The article focuses on the systemic effects of censorship that manifest themselves in the content of published materials that successfully passed the censorship filters. We understand censorship as a special kind of collective imagination about the (in)acceptable, inherent in a particular political context and influencing the decision-making logic by different actors. The idea is that censorship affects the ability of the authors to navigate the topical space, so that juxtaposition of certain topics (e.g. literature and politics) is specifically avoided. To detect this effect, we suggest an idea of topical dissociation, operationalized as a probability that either one or the other topic appear in the same article, but not both. We apply LDA topic modeling to the corpus of Russian literary magazine Sovremennik (1847--1866) to trace topic dissociation across the period. We hypothesize how the strength of topical dissociation should change with respect to the historical data on the changing censorship practices of the period. Empirical data only partially supported our hypotheses. The method has a potential for wider application to study censorship effects on the published materials.