Skip to main content

Is there a freedom requirement for moral responsibility?*

  • Phillip D. Gosselin (a1)

The Principle that freedom is necessary for moral responsibility (hereafter referred to as “the freedom principle”) has received a variety of explications, but few philosophers have doubted that in some plausible sense it is true. However, two philosophers have recently challenged it using very different but equally ingenious arguments. J.F.M. Hunter has provided the more obviously direct attack in arguing that considerations of freedom as such are in no way relevant to assessments of moral responsibility. Harry Frankfurt has directed his fire at the version of the freedom principle which says that a person is morally responsible for what he has done only if he could have done otherwise. Both Frankfurt and Hunter point out the significance of their arguments for the determinism/moral responsibility debate: if there is no freedom requirement for moral responsibility, then even if determinism threatens freedom, it does not follow that determinism threatens moral responsibility.

Hide All


1 Hunter J.F.M., “Acting Freely and Being Held Responsible,” Dialogue, XII (06, 1973), 233–45.

2 Frankfurt Harry, “Alternate Possibilities and Moral Responsibility,” The Journal of Philosophy, LXVI (12, 1969), 829–39. Frankfurt's label for the principle he attacks i s not “the freedom principle,” but rather “the principle of alternate possibilities.”

3 Frankfurt, “Alternate Possibilities,” 835–36.

4 Ibid.,831.

5 Ibid., 832–33.

6 Frankfurt's objections to the freedom principle were considered by Blumenfeld David in “The Principle of Alternate Possibilities,” The Journal of Philosophy, LXVIII (06, 1971), 339–45. Blumenfeld accepted those objections arguing only that the principle could be modified to meet them and that the modified version (entirely different from the version I shall propose) does not help the compatibilist cause. However, after completing this paper I discovered an article by Robert Audi that I had previously over-looked and that makes some points very similar to ones made in this paper. See Audi Robert, “Moral Responsibility, Freedom and Compulsion,” The American Philosophical Quarterly, Vol. II (01, 1974), 114. One point of similarity is the examination and rejection of Frankfurt's argument against “the principle ofalternate possibilities.” Audi's criticism of Frankfurt, though briefer and somewhat different from mine, is, I think, effective.

7 Hunter , “Acting Freely,” 244.

8 Dworkin Gerald, “Acting Freely,” Noûs, 4 (11, 1970), 367–83. See also Locke Don and Frankfurt Harry, “Three Concepts of Free Action,” PAS Supplement, XLIX (1975). 95125.

9 Hunter, 237.

10 Ibid., 237–38.

11 Ibid., 238.

12 Although the concept of unfreedom he discusses is rather different from those distinguished above, Robert Audi in “Moral Responsibility” (pp. 11 and 12) makes the point that “acting unfreely” is not always sufficient to excuse.

13 Notice that ifone agrees with Hunter that ignorance ofthe nature or consequences of one's act can render one unfree in a legitimate sense, and ifone accordingly distinguishes a third variety of unfreedom (“the unfreedom of ignorance”?) this does not damage my reply to Hunter's objections. However, as I have implied above, I believe one is less apt to mislead if one distinguishes (non-culpable) ignorance from unfreedom as a ground for excusing.

14. Hunter, 240.

15 Ibid., 243.

16 Winston Nesbitt and Stewart Candlish describe the normative sense of “couldn't” in On Not Being Able To Do Otherwise” in Mind, LXXXII (07, 1973), 321–30. However, they mistakenly maintain that “couldn't have done otherwise” when used as the statement of an excusing condition is always used in the normative sense. They argue (1) that, if a man couldn't have done otherwise in the “powerlessness sense” then he did not do anything (in the narrow sense of “do” which implies intentional action); and (2) that a man can be excused only for what he does (p. 323). But (1) is disproved by extreme cases of compulsion and addiction and (2) is disproved by the person who is excused for accidentally breaking a vase.

It might be thought that the preceding distinction between the normative and the power senses of “could” could be used to mount a critique of Frankfurt's counter-examples alternative to the one provided in part I of this paper. It is true that Frankfurt does not seem to be aware of this distinction; however, what he needs for his argument to have a bearing on the determinism/responsibility issue is the power sense of “could” and the “could's” of both his counter-examples can be taken in that sense. Accordingly, I have interpreted them in the power sense.

17 It is true that the second disjunct of this excusing condition is extremely vague; some might say that it is intolerably vague. The agent is excused if doing otherwise was “sufficiently difficult” for him. But what criterion are we to use in assessing the degree of difficulty? And what degree of difficulty is sufficient to excuse?

I can offer little by way of answer to these questions. As regards thefirstit might be helpful to ask “How weighty are the reasons that would have been required to move P to do what he did not do – or are there any conceivable reasons to which he would have responded?” This procedure has its limitations of course for the weightiness of reasons doesn't admit ofprecise measurement and, furthermore, it is the subjective weight (i.e. the weight for the agent) of the reasons that must be considered and that is something notoriously variable. As regards the second question we must, I believe, sometimes refer to more fundamental moral principles to determine whether it isfittingto blame under the particular conditions in question, i.e. given the degree ofunfreedom that is present. A clear answer will not always be apparent – but this, of course, is not surprising. In spite of this vagueness, however, the second disjunct of the principle tells us something important, viz. that the unfreedom of difficulty can excuse. There are cases where it is very far from clear whether the degree of difficulty is sufficient to excuse; but there are many clear cases as well.

In any event my object in enunciating this principle has not been to state a principle that will be of greater practical value for assessing moral responsibility in particular instances than other principles commonly employed. The point rather has been to show that there is a freedom requirement for moral responsibility and, more specifically, to show that whether doing otherwise was within the agent's power, and if so to what extent, is relevant to the question of whether he is morally responsible for what he did. And that point, of course, bears on the determinism/responsibility question.

18 Robert Audi in “Moral Responsibility” (pp. 2 and 3) uses a similar example to make a similar point.

Notice that I do not maintain that the drunken driver who accidentally kills a pedestrian bears the same moral responsibility as the driver, drunken or otherwise, who does the same with deliberation and malice. The point here is merely that the incapacity of the driver does not serve to absolve him; he is to some extent morally responsible for the pedestrian's death.

It is interesting to compare the preceding case to the case of the person who is equally blameworthy for drunken driving, but who luckily doesn't strike any pedestrians on his way home. The unlucky drunk is to blame not only for drunken driving, but also for killing a pedestrian; the lucky drunk is to blame only for drunken driving. Yet, since this difference is due to nothing but luck, it would seem that the degree of their moral blameworthiness (though not, of course, their legal culpability) must be precisely the same.

19 Notice that in these examples there is question of holding the agent morally responsible for an inability. A person may be morally responsible, then, not only for his actions, but also for states of his mind or body. Indeed he may be morally responsible for the behaviour or state of another person or for impersonal events or states of affairs. However, whatever he is morally responsible for, it is always in virture of something that he did or failed to do and for which he is morally responsible. It is because moral responsibility for actions is in this way basic that the present discussion focusses on actions.

* I wish to thank David Zimmerman of Simon Fraser University for his helpful comments on an earlier version of this paper delivered to the 1977 meeting of the Western Canadian Philosophical Association.

Recommend this journal

Email your librarian or administrator to recommend adding this journal to your organisation's collection.

Dialogue: Canadian Philosophical Review / Revue canadienne de philosophie
  • ISSN: 0012-2173
  • EISSN: 1759-0949
  • URL: /core/journals/dialogue-canadian-philosophical-review-revue-canadienne-de-philosophie
Please enter your name
Please enter a valid email address
Who would you like to send this to? *


Full text views

Total number of HTML views: 0
Total number of PDF views: 5 *
Loading metrics...

Abstract views

Total abstract views: 77 *
Loading metrics...

* Views captured on Cambridge Core between September 2016 - 21st January 2018. This data will be updated every 24 hours.