: Feinberg and the Failure to Act. Leg. Theory 11, 237-250 (2005) 51. Smith, P.G.: Contemplating Failure. Phil. Stud. 59, 159-176 (1990) 52. Smith, P.G.: Recklessness, Omission, and Responsibility. SJP 27, 569-583 (1989) 53. Stoecker, R.: Action and Responsibility. In: Lumer, C., Nannini, S. (eds.) Intentionality,
Deliberation, and Autonomy, pp. 35-46. Ashgate, Aldershot, UK (2007) 54. Stoecker, R.: Agents in Action. Grazer Philos. Stud. 61, 21-42 (2001) 55. von Wright, G.H.: Norm and Action. Routledge and Kegan Paul, London (1963) 56. Williams, B.: Acts and Omissions, Doing and Not Doing. In: Making Sense of Humanity and Other Philosophical Essays, pp. 56-64. Cambridge Univ. Press, Cambridge (1995) 57. Wilson, G.M.: The Intentionality of Human Action. Stanford Univ. Press, Stanford (1989) 58. Zimmerman, M.J.: Taking Some of the Mystery out of Omissions. SJP 19, 541-554 (1981)