Self-deception and the logic of belief

                                  Andrew Jones

                            King’s College London, UK

    The point of departure for this presentation is the brief discussion of self-
deception that appears in Hintikka’s book Knowledge and Belief: An Introduc-
tion to the Logic of the Two Notions, Cornell UP, 1962. Hintikka starts from
a remark by Montaigne: “Some make the world believe that they believe what
they do not believe; others, in greater number, make themselves believe it”, and
gives a formal treatment of (the second part of) Montaigne’s remark that paral-
lels Hintikka’s analysis of Moore’s puzzle about saying and disbelieving. Those
analyses depend crucially on the 4. schema for the logic of belief (later dubbed
the ‘positive introspection schema’): Ba p → Ba Ba p.
    It will be argued that Montaigne’s remark indicates just one of a small group
of ‘self-deception positions’, the others of which are inconsistent if the logic of
belief is that of a (relativised) modal system of type KD4 (Hintikka’s choice),
and all of which are inconsistent if KD45 is adopted (commonly the choice in
AI).
    The presentation will show how to characterise that group of ‘self-deception
positions’ consistently using KD as the logic of belief, and provides an alternative
treatment of Montaigne’s remark and Moore’s puzzle.
    Why should researchers in Informatics concern themselves with self-deception?
At least two reasons: first, there is already a good deal of interest in the phe-
nomenon of awareness in Cognitive Science, and among those computer scientists
who are developing models of self-organising, adaptive systems. Self-awareness,
and thus also constrained self-awareness, of which self-deception is arguably an
instance, is central to those interests. Secondly, the recent book by the distin-
guished evolutionary biologist Robert Trivers provides a fascinating new per-
spective on the importance of self-deception. While the traditional view among
psychiatrists and psychologists has perhaps been that self-deception is essentially
a defence mechanism, Trivers assembles evidence from various sources suggesting
that a distinct strategic (i.e., offensive in contrast to defensive) advantage may
arise from the capacity to self-deceive: it enhances the ability to deceive others.
Many computer scientists have long been interested in communicative decep-
tion, for obvious reasons. If Trivers’ central thesis is right, then the study of
deception in communication among complex, reflective systems should perhaps
go hand-in-hand with the study of self-deception.