By Crystal Thorpe

Critical Commentary on Alfred Mele’s Self-Deception Unmasked (Princeton and Oxford: Princeton UP, 2001)

Crystal Thorpe, University of Florida 

In the book, Self-Deception Unmasked, Alfred Mele offers a fresh new approach for dealing with the seemingly intractable problem of self-deception. Traditionally, philosophers have taken self-deception to be a phenomenon in which an individual who initially knows or believes the truth, p, intentionally causes herself to believe ~p.1 Two things are important to note about the traditional view. First, self-deception is taken to be an intentional process; and second, the self-deceived agent is characterized as holding conflicting beliefs, that is, she holds both her initial belief that p and her newly acquired belief that ~p. Mele rejects the traditional view by arguing that garden-variety cases of self-deception are not intentional and do not involve the holding of contradictory beliefs. His main ground for rejecting the traditional view is that it does not adequately explain what happens in garden-variety cases of self-deception.

Mele argues that we have evidence from empirical psychology that suggests that typical cases of self-deception are not intentiona land do not involve the holding of contradictory beliefs. 2 I think Mele is right to reject the traditional view. The reason I think the traditional view should be rejected, however, has less to do with evidence we get from empirical psychology and more to do with our attitudes toward self-deceived people. The traditional view fails to capture our attitudes toward self-deceived people. If the traditional view were correct, one would expect self-deceived people to be regarded as desperate, epistemically irresponsible, irrational and perhaps even mentally ill. Yet considering the prevailing social attitudes, how do we customarily regard the broker who thinks she is easier to get along with than all the other associates, or the wife who is self-deceived in believing her husband is not having an affair, or the parent who is self-deceived in believing that her child has not committed a felony? Even if we tend to perceive such individuals to be desperate, epistemically irresponsible, irrational or mentally ill, surely we do not view the majority of self- deceived people in this way. Rather, we tend to regard self-deceived people as being irritating, pitiable, silly or laughable. The prevailing attitudes are compatible with Mele’s model of self- deception and incompatible with the traditional model.

I take it that to be self-deceived is a bad thing. It is not something for which one should strive. Quite to the contrary, it is something that one ought to avoid. If one takes the traditional view of self-deception, it is easy to see why it is bad. For starters, the self-deceived person violates several epistemic norms. A self-deceiver starts off with the belief that p, and then causes herself to have a belief that ~p. She ends up, therefore, with the belief that p and the belief that ~p. This clearly violates epistemic norms governing the consistency of our beliefs. Furthermore, self- deceivers violate norms governing how evidence is to be gathered and interpreted. Now, if the individual violates these norms unintentionally, we are not likely to attribute epistemic irresponsibility to her. The traditionally conceived self-deceiver, however, violates these norms intentionally. To make matters worse, the reasoning that leads the self-deceiver to violate these norms is itself irrational.

Let me say a bit more about what I mean here. Assume that A wants a certain proposition, p, which she believes to be false, to be true. For example, she may believe that the proposition, “My spouse is not having an affair,” is false, yet she may want it to be true. Given that she believes that her spouse is having an affair, she presumably has evidence that supports this belief. For example, her spouse arrives home late, smelling of perfume, at least two nights a week. Her spouse takes two hour lunch breaks out of the office when in the past he routinely ate at his desk. Her spouse has been seen with another woman in intimate settings. And so on. Now, the thought that her husband is having an affair is devastating to our agent. If she thinks about it, she finds herself unable to eat, sleep or function. She desperately wants it to be the case that he is not having an affair. What can she do about this? On the one hand, she could break up the relationship. Notice that if she were to do this, the proposition “My spouse is not having an affair” would be true. This, however, will not solve her problems, for not only does she want the proposition “My spouse is not having an affair” to be true, she wants the proposition “My spouse didn’t have an affair” to be true as well. So, in a desperate attempt to gain control over a situation that is completely out of her control, she decides to cause herself to believe that her husband is not having an affair.

Now, intentional action is made up of three components. One, the agent sets a goal for herself. Here the goal is to believe something she knows to be false. Two, the agent figures out, through a process of rational deliberation, the means to that goal. Here, the means involve deliberating badly at just the right points. And three, the agent acts on this means in order to further that goal. That is, she deliberates badly in just the right way to cause herself to believe that her husband did not have an affair. Clearly, there is something wrong here. Although this intentional act may be successful, that is, although our agent may succeed in causing herself to believe that her husband is not having an affair, and although there may be something rational about the strategy itself, the thinking that led her to perform this intentional act is itself irrational.

If the traditional view were correct, and the self-deceived agent behaved as I have described above, we would tend to think her desperate and epistemically irresponsible, not to mention irrational. We would find her desperate, for only a desperate individual would intentionally engage in such irrational behavior. We would see her as being epistemically irresponsible as well, for it is epistemically irresponsible to intentionally flout epistemic norms. Furthermore, we would perceive her to be irrational, for intentionally causing oneself to believe something that is false is clearly irrational. It is no help to say that the traditionally conceived self-deceived agent unconsciously intends to deceive herself or that there is some mental partitioning that takes place. Hypothesizing such mechanisms does help to explain how self-deception, traditionally conceived, is psychologically possible. Furthermore, it takes some of the sting out of its irrationality. However, if such mechanisms were indeed at work in the self-deceived person, most of us would tend to think that the self-deceived person verges on being mentally ill.

I realize that my portrayal of the traditionally conceived self-deceived person may be a bit unfair. However, if you take seriously the claim that self-deception is an intentional phenomenon, the self-deceived person becomes one to whom you would attribute desperation, epistemic irresponsibility, and worst of all perhaps even mental illness. Mele rejects the claim that self- deception is intentional and in so doing he presents us with a characterization of the self-deceived person that more closely matches our attitudes towards the self-deceived. On Mele’s view, an agent does not intentionally cause herself to believe something that is not the case. Rather, cognitive mechanisms are activated in us by strong feelings and desires that have the effect of making us believe something that is not true. Often these mechanisms are activated without our being aware of it. Sometimes, on the other hand, we intentionally activate them. In these cases, however, we don’t activate them in order to deceive ourselves. Rather, we have another goal in mind. Perhaps we do it to avoid pain or to make ourselves feel happy—as Mele’s character Beth intentionally focuses on pleasant memories of her father in order to avoid the painful truth that he did not love her best. 3 Although Beth’s goal is to avoid pain, she inadvertently causes herself to believe that her father cared for her more than he actually did.

On Mele’s account, self-deception is something that happens to us rather than something that we do. This isn’t to say, however, that we cannot avoid it. One can reflect on one’s motivations and cognitive processes and try to stop oneself from becoming self-deceived. Notice, however, that when one fails to do this, we do not tend to attribute epistemic irresponsibility to that person. Rather, we tend to pity that person or become irritated with her. Furthermore, there is nothing desperate about Mele’s self deceived person. This, in part, is because the phenomenon as Mele sees it is all too human. It happens to the best of us. Mele’s self-deceived agent captures our attitude towards the self-deceived. The traditional view does not.


Works Cited 

Davidson, D. “Deception and Division.” Actions and Events. Eds. E. LePore and B. McLaughlin. Oxford: Basil Blackwell, 1985. 138-48.

Mele, Alfred. Self-Deception Unmasked. Princeton and Oxford: Princeton UP, 2001.

Pears, D. Motivated Irrationality. Oxford: Oxford UP, 1984.

  1. See D. Davidson, “Deception and Division” in E. LePore and B. McLaughlin, eds., Actions and Events (Oxford: Basil Blackwell, 1985) 138-48 and D. Pears, Motivated Irrationality (Oxford: Oxford UP, 1984).
  2. See Chapter 2 of Alfred Mele’s Self-Deception Unmasked. Princeton and Oxford: Princeton UP, 2001.
  3. Mele 18-19.

Crystal Thorpe

Crystal Thorpe is Assistant Professor at the University of Florida. Her primary research interests include practical rationality, reasons for action, Kantian ethics and normative ethics.