In this paper I explore the tension that arises when we describe ourselves as intentional beings, on one side, and biological or physical beings on the other. That is, I take a closer look at the age old problem of how the mind relates to the brain. In order to get a better grip on the tension between mental and physical descriptions I start of the discussion with a closer look at what it means to explain the same object, the brain, as two different kinds of thing. The pragmatic kind of naturalism endorsed in this paper draws on the idea that different vocabularies instantiate different interests and commitments. The vocabulary-vocabulary as Brandom has dubbed it allows you to talk about mental properties and physical properties without assuming the ontological primacy of the one over the other. Since the idea of vocabularies is significant to this paper I devote some time to explaining the central idea behind it and lay down some of it central virtues and spell out how it stands in relation to more traditional approaches to the mind-body problem. As the idea is elaborated it will be clear that it helps to fend off some major metaphysical and ontological problems inherent in the more orthodox approaches to the mind. I then proceed to evaluate a position that is intended to remove the tension by arguing that the intentional vocabulary is widely erroneous. This is the eliminative materialism of Paul and Patricia Churchland. They claim that Folk Psychology fail to meet up with the current findings of neuroscience and consequently should be eliminated. I assess their position by first stating what they mean by eliminativism and then examine five different objections against their position. The reason for examining these objections is to determine what the opposing sides in the debate believe are the important elements in the commonsensical and in the scientific description of our mental lives. In essence the Churchlands claim that the commonsensical framework of the mental will have to adapt according to the findings of neuroscience. The purpose of scientific explanation will be discussed further in the final part of the paper. For this purpose I turn to Paul Griffiths’ new and more flexible account of natural kinds. Griffiths thinks that the traditional view on natural kinds, with universal exceptionless laws, is too rigid to be of much help to science. He wants to leave the idea of natural kinds as the most fundamental category of nature and instead view them as non-arbitrary ways of classifying the subject matter under examination. He uses his findings to argue that the vernacular concept of emotion does not form a single natural kind and concludes that it has to be eliminated in favor of a better informed theory of emotion. I then compare and contrast the different reasons the Churchlands and Griffiths hold for arriving at the eliminativist conclusion. I take a look at the normative role the intentional vocabulary plays, since this is something they seem to disagree about.