Pairwise comparisons are an important tool of modern (multiple criteria) decision making and form a fundamental part of the analytic hierarchy/network process (AHP/ANP), ELECTRE, PROMETHEE and other sophisticated methods. One of the most important features of logically consistent preferences is their transitivity. The aim of this paper is to compare occurrence and violation of transitivity of preferences in pairwise comparisons for pairwise comparisons matrices (PCMs) provided by human experts and PCMs generated randomly (Monte Carlo method) with the use of Saaty's fundamental scale. Our results suggest that the occurrence of transitivity is more frequent for humanmade PCMs, while transitivity violation is more frequent for random PCMs. This indicates that transitivity analysis can be used for discrimination between human-made and random PCMs.