'Doomed' from the Start: The Role of 'Cultural Pollutants' in the 'Cultural Death' of Late Victorian Bestsellers
Image by Ivana Divišová from Pixabay
Abstract
In this presentation, I test the general hypothesis that a significant factor in the “cultural death” of late Victorian bestsellers is the presence of “cultural pollutants” (retrospectively, a better term would have been “transmission blockers”) in their textual fabric. “Cultural pollutants” are inherent properties of the text linked to the zeitgeist of a particular period that block its vertical cultural transmission across time to new generations of readers — they neatly fit the cultural tastes of a certain period or generation but prevent the novel to thrive when these tastes change. As such, they are direct opposites to Dan Sperber’s “cultural attractors”, aspects of the cultural product that facilitate its transmission. Textual features of the texts of the novels are “operationalized” as proxies for three types of “cultural pollutants”. The first one, linguistic complexity, is measured through the token-type ratio of the text — the initial hypothesis is that a higher TTR would be correlated with lower success. The second, named entity ratio, indicates the level of saturation with concepts inextricably linked to the sociocultural context of the novels (i.e. references to places, people, professions that are no longer relevant or no longer exist) — the hypothesis is that a higher NER would correlate with lower success. The final one, dialogic liveliness, is a basic measure of the proportion of dialogues in the texts (the number of utterances divided by the number of words) — the hypothesis is that a lower dialogic liveliness would correlate with lower success. The results reject the first hypothesis as TTR is positively correlated with success, possibly indicative of the high “literariness” of the “canon”. The second and third hypotheses are confirmed, but the effects sizes are quite small. Although a step in the right direction, this paper is far from the final word of the topic. The corpus is quite limited, which affects the statistical significance of the study. Instead of just focusing on bestsellers and the most “canonical” novels from the period, future research will need to include the “archive”, i.e. books that fall outside of these two narrow categories.
Presentation slides available upon request! I am currently trying to rerun some of the analyses of this paper using additional data and fit it in my dissertation. I am open for any suggestions, especially for other “cultural pollutants” or “transmission blockers”, and how these can be “operationalized”.
