Self-deception is a form of motivated misbelief, but not every motivated misbelief counts as self-deception. Sometimes, motivated misbeliefs are accidental consequences of motivational states. We refer to these as “wishful thinking” rather than self-deception. Suppose S strongly desires p to be the case, and this desire distorts S's assessment of the evidence so that S comes to unjustifiably believe p. In this scenario, neither S nor any process in S aims at producing misbelief. Because it has no aim, wishful thinking cannot succeed or fail. In the paradigmatic case of self-deception, S believes not-p, desires strongly p, and either S (or some mechanism in S) aims at causing S to misbelieve that p. In this case S (or some mechanism in S) succeeds in producing a misbelief if S ends up believing p, and fails if S ends up believing not-p. I will refer to this as the normative distinction between self-deception and wishful thinking.
Self-deception and wishful thinking both have interpersonal counterparts. Suppose you are driving through an unfamiliar town, and you ask a man for directions to the nearest gas station. He tells you to turn right at the next light, but unbeknownst to him, the gas station has closed down. He has not deceived you, because he did not intend to cause you to have a false belief, and therefore did not succeed in causing you to have a false belief. However if he had intended to lead you astray, he would have succeeded in deceiving you.
Modeling self-deception on interpersonal deception requires the assumption that the self-deceiver intends to cause himself to misbelieve. This way of looking at things requires that self-deceivers simultaneously believe that p and believe that not-p, and it also requires that self-deceivers conceal their self-deceptive intentions from themselves, both of which seem impossible. One response to these problems is to posit that self-deceivers have divided minds. The self-deceiver's mind contains a homunculus that intentionally causes him or her to misbelieve. There are numerous problems with this suggestion, not the least of which is its utter implausibility.
These sorts of problems motivate the deflationist approach to self-deception. Deflationists abjure modeling self-deception on interpersonal deception and argue that self-deception is an unintended consequence of information-processing biases. In Mele's (Reference Mele2001) formulation, these cognitive biases aim at avoiding costly errors, not at causing misbelief. Misbelief is, so to speak, a side effect. The problem is that this eliminates the normative distinction between self-deception and wishful thinking. The processes that cause S to misbelieve do not aim at doing this, so self-deception cannot succeed or fail, and no principled distinction can be drawn between self-deception and wishful thinking.
Von Hippel & Trivers (VH&T) make it clear that they are deflationists, but they also want to model self-deception on interpersonal deception and recognize (in the example of the husband coming home late from work) that intent is necessary for interpersonal deception. In short, they appear to be advocating two incompatible models of self-deception: deflationism, which denies that self-deception is intentional, and intentionalism, which requires it.
There is a reason VH&T have put themselves in this unenviable position. They recognize that the deflationist approach is empirically appealing, but it is important for their whole conception of self-deception that it is purposive. My guess is that VH&T are drawn to think about self-deception as similar to interpersonal deception because they want to capture its goal-directed character, but they do not notice that this plays havoc with their commitment to deflationism.
My diagnosis is that VH&T have made the erroneous assumption almost everyone working on self-deception makes: They mistakenly equate other-deception with interpersonal deception. This is a mistake because nonhuman organisms also engage in other-deception. A comprehensive theory of other-deception needs to encompass the whole spectrum, ranging from bee orchids' deception of their pollinators (which is clearly not intentional) to deliberate human lying (which clearly is).
The most promising formulation of a general theory of deception draws upon Millikan's (Reference Millikan1984; Reference Millikan1993) analysis of biological purpose. Due to limitations of space, I will not set out the details of Millikan's theoretical apparatus but will present a somewhat simplified version of an analysis based on it.
Deception=df For organisms O1
and O2
, O1
deceives O2
iff O2
possesses a character C with the purpose F of representing truly and O1
possesses a character C* with purpose F* of causing C to misrepresent, and it is in virtue of performing F* that C* causes C to misrepresent.
This can be crudely paraphrased as follows. One organism deceives another just in case the deceiving organism has some characteristic that has the biological purpose of preventing the representational apparatus of the deceived organism from correctly representing its environment (i.e., fulfilling its biological purpose). The deceiving organism succeeds in causing the deceived organism's representational apparatus to fail. Using this analysis as a model for self-deception, we get:
Self-deception=df
O deceives itself iff O possesses character C with purpose F of representing truly, and character C* with the purpose F* of causing C to misrepresent, and it is in virtue of performing F* that C* causes C to misrepresent.
Roughly paraphrased, an organism deceives itself if and only if it has some characteristic with the biological purpose of causing the representational apparatus in the same organism from correctly representing its environment (fulfilling its biological purpose), and that characteristic succeeds in causing the representational apparatus to fail. This analysis of self-deception can then be made consistent with deflationism by adding that C* fulfills its purpose by biasing the manner in which the organism processes information.
Self-deception is a form of motivated misbelief, but not every motivated misbelief counts as self-deception. Sometimes, motivated misbeliefs are accidental consequences of motivational states. We refer to these as “wishful thinking” rather than self-deception. Suppose S strongly desires p to be the case, and this desire distorts S's assessment of the evidence so that S comes to unjustifiably believe p. In this scenario, neither S nor any process in S aims at producing misbelief. Because it has no aim, wishful thinking cannot succeed or fail. In the paradigmatic case of self-deception, S believes not-p, desires strongly p, and either S (or some mechanism in S) aims at causing S to misbelieve that p. In this case S (or some mechanism in S) succeeds in producing a misbelief if S ends up believing p, and fails if S ends up believing not-p. I will refer to this as the normative distinction between self-deception and wishful thinking.
Self-deception and wishful thinking both have interpersonal counterparts. Suppose you are driving through an unfamiliar town, and you ask a man for directions to the nearest gas station. He tells you to turn right at the next light, but unbeknownst to him, the gas station has closed down. He has not deceived you, because he did not intend to cause you to have a false belief, and therefore did not succeed in causing you to have a false belief. However if he had intended to lead you astray, he would have succeeded in deceiving you.
Modeling self-deception on interpersonal deception requires the assumption that the self-deceiver intends to cause himself to misbelieve. This way of looking at things requires that self-deceivers simultaneously believe that p and believe that not-p, and it also requires that self-deceivers conceal their self-deceptive intentions from themselves, both of which seem impossible. One response to these problems is to posit that self-deceivers have divided minds. The self-deceiver's mind contains a homunculus that intentionally causes him or her to misbelieve. There are numerous problems with this suggestion, not the least of which is its utter implausibility.
These sorts of problems motivate the deflationist approach to self-deception. Deflationists abjure modeling self-deception on interpersonal deception and argue that self-deception is an unintended consequence of information-processing biases. In Mele's (Reference Mele2001) formulation, these cognitive biases aim at avoiding costly errors, not at causing misbelief. Misbelief is, so to speak, a side effect. The problem is that this eliminates the normative distinction between self-deception and wishful thinking. The processes that cause S to misbelieve do not aim at doing this, so self-deception cannot succeed or fail, and no principled distinction can be drawn between self-deception and wishful thinking.
Von Hippel & Trivers (VH&T) make it clear that they are deflationists, but they also want to model self-deception on interpersonal deception and recognize (in the example of the husband coming home late from work) that intent is necessary for interpersonal deception. In short, they appear to be advocating two incompatible models of self-deception: deflationism, which denies that self-deception is intentional, and intentionalism, which requires it.
There is a reason VH&T have put themselves in this unenviable position. They recognize that the deflationist approach is empirically appealing, but it is important for their whole conception of self-deception that it is purposive. My guess is that VH&T are drawn to think about self-deception as similar to interpersonal deception because they want to capture its goal-directed character, but they do not notice that this plays havoc with their commitment to deflationism.
My diagnosis is that VH&T have made the erroneous assumption almost everyone working on self-deception makes: They mistakenly equate other-deception with interpersonal deception. This is a mistake because nonhuman organisms also engage in other-deception. A comprehensive theory of other-deception needs to encompass the whole spectrum, ranging from bee orchids' deception of their pollinators (which is clearly not intentional) to deliberate human lying (which clearly is).
The most promising formulation of a general theory of deception draws upon Millikan's (Reference Millikan1984; Reference Millikan1993) analysis of biological purpose. Due to limitations of space, I will not set out the details of Millikan's theoretical apparatus but will present a somewhat simplified version of an analysis based on it.
Deception=df For organisms O1 and O2 , O1 deceives O2 iff O2 possesses a character C with the purpose F of representing truly and O1 possesses a character C* with purpose F* of causing C to misrepresent, and it is in virtue of performing F* that C* causes C to misrepresent.
This can be crudely paraphrased as follows. One organism deceives another just in case the deceiving organism has some characteristic that has the biological purpose of preventing the representational apparatus of the deceived organism from correctly representing its environment (i.e., fulfilling its biological purpose). The deceiving organism succeeds in causing the deceived organism's representational apparatus to fail. Using this analysis as a model for self-deception, we get:
Self-deception=df O deceives itself iff O possesses character C with purpose F of representing truly, and character C* with the purpose F* of causing C to misrepresent, and it is in virtue of performing F* that C* causes C to misrepresent.
Roughly paraphrased, an organism deceives itself if and only if it has some characteristic with the biological purpose of causing the representational apparatus in the same organism from correctly representing its environment (fulfilling its biological purpose), and that characteristic succeeds in causing the representational apparatus to fail. This analysis of self-deception can then be made consistent with deflationism by adding that C* fulfills its purpose by biasing the manner in which the organism processes information.