We have shown that propositionality is, in none of its three main senses, an adequate and useful demarcation criterion between knowledge-how and knowledge-that.
First, in its semantic sense (e.g., Stanley 2011a), propositionality applies to both knowledge-how and knowledge-that, and thus a fortiori cannot be successfully used as a demarcation criterion.
Second, in its “language of mind”-sense, propositionality applies to knowledge representation. As we have shown, the way in which a particular piece of knowledge is represented is independent from the type of knowledge exemplified by this piece of knowledge. Thus, again, this sense of propositionality is not useful as a demarcation criterion.
Third, propositionality in the sense of linguistic, consciously available propositions is without doubt a central phenomenological trait of knowledge-that as opposed to knowledge-how. On the one hand, this sort of propositionality offers a rather trivial demarcation criterion. On the other hand, as a mere replication of a well-known phenomenological distinction, it can in no way be used to explain the different peculiarities characteristic of knowledge-how versus knowledge-that. Anti-intellectualists have tried to fill the void corresponding to non-propositionality, according to this third sense of propositionality, by declaring specific knowledge formats such as sensorimotor or image-like knowledge (Newen & Jung 2011). In our view, it is doubtful whether, with such an eclectic way of characterizing knowledge-how, a satisfactory and complete classification of knowledge-how could be achieved. We have, for example, argued that “intuitive” knowledge would be a further legitimate candidate for the list, and that it is, in all probability, not the only further candidate. Identifying different forms of knowledge-how without any well-grounded theoretical basis for the different forms will probably be of limited use for empirical research in cognitive science, neuroscience, and psychology.
In sum, “propositionality” can in none of its different senses provide a useful demarcation criterion for an empirically-fruitful theory of knowledge-how. Therefore, we go with the intellectualists, at least with respect to rejecting the propositionality criterion, but we depart where intellectualists fail to provide positive accounts of the obvious phenomenological and empirical peculiarities making knowledge-how distinct from knowledge-that. In contrast to the intellectualist position, we have provided a minimal notion of conceptuality as an alternative demarcation criterion. We suggest that conceptuality gives a sound basis for a fruitful theory of knowledge-how, and we have tried to provide support to this suggestion by showing that by means of an adequate notion of conceptuality, five central peculiarities of knowledge-how as compared to knowledge-that can be accounted for. Future research will have to show whether the framework for practical knowledge described here fulfills the empirical promise we think it has.