Unknown Knowledge: A Competence That Is Harder to Recall Than Receptive Vocabulary

Specialist research can produce results without memorised knowledge of the field. This short communication examines that possibility through a single case: an independent researcher with a bachelor's degree, an outsider to every field outside English education and positive behavior support, who entered artificial intelligence research with no prior knowledge and had fourteen papers accepted within six months, including acceptances after editorial review. In almost every paper, the researcher anticipated that circumstantial evidence existed before searching for it, and every paper was written on the basis of the search results. In one recorded conversation, ten such predictions were checked: three had already been argued, six had closely related research, and one found nothing. The paper proposes the term unknown knowledge for knowledge that is harder to recall than receptive vocabulary, let alone productive vocabulary, and yet can become a critical competence. Its paradox is that it appears less firmly retained than vocabulary while the understanding of its structure is retained more firmly than verbal knowledge. The paper relates this proposal to work on expertise, analogy, tacit knowing and powerful knowledge, and draws a consequence for assessment: tests that require named terms can miss the competence that free hypothesis formation reveals.