Ongoing advancements in machine learning are providing new avenues for the analysis of large-scale visual data, particularly in the realm of historical political economy. Researchers at the Massachusetts Institute of Technology (MIT), led by Valentine Figuroa, have developed a framework aimed at interpreting the rich information encoded in paintings from museums and private collections. This framework is crucial, as it allows scholars to assess what these artworks reveal about historical contexts and the assumptions underlying their interpretation.
The framework is founded on traditional humanities concerns and is illustrated through three distinct applications using a comprehensive database of 25,000 European paintings spanning from 1000 CE to the First World War. Each application focuses on a different aspect of the information conveyed in these artworks: depicted content, communicative intent, and incidental information. The research also highlights significant cultural transformations that occurred during the early-modern period.
Exploring the Civilizing Process through Art
One application revisits the concept of a European “civilizing process,” which refers to the internalization of stricter behavioral norms that coincided with the expansion of state power. By examining paintings that depict meals, researchers sought to determine whether there is evidence of increasingly complex etiquette over time. This analysis provides insights into social dynamics and the evolving standards of conduct within European societies.
Shifts in Political Imagery
The second application focuses on portraiture to explore how political elites constructed their public images. The study highlights a long-term shift in representation, moving from chivalric depictions to more rational-bureaucratic portrayals of men. This transition reflects broader societal changes and the growing influence of state institutions on personal identity and public perception.
The third application documents a long-term process of secularization, measured by the proportion of religious paintings produced. This trend began before the Reformation and accelerated in its aftermath, showcasing how artistic expressions mirrored the shifting religious landscape of Europe.
The findings presented in this research not only contribute to the understanding of political and cultural history but also demonstrate the potential of computational methods to uncover new narratives within traditional art forms. By establishing a systematic approach to analyzing visual data, Figuroa and his team open up exciting possibilities for interdisciplinary research that bridges art, history, and technology.
This innovative work underscores the importance of reevaluating historical artifacts through modern lenses, ultimately enriching our comprehension of the past. As machine learning continues to evolve, so too will the opportunities to unlock the stories hidden within art, providing a deeper understanding of the political landscapes that shaped historical societies.
