Does language work to make the world a happier place?

Analysis of books, films and media suggests that humans tend to use more happy words than sad ones. Katherine Kizilos reports.

People were asked to rate about 10,000 words on a nine-point scale from saddest to happiest. Researchers then analysed how often they were used in literature and social media. – istock

In fiction, Pollyanna was a young girl who tried to cheer everybody up by urging them to look on the bright side of life. It turns out she was simply encouraging a natural human tendency. Recent research published in the Proceedings of the National Academy of Sciences, shows that human language has a “universal positivity bias”.

The idea that we humans are essentially a positive bunch is not new.

In 1978 psychologists Margaret Matlin and David Stang found that people tended to remember positive experiences more vividly than sad or painful ones. They dubbed their finding “the Pollyanna principle”.

But the latest research suggests language itself helps create that frame of mind.

Peter Dodds and Chris Danforth at the Computational Story Lab at the University of Vermont led the research that looked at the most commonly used words in English, Spanish, French, German, Brazilian, Portuguese, Korean, Chinese, Russian, Indonesian and Arabic. Native speakers from each language were asked to rate their responses to about 10,000 words on a nine-point scale (one being the saddest and nine the happiest). A computer program then analysed the rated words as they appeared on social media, in web searches, The New York Times, Google books (including Melville’s Moby Dick, Dumas’s The Count of Monte Cristo and Dostoyevsky’s Crime and Punishment) and film scripts.

Their analysis found that, to quote the old song, humans naturally tend to accentuate the positive, although there are slight cultural variations. Spanish speakers in Mexico and Portuguese speakers in Brazil were found to be the most positive overall, while the Chinese were at the dour end of the spectrum, or, as the academics put it, the “most constrained”. English speakers were the third most positive in the group.

The hedonometer found a dip in happiness in France after the Charlie Hebdo terror attack, and a spike in Mexican happiness over the Christmas holidays.

The latest research builds on work begun in 2009, when the same team constructed a hedonometer, an online instrument that its creators claim can “measure the happiness of large populations in real time”. It chiefly does this by analysing the number of positive words in Twitter feeds. Recently the hedonometer found a dip in happiness in France after the Charlie Hebdo terror attack, and a spike in Mexican happiness over the Christmas holidays.

The current research “takes a similar sort of idea and looks at the (emotional) temperature of a large body of work,” says Lewis Mitchell, a mathematician at the University of Adelaide who worked on the hedonometer and also took part in the latest research. He says it shows we are “inherently positive creatures” – although whether our language is the cause is hard to say.

But what excites him most about the research is that it gives mathematicians the opportunity to apply their skills to areas that were once the domain of linguists, psychologists and social scientists. Mitchell says the analysis of big data gives mathematicians the opportunity “to measure what was previously unmeasurable”.

It also raises intriguing questions about how we frame the stories we share about the human experience. For his part in the language project, Mitchell downloaded 1,000 English language film scripts from the internet to see if he could devise a simple algorithm by which it might be possible to identify the emotional signatures of basic story types.

The idea isn’t new. Literary theorists, psychologists and script-writing manuals have all argued that human stories follow predictable patterns. An inspiration for Mitchell’s story research was a brief talk by the US author Kurt Vonnegut on the shapes of stories, in which the author draws a graph of the biggest story of all, which he summarises as “man falls in a hole”. Bad or unexpected events can overtake our protagonist, but experience and resilience can redeem him (he climbs out of the hole).

Which brings us back to the positivity bias in language. The big story sees man falling into a hole as an opportunity – a set up that calls forth the adventure and happy ending.

Professor Hugh Craig, the director of the Centre for Linguistic and Literary Computing at the University of Newcastle says the research is “impressive” and “something quite new”.

“When you start to think about it, it does make a lot of sense,” he says. “When you ask people how they are, they tend to say ‘fine’ rather than ‘terrible’ … people do get rewarded for being optimistic and positive.”

Katherine Kizilos is a staff writer at Cosmos.