In an analysis of the role of emotion in self-deception is presented. It is argued that instances of emotional self-deception unproblematically meet Mele's jointly sufficient criteria. It is further proposed that a consideration of different forms of mental representation allows the possibility of instances of self-deception in which contradictory beliefs (in the form p and ~p) are held simultaneously with full awareness.

There are other options than the opposition set by Mele between his own account and the strict interpersonal model. The notion of collective self-deception or “social hypocrisy” is discussed and shown to be nonparadoxical. When an individual consciousness lies to itself, there is often a form of “negative collaboration” with another.

Mele desires to believe that the self-deceived have consistent beliefs. Beliefs are not observable, but are instead ascribed within an explanatory framework. Because explanatory cogency is the only criterion for belief attribution, Mele should carefully attend to the logic of belief-desire explanation. He does not, and the consistency of his own account as well as that of the self-deceived, are the victims.

A distinction between the self and its superstructure, the ego, supports Mele's conclusions. The dynamics of the limbic system generates the self through behavior that is subject to societal observation. The rest of the brain contributes awareness that, by ingenious back-dating and rationalization, gives the ultimate in self-deception: the illusion of control of the self by its own derivative.

The mechanisms invoked to demonstrate how self-deception can occur without intention or awareness imply that self-deceptive beliefs are nevertheless the outcome of inappropriate and often egoistically driven processes. In contrast, models of pragmatic reasoning suggest that self-deception may well be the “reasonable” output of a more generalized, adaptive approach to hypothesis testing.

I find three major shortcomings in Mele's account. First, verbal ambiguities suggest that the analysis is irrelevant to self-deception and/or that the traditional conception is subtly reinstated. Second, the data offer no means of establishing the superiority of the present account. Finally, as political rhetoric, Mele's proposal not only operates to disqualify others, but establishes science as their judge.

A major worry in self-deception research has been the implication that people can hold a belief that something is true and false at the same time: a logical as well as a psychological impossibility. However, if beliefs are held with imperfect confidence, voluntary self-deception in the sense of seeking evidence to reject an unpleasant belief becomes entirely plausible and demonstrably real.

A form of self-deception exists that is both intentional and common. In it, people act as if they are undergoing a certain state of mind as a tactic for experiencing the state. This kind of self- deception can be illustrated by what happens to players of simulation games. Someone playing a pilot in a flight simulator game, for example, comes to experience aspects of the world of a pilot. Research on hypnotic responding is used to illustrate the nature and effectiveness of such a strategy of self-deception.

Mele's modified definition of self-deception is consistent with evolutionary theory. Self-deception is most likely whenever ignorance confers (reproductive) advantage, namely, in impression management, deception, conformity, social norms, reproductive knowledge, and existential conflicts. Second-order self-deception (unawareness of unawareness) perpetuates self-deception and may be the reason for our misguided definitions.

Mele's jointly sufficient conditions for self-deception preclude definitive ascriptions of self-deception in practice. Consequently, actual ascriptions of self-deception require large inferences and may frequently be in error. It is recommended that attention be directed toward actual practices of ascription to understand how children learn and adults dispense what is ultimately a moral judgment.

As understood by neodissociation and sociocognitive theorists, hypnotic responses are instances of self-deception. Neodissociation theory matches the strict definition of Sackeim and Gur (1978) and sociocognitive theory matches Mele's looser definition. Recent data indicate that many hypnotized individuals deceive themselves into holding conflicting beliefs without dissociating, but others convince themselves that the suggested state of affairs is true without simultaneously holding a contrary belief.

Those who invoke the word self-deception to represent one phenomenon often argue that those who use it to represent another are misusing the construct. Better to recognize that self-deception is a fuzzy concept that may be used to represent a variety of mental processes and states, and to direct our energy toward distinguishing empirically among its forms and functions.

This commentary concentrates on two flaws in Mele's account. The first is Mele's attempt to account for self-deception by appealing to a desire to believe, together with an instrumental belief concerning the means of satisfying this desire. Contrary to Mele, it is argued that such an account requires a recognition on the part of agents that their actions instantiate these means. Second, Mele misidentifies the most essential – and flawed – ingredient of the standard approach to self-deception, the agent's desire to form the belief (the belief that is undermined by the evidence). This ingredient is retained in Mele's own account of self-deception.

There is little fear that the concept of motivational bias as proposed by Mele is likely to dampen the current academic ferment (see Mele's Introduction) with respect to self-deception for several reasons: (a) like philosophy, science has more recently abandoned the heuristic of a rational human mind; (b) the concept is parsimonious, applicable to many research topics other than self-deception, and, therefore, scientifically serviceable; (c) as a proximal mechanism it addresses process rather than function, that is, how rather than why questions; (d) it is not as interesting a question as why there is a high prevalence of “real” self-deception (i.e., “garden-variety self-deception” as described by Mele, see sect. 6); and (e) a more penetrating issue is whether “real” self-deception is adaptive.

Although Mele's four sufficient conditions for self-deception are on track insofar as they avoid the requirement that self-deception involves contradictory beliefs, they are too weak, because they are broad enough to include cases of bias or prejudice that are not typical cases of self-deception. I discuss what distinguishes self-deception from other forms of bias.

Contrary to Mele's suggestion, not all garden-variety self-deception reduces to bias-generated false beliefs (usually held contrary to the evidence). Many cases center around self-deceiving intentions to avoid painful topics, escape unpleasant truths, seek comfortable attitudes, and evade self-acknowledgment. These intentions do not imply paradoxical projects or contradictory belief states.

An important way to become self-deceived, omitted by Mele, is by intentionally ignoring and avoiding the contemplation of evidence one has for an upsetting conclusion, knowing full well that one is giving priority to one's present peace of mind over the search for truth. Such intentional self-deception may be especially hard to observe scientifically.

The essence of a mental event such as self-deception lies in its function – its place in the life of an animal. But the function of self-deception corresponds to that of interpersonal deception. Therefore self-deception, contrary to Mele's thesis, is essentially isomorphic with interpersonal deception.

Mele questions the prevalence and ontological status of strong forms of self-deception, as well as our attempt at experimental demonstration. Without validated indicators outside laboratory contexts, statements about prevalence are purely speculative. Conceptualizing self-deception without positing the motivated lack of awareness of a contradictory belief is unsatisfactory in dealing with issues of “agency,” that is, how can we stop the processing of threatening information unless we recognize that the information is threatening?

Empirical evidence of self-deception's propositional duality is not sought; philosophically relevant links between propositions proper and mind are explored instead. Speech in unison ably indicates the social grounding of such attitudinal structures. An extra-theoretical eye – with regard to cognitivism – is cast on a case of “illusory communication.” The reinforcing of lexical analysis shows Mele's approach to be in need of non-ego concepts, wherefore it lacks soundness with respect to reference.