Indian cinema is gradually abandoning a set of gender and social prejudices behind it, but resumes to be fixated upon the idea of ‘fair skin’ as the fundamental explanation of beauty, large-scale research of 700 Bollywood movies bridging the last 70 years, has found.
Conducted by researchers from the Carnegie Mellon University in the U.S., the research points out that “popular movie content reflects social norms and beliefs in some form or shape,” and so, the researchers wanted to examine the differences in cultural standards through the lens of movies that audiences have been consuming over the prior seven decades.
The team compiled 100 Bollywood movies stretching 70 years and juxtaposed them against 100 top-grossing Hollywood films made during a similar period. They then applied Natural Language Processing (NLP) techniques and analyzed subtitles of 1,400 films for gender and social prejudices in their statistical dialect models. And ultimately utilized artificial intelligence (AI) to make the observance.
They utilized fill-in-the-blank training to evaluate specific assertions like “A beautiful woman should have _____ skin”. The model anticipated ‘fair’ after being equipped on Bollywood subtitles data.
The prejudice was less evident when the AI model was utilized on Hollywood subtitles, the team reported.
According to Tom Mitchell, a computer scientist, and professor at the CMU, who co-authored the study, “The study gives us a finer probe for understanding the cultural themes implicit in these films”.
Bollywood’s preoccupation with fair skin is reflective of the colorism prevalent in the nation, driving people to face racism in schools, at workplaces, and in India’sarranged marriage market’ because of their skin complexion.
To evaluate the preponderance of male characters, the researchers utilized a metric named Male Pronoun Ratio (MPR), which distinguishes the occurrence of male pronouns such as “he” and “him” with the total occurrences of male and female pronouns.
From 1950 through today, the MPR for Bollywood and Hollywood films varied from roughly 60 to 65 MPR.
Ashique KhudaBukhsh, a project scientist at the School of Computer Science in CMU, who co-authored the research stated, “ Monopolized by fair-skinned actors, and still submitting in practices like ‘brown face’, Bollywood has reflected societal prejudices for years, and “now we have numbers to quantify them And we can also see the progress over the last 70 years as these biases have been reduced”.
As KhudaBaksh noted, the study did examine many optimistic trends as well.
Dowry, reported to be behind the deaths of 21 women every day, was found to be more socially reasonable in the 50s and 60s, approximated to the present.
In accumulation, the birth of a child, depicted by the study as “a popular Bollywood plot point,” was found to be more sex-balanced now: in the past, infants were often male; now they’re more evenly split between male and female, the research proved.
The research also found that religious proportion in movies, at present, fits the real-life religious ratio among the Indian population, with the representation of non-Hindu religions having heightened gradually over the years. At the same time, “the representation of Muslims is slightly less than the community’s population share,” the research notes.
In terms of regions depicted by India’s mainstream movie industry, the study found the current subject is geographically more extensive, and not as focused in cities like Mumbai and Delhi — but areas like the northeast remain least represented.
The study shows that NLP can uncover how popular culture reflects social biases. The next step could be using the tech to show how popular culture influences those biases.