It is a commonly prevailing persuasion that a theory of education is a quite autonomous sub-discipline of pedagogical science. Meanwhile, it has more and more increasing problems with defining its own identity. A different way of teaching it as a learning course at the universities confirms it as well. Certainly, for long time now, we have been dealing with the ambiguous understanding of the theory of education, with its both methodological and ontological weaknesses. Such ambiguity arises not only out of the various orientations, systems or concepts of directions of pedagogical thinking, but mostly out of the ambiguity of terms 'theory' and 'education', which jointly give name to this field of knowledge. It can be assumed that the term 'theory' should be understood as a set of inter-related statements related to an explored fragment of reality, which are strictly, coherently and consistently interconnected. Search for the definition of the term 'education' can be based on an explication method. It allows for acknowledging the education as a specific social entity which is a creation of at least two people, and which characteristic feature lies in the relationship occurring between them, in which an educator, in other words, a teaching person, who is guided by the universally acknowledged good, enables a learner to get to the subsequent maturity stages expressed by reaching self-identity. The education which has been understood in such a way takes into consideration its basic properties: continuity, dynamics and long duration. From the point of view of clarity and usefulness, in the scope of the theory of education, two areas can be singled out: teleology of education and technology of education. They provide basis for the definition of its most important functions: ideographical and diagnostic, analytical and prognostic, ordinary and intentional, communicative and complex, utilitarian and formative, heuristic and cognitive ones. Out of the above it arises that the theory of education is a pluralistic sub-discipline and to a small degree, it is conclusive.