Frankfurt Cases and 'Could Have
Done Otherwise'

Does Frankfurt's counterexample succeed in showing that for a person to act freely, they need not have been able to act other than they did?

Citation Information

Allan, Leslie 2016. Frankfurt Cases and 'Could Have Done Otherwise', URL = <>.

Person standing at fork in path deciding which to take

1. Introduction

Harry Frankfurt, in his seminal paper, challenges the principle that states that 'a person is morally responsible for what he has done only if he could have done otherwise' [Frankfurt 1969: 829]. In my Allan [2016: §7], I provide a counterfactual conditional analysis of free agency as the ability to have done otherwise, given a person's character. My analysis assumes this very principle that Frankfurt denies. In the same essay [Allan 2016: §4], I also advance four requirements that an intentional act needs to meet for it to be properly regarded as a 'free act' and therefore as morally praiseworthy or blameworthy. I call these four requirements the '4C theory' of free acts.

In this essay, I want to explore how adequately the 4C theory and a character-based counterfactual conditional analysis deal with Frankfurt's argument. In particular, I want to answer these three questions:

  1. How do the four requirements (4C) for an act to be free measure up against Frankfurt's argument?
  2. Does a character-based counterfactual conditional analysis of 'could have done otherwise' survive Frankfurt's critique?
  3. Does a character-based counterfactual conditional analysis vindicate the principle that sources moral responsibility in an agent's capacity to have done otherwise?

I will first recount Frankfurt's argument, as advanced in his original essay, and then proceed to answer each of my questions in turn. Finally, I will offer some comments on why Frankfurt's argument appears so convincing and how we can untangle the confusion contained within it.

2. Frankfurt's Counterexample

Frankfurt advances his argument by way of the following counterexample:

Suppose someone—Black, let us say—wants Jones4 to perform a certain action. Black is prepared to go to considerable lengths to get his way, but he prefers to avoid showing his hand unnecessarily. So he waits until Jones4 is about to make up his mind what to do, and he does nothing unless it is clear to him (Black is an excellent judge of such things) that Jones4 is going to decide to do something other than what he wants him to do. If it does become clear that Jones4 is going to decide to do something else, Black takes effective steps to ensure that Jones4 decides to do, and that he does do, what he wants him to do. Whatever Jones4's initial preferences and inclinations, then, Black will have his way.

What steps will Black take, if he believes he must take steps, in order to ensure that Jones4 decides and acts as he wishes? Anyone with a theory concerning what "could have done otherwise" means may answer this question for himself by describing whatever measures he would regard as sufficient to guarantee that, in the relevant sense, Jones4 cannot do otherwise. Let Black pronounce a terrible threat, and in this way force Jones4 to perform the desired action and prevent him from performing a forbidden one. Let Black give Jones4 a potion, or put him under hypnosis, and in some such way as these generate in Jones4 an irresistible inner compulsion to perform the act that Black wants performed and to avoid others. Or let Black manipulate the minute processes of Jones4's brain and nervous system in some more direct way, so that causal forces running in and out of his synapses and along the poor man's nerves determine that he chooses to act and that he does act in the one way and not in any other. Given any conditions under which it will be maintained that Jones4 cannot do otherwise, in other words, let Black bring it about that those conditions prevail. The structure of the example is flexible enough, I think, to find a way around any charge of irrelevance by accommodating the doctrine on which the charge is based.

[Frankfurt 1969: 835f]

If Jones does what Black wants him to do without necessitating Black's intervention, our common intuition is that Jones acted freely. For the sake of illustration, let's say the act that Black wanted Jones to perform was shooting Smith. In the absence of Black's intervention, how does the 4C compatibilist theory regard Jones' freedom in shooting Smith? On this theory, there are four requirements for an act to be freely chosen: (1) the act must not feel compelled; (2) the act must not be controlled by a third party; (3) the act results from the agent's character; and (4) the agent is cognitively capable of reasoning about the act.

Applying the 4C theory, when Jones shot Smith without Black's intervention, Jones did not feel compelled to do so. So, condition (1) Compelled is satisfied. As Black did not intervene, we can say that Jones was not controlled by a third party when he did what he did. Condition (2) Control is also satisfied. Condition (3) Character is also easily satisfied. The act emanated from Jones' internal psychological dispositions. Finally, if asked, Jones could offer cogent reasons for wanting to shoot Smith. So condition (4) Cognition is also satisfied. After applying the 4C requirements to Jones' act, we can conclude that Jones acted freely in shooting Smith. Our intuition that Jones acted freely is vindicated in the face of all of the suggested types of interventions that Black could have implemented. In answer to the first question I raised in the Introduction, applying the 4C requirements yields the same judgement as Frankfurt: that Jones acted freely and was therefore morally responsible for his action.

The next question is: 'Could Jones have done otherwise than to have shot Smith, and, if so, how should we understand this capability?' In my Allan [2016: §7.2], I proposed an analysis of free acts that made an agent's character a central element of their freedom to choose. On this counterfactual conditional analysis, an agent could have done otherwise if:

Given the agent's character, the agent would have done otherwise in this situation if certain external circumstances were different.

On this analysis, different 'external circumstances' means possible circumstances that are external to the agent's character and that preserve the opportunity for the agent to perform the act and the situational context of the act. Let us apply this to the Frankfurt situation. For Jones to have acted freely when he shot Smith, he must have been able to do otherwise; that is, to have not shot Smith. It must be the case that:

Given Jones' character, Jones would not have shot Smith in this situation if certain external circumstances were different.

Book cover: Introduction to Ethics by Fred Feldman

In Frankfurt's example, Jones was making up his mind whether to do the act or not. So, Jones' 'character' was of the kind that he was partial to shooting Smith without possessing a compulsive desire to do so. The 'situation' is the specific environment in which Jones committed the act (for example, late at night in a dark alley close to Smith's house). The question of Jones' freedom hinges on whether there were any possible external circumstances in which Jones would not have shot Smith.

In considering different external circumstances, we must preserve Jones' opportunity to shoot Smith. So, for example, we cannot consider circumstances such as Smith forgetting to load the gun and his eyesight failing in the poor light. Given Jones' vacillation, it appears there are possible circumstances in which Jones would not have shot Smith. For example, Smith may have successfully pleaded for his life, Jones may have suddenly recalled the time Smith did Jones a huge unrequited favour or Jones realized he was late for his son's graduation. The important point here is that Jones was not being coerced or manipulated or under the influence of drugs or suffering some kind of compulsive mental disorder that made it inevitable that he would shoot Smith.

Now, part of the situation description is Black's waiting in the background to see if Jones pulls the trigger. But note that Black's laying in wait does not impact at all Jones' capacity to change his mind at the time of the incident and to fail to shoot Smith. Taking into account this limitation of Black's capacity at the time to affect what Jones does, we can rightfully conclude that the counterfactual conditional analysis above shows that Jones was free to shoot Smith in the situation described by Frankfurt.

Let's say Jones does change his mind at the last moment and ends up not shooting Smith. How will his abrogation affect his freedom to shoot Smith? In this case, no doubt Black will implement his chosen plan to bring it about that Jones willingly shoots Smith. Let's say when Jones arrives back at his house, he finds Black waiting for him, ready to issue Jones a terrible threat that Jones can't ignore. Jones then goes on to shoot Smith, as Black wants.

On the 4C theory of free acts, Jones did not shoot Smith freely as his act did not satisfy the first requirement; that Jones not be compelled to act as he did. For the same reason, applying a character-based counterfactual conditional analysis reveals that he also could not have done anything other than to shoot Smith. (See Allan [2016: §7.1] for a representative analysis.) On both counts, then, the act of Jones shooting Smith was not freely done. The same conclusion holds for whatever method Black used to bring about the inevitability of Jones shooting Smith. These include all of the other methods suggested by Frankfurt: filling Jones with a mind-bending potion, hypnotizing him and directly manipulating the neurons in his brain.

3. Frankfurt Problem Diagnosis

Frankfurt's case is both unusual and philosophically taxing because it encapsulates a feedback loop in the agent's decision-making. It is a case in which the agent's intention to act sets in train a set of external events that cause the initial intention to change to the opposite intention. In Jones' case, there exists a feedback loop starting from the output of the psychological mechanism that causes him to avoid shooting Smith, progressing through an external mechanism comprising Black's coercion/manipulation and then feeding back to the input of Jones' psychological mechanism that causes him to intend to shoot Smith.

Difficulties can arise for compatibilists who fail to recognize that the events traversing the feedback loop neither occur instantaneously nor simultaneously. In Jones' case, there are two entirely different events for which we need to consider separately whether Jones acted freely:

  1. his first encounter with Smith and
  2. his second encounter with Smith following Black's intervention.

Once we clearly separate the two distinct events, the counterfactual conditional analysis follows through smoothly. We are thus saved from modifying the accepted view that an agent is morally responsible for what they have done only if they could have done otherwise.

Philosophers wishing to make Frankfurt's argument stronger could modify his counterexample by giving Black the ability to tell whether Jones will vacillate and change his mind when he is about to shoot Smith. The scenario can be redrawn such that Black will resolve not to implement his coercion/manipulation strategy only in those circumstances where he is convinced that there is no possibility that Jones will change his intention to shoot Smith. This scenario does away with any decision-making feedback loop, leaving only one act of Jones' to consider. Isn't it the case, then, that in this scenario, when Jones decides to shoot Smith without any intervention from Black and does so, that Jones could not have done otherwise?

The answer, I think, is yes. Jones could not have done anything other than to have shot Smith. However, in this case, it is clear that Jones did not act freely. Consider this. How does Black know with such certainty that Jones will not change his mind at the last moment; that Jones is immune to all contingencies? If there is no possible memory, internal feeling or external event that can persuade Jones to do otherwise, he is either drug induced, hypnotized or otherwise mind-controlled, or suffering some mental aberration that has severely compromised his cognitive capabilities. In other words, he is not a free, autonomous agent able to reason and deliberate about possible courses of action. We can conclude that this modified scenario also gives us no reason to abandon our moral maxim linking an agent's freedom and moral responsibility with their ability to do other than they did.

4. Conclusion

In this essay, I sought to answer three questions. Firstly, I explored whether the 4C theory of free acts yielded the same conclusions about Jones' freedom to act as Frankfurt's. The answer here is yes. In both schemes, Jones acted freely when he acted without Black's intervention and acted unfreely when he acted given Black's coercion/manipulation.

Secondly, I examined whether my counterfactual conditional analysis of Jones' act withstood Frankfurt's claim that Jones acted freely even though he could not have done otherwise. I argued that Frankfurt's conflation of Jones' two acts under consideration as just a single act led him to his mistaken conclusion that Jones could not have acted otherwise, even where Black in fact did not intervene. My analysis led me to answer the final question in the affirmative: a counterfactual conditional analysis of free acts based on an agent's character supports the notion that moral responsibility and freedom of will translate into an agent's capacity to do other than they did.


  • Allan, Leslie 2015. Plantinga's Free Will Defence: Critical Note, URL = <>.
  • Allan, Leslie 2016. Free Will and Compatibilism, URL = <>.
  • Berofsky, Bernard, ed., 1955. Free Will and Determinism, New York: Harper and Row.
  • Campbell, Charles 1967. In Defence of Free Will, In Defence of Free Will, London: Allen and Unwin.
  • Frankfurt, Harry 1969. Alternate Possibilities and Moral Responsibility, Journal of Philosophy 66: 829–39.
  • McKenna, Michael and D. Justin Coates 2015. Compatibilism, The Stanford Encyclopedia of Philosophy (Summer 2015 Edition), ed. Edward N. Zalta, URL = <>.
  • O'Connor, Timothy 2014. Free Will, The Stanford Encyclopedia of Philosophy (Fall 2014 Edition), ed. Edward N. Zalta, URL = <>.

Copyright © 2016

You will be interested in
Book cover: Epistemology: A Contemporary Introduction to the Theory of Knowledge by Robert Audi
Book cover: The Logic of Scientific Discovery by Karl Popper
Book cover: Progress and Its Problems: Towards a Theory of Scientific Growth by Larry Laudan
Book cover: Fads and Fallacies in the Name of Science by Martin Gardner
Book cover: An Inquiry into Meaning and Truth by Bertrand Russell
Book cover: Philosophy of Science: An Historical Anthology by Timothy McGrew, Marc Alspector-Kelly and Fritz Allhoff

Share This

  • twitter
  • facebook
  • linkedin
  • googleplus
  • gmail
  • delicious
  • reddit
  • digg
  • newsvine
  • posterous
  • friendfeed
  • googlebookmarks
  • yahoobookmarks
  • yahoobuzz
  • orkut
  • stumbleupon
  • diigo
  • mixx
  • technorati
  • netvibes
  • myspace
  • slashdot
  • blogger
  • tumblr
  • email
Short URL: