Deception is an important concept in philosophy, psychology, *and*  biology.  When cognitively unsophisticated species, such as fireflies and butterflies, have evolved to mislead potential predators and prey, it is no accident that they are misled.  However, the traditional analysis of the concept, which requires that a deceiver*intentionally*  cause her victim to have a*false belief*, rules out the possibility of such "functional deception."  Following Brian Skyrms (2010), Fallis and Lewis argue that deception only requires that a deceiver systematically benefit from putting her victim into a worse epistemic state.  Unfortunately, Skyrms's specific analysis along these lines in the context of Lewisian sender-receiver games is seriously flawed.  In this talk, the researchers suggest how to correct it.  In addition, they will show that two recent alternative analyses (Godfrey-Smith 2011 and McWhirter forthcoming) are unsuccessful.