Modeling Linguistic Changes in Propaganda Across USSR, China, and Present

Session Number

3

Advisor(s)

Dr. Patrick Buck, IMSA

Location

B115

Discipline

History

Start Date

15-4-2026 2:15 PM

End Date

15-4-2026 3:00 PM

Abstract

In the Soviet Union and the People's Republic of China, communist state propaganda was historically disseminated through centralized, state-owned newspapers. Recently, algorithm-based, short video platforms such as TikTok and its Chinese counterpart, Douyin, have transformed the speed and personalization of propaganda distribution. This study investigates ideological linguistic changes in historical print-based and modern video content propaganda using computational methods. Over 35,000 English-translated digital works by Karl Marx, Vladimir Lenin, Joseph Stalin, Mao-era Peking Review, and transcribed Douyin videos were analyzed for common political terms such as “party”, “worker”, “peasant”, and “revolution.” Results show significant vocabulary changes across authors and time periods, relating to class, national, and party ideology. Consistently, “party” was one of the most frequently used words, definitionally differing with class struggle under Marx, organizational structure under Lenin, centralized authority under Stalin, and collectivist leadership under Mao. Contemporarily, CCP news accounts for roughly half of sampled Douyin videos, with 8,910 labeled as government news (propaganda, business, or other). Popular categories (>40 videos) include state achievement announcements, foreign policy, and crisis responses. These findings show the progression of linguistic patterns in propaganda, where categories/words remain similar across eras, but their uses/definitions differ.

Share

COinS
 
Apr 15th, 2:15 PM Apr 15th, 3:00 PM

Modeling Linguistic Changes in Propaganda Across USSR, China, and Present

B115

In the Soviet Union and the People's Republic of China, communist state propaganda was historically disseminated through centralized, state-owned newspapers. Recently, algorithm-based, short video platforms such as TikTok and its Chinese counterpart, Douyin, have transformed the speed and personalization of propaganda distribution. This study investigates ideological linguistic changes in historical print-based and modern video content propaganda using computational methods. Over 35,000 English-translated digital works by Karl Marx, Vladimir Lenin, Joseph Stalin, Mao-era Peking Review, and transcribed Douyin videos were analyzed for common political terms such as “party”, “worker”, “peasant”, and “revolution.” Results show significant vocabulary changes across authors and time periods, relating to class, national, and party ideology. Consistently, “party” was one of the most frequently used words, definitionally differing with class struggle under Marx, organizational structure under Lenin, centralized authority under Stalin, and collectivist leadership under Mao. Contemporarily, CCP news accounts for roughly half of sampled Douyin videos, with 8,910 labeled as government news (propaganda, business, or other). Popular categories (>40 videos) include state achievement announcements, foreign policy, and crisis responses. These findings show the progression of linguistic patterns in propaganda, where categories/words remain similar across eras, but their uses/definitions differ.