The aim of this article is to challenge a time-honoured myth which claims that the Aristotelian demarcation of the mutual relations between particular types of categorical judgements, as represented in the “square of opposition”, and also a part of Aristotelian syllogistics, are not valid from the point of view of modern logic without the assumption of the non-emptiness of the concepts under consideration. In reality, however, the Aristotelian tradition works with empty concepts and the square of opposition is valid for those concepts. The only problem is with the inadequate formal transcription of categorical judgements which modern logic usually offers: it assumes, that is, an “existential import” (if the judgement is true then there must exist an instance of the subject’s concept) in partial judgements, while the Aristotelian logic assumes it in positive judgements. On this Aristotelian assumption, the relations of the square of opposition, and syllogistics in general, function without difficulties, even for empty concepts. This had been quite explicitly formulated by medieval logicians. In the following part of the article, some traditionally controversial expressions of Aristotle concerning just this question of existential import are analysed. Some of these controversies probably stem from the confusion of two different problems – on the one hand, the question of the mutual logical relations of the categorical judgements under analysis, on the other hand the problem of the possibility of a truthful account of non-existent objects. To this second problem the Aristotelian tradition takes a different approach than modern logic, which considers existence as a second-order property and identifies it with presence in the universe.