In this dissertation I direct critique at a conception of self-deception prevalent in analytical philosophy, where self-deception is seen as a rational form of irrationality in which the self-deceiver strategically deceives himself on the basis of having judged that this is the best thing to do or, in order to achieve something advantageous. In Chapter One, I criticize the conception of self-deception as analogous to deceiving someone else, the so-called “standard approach to self-deception”. The account under investigation is Donald Davidson’s. I criticize Davidson’s outline of self-deception as involving contradictory beliefs, and his portrayal of self-deception as a rational and strategic action. I trace the assumptions involved in Davidson’s account back to his account of radical interpretation and argue that the problems and paradoxes that Davidson discusses are not inherent in self-deception as such but are problems arising in and out of his account. In Chapter Two, I present Sebastian Gardner’s account of self-deception. Gardner is concerned with distinguishing self-deception as a form of “ordinary” irrationality that shares the structure of normal, rational thinking and action in being manipulation of beliefs from forms of irrationality treated by psychoanalysis. I object to the way in which Gardner makes this distinction and further argue that Gardner is mistaken in finding support in Freud for his claim that self-deception involves preference. In Chapter Three, I present a different understanding of self-deception. I discuss self-deception in the context of Sigmund Freud’s writings on illusion, delusion, different kinds of knowledge, etc., and propose a view of self-deception where it is not seen as a lie to oneself but rather as motivated lack of self-knowledge and as a flight from anxiety. In Chapter Four, I discuss some problems inherent in the three accounts under investigation, for example, problems arising because first-person awareness is conflated with knowledge of objects.