This article is aimed at revealing a kind of methodological narrowness in the descriptions of social networks in Soviet society. In sociological research, the term “social network” usually refers to a complex of close and distant acquaintances that connect people in a certain geographical locality, and/or in a certain social cluster. However, if regarding the studies of Soviet and post-Soviet society, sociologists and political scientists usually use this term referring to the networks of corruption, coat-tails, a shady economy, etc., that “corrode” modern social and political institutes, hindering modernization. These can also refer to the secret networks of samizdat and dissidence in the 1960–70s. All of these networks can be defined as “anti-disciplinary”. Apparently, in both the late Soviet and post-Soviet societies, networks of different types existed, and examples of these are described in the article. Nevertheless, all other networks aside from the “anti-disciplinary” ones are studied significantly less, and were never studied systematically. There are at least two types of causes that have entailed this “blind spot”. The “hidden” networks in the late Soviet and post-Soviet societies were and are of great political significance and of a specific organization; thus, they are considered as structures unavoidable for the understanding of Russia’s social dynamics and power relations. Secondly, Soviet society was based on a very complicated combination of modernizing and “archaizing” trends. This tension can be designated as “conservative modernization”, as coined by the demographer Anatoly Vishnevsky. A study of the interrelation between “pro-disciplinary” and “anti-disciplinary” networks demands a reconsideration of some the basic theses of the theory of social networks. Here, I use Harrison White’s theory in order to demonstrate the limits of its applicability to late Soviet society.