<p>According to the so-called ‘functional view’, we have deception if and only if the thing whose function is to mislead misleads. An idea of this sort was first introduced by Fallis, then defended in more detail by Artiga and Paternotte, and finally refined and applied to both psychological (human) and biological deception and self-deception by me. I now argue that this is currently our best analysis of deception. My argument involves contrasting Artiga and Paternotte’s version of the functional view with their latest view that deception is an instance of mimicry. My reason for this selection is that only these two views can offer a unified analysis of all kinds of deception (including self-deception), they do not require the deceiver to benefit from deception, and they are not limited to deception involving signalling strategies. However, deception need not involve mimicry either. Therefore, the functional view should be our preferred analysis of deception.</p>