One way to judge whether sensations are merely part of the causal order and not part of the cognitive or epistemic order is to determine whether or not sensations control to any extent the meaning of our observation terms. Should our observation terms have their meanings even in part determined by sensations then this would seem to be evidence that sensations are of the cognitive order. In a recent and noteworthy book, Scientific Realism and the Plasticity of Mind, Paul Churchland offers a development of the Sellarsian idea that sensations play a merely causal role in perception. Churchland also cites Paul Feyerabend, whose adaptation of the Sellarsian position he finds more to his purpose. The central argument of Churchland's book and the one which will be most particularly dealt with herein is an attempt to show that facts about sensations are totally irrelevant to the meaning of observation terms, even to the meaning of common observation terms such as 'hot,' 'cold,' 'white,' and 'black.' It is Churchland's contention, then, that facts about the intrinsic nature of sensations (as opposed to facts about their roles in causal chains) are semantically irrelevant.