Recent advancements in machine learning are paving the way for new analyses of historical political data, particularly through visual art. Research led by Valentine Figuroa from MIT explores how paintings from museums and private collections can serve as valuable, yet underutilized, sources of information in the field of historical political economy. This research emphasizes the necessity of developing a comprehensive framework to assess the information encoded in paintings before applying computational methods to analyze them.
The study draws upon a vast database of 25,000 European paintings spanning from 1000 CE to the First World War. Figuroa’s framework categorizes the information conveyed in these artworks into three distinct applications: depicted content, communicative intent, and incidental information. Each application targets specific cultural transformations that occurred during the early-modern period.
Examining the Civilizing Process
One of the key applications investigates the concept of a European “civilizing process,” which refers to the internalization of stricter behavioral norms alongside the expansion of state power. By analyzing paintings of communal meals, the research aims to determine whether there is a noticeable increase in the complexity of etiquette depicted in these artworks over time. This examination could provide new insights into how societal norms evolved alongside political authority.
Political Portraits and Public Image
Another application focuses on portraits to understand how political elites crafted their public personas. This analysis highlights a significant long-term shift in representation, moving from traditional chivalric depictions to more rational and bureaucratic portrayals of men. By studying these changes in artistic expression, researchers can gauge the evolving dynamics of power and identity among political figures.
The third application tracks the long-term process of secularization within European art. This aspect measures the proportion of religious paintings produced over the centuries, revealing a trend that began prior to the Reformation and intensified thereafter. Such findings could reshape our understanding of the cultural landscape during this pivotal historical period.
Overall, Figuroa’s research presents a novel intersection of art and data analysis, demonstrating how machine learning can illuminate aspects of political history that were previously obscured. As scholars continue to explore the potential of visual data, the implications for both historical understanding and contemporary political analysis remain significant.
