Humans tend to be optimistic by nature. For proof of this just look at an average weeks lottery ticket sales. In 1969 two University of Illinois psychologists proposed the “Pollyanna Hypothesis” which suggests that this general positivity extends to human language. They suggested that there is a universal tendency to use positive rather than negative words in communication.

Recently, researchers at the University of Vermont and the MITRE Corporation employed big data to put the Pollyanna Hypothesis to the test.

The team gathered billions of words from movie subtitles, twitter feeds, literature, websites, music lyrics and newspapers in English, Arabic, Chinese, Korean, Russian and a variety of other languages. Overall they found that all of these languages, and probably all of human languages, skews in favor of the positive.

“We looked at ten languages and in every source we looked at, people use more positive words than negative ones,” said University of Vermont mathematician Peter Dodds who co-led the study, in a statement.

The researchers found that even the torrential cursing on sites like Twitter and the endless bad-news “if it bleeds it leads” stories in the news did not have an impact.
According to Dodds the language itself has a positive outlook and it appears that positive social interaction is built into the fundamental structure of language.

This is supported by a Association for Psychological Science (APS) article published in 2012. The authors, Timothy Jay and Kristin Janschewitz, do not find the negative associations with swearing that people might expect.

“Swearing can occur with any emotion and yield positive or negative outcomes. Our work so far suggests that most uses of swear words are not problematic,” say the authors.

Using the University of Vermont’s Computational Story Lab, the researchers of the Pollyanna study gathered words from twenty-four types of sources, including more than one-hundred billion words from Twitter.

Next the team identified the ten thousand most common words in each of 10 languages including English, Spanish, French, German, Brazilian Portuguese, Korean, Chinese (simplified), Russian, Indonesian and Arabic. They paid native speakers to rate these words from very negative (1) to very positive (9). So, for example, in English “laughter” rated an average 8.5 while “terrorist” rated a 1.3 and “the” averaged 4.98.

Finally, the researchers crunched the numbers. The most positive results came from a Google web crawl of Spanish-language websites and the most negative from an analysis of Chinese books but in every language, from every source the words skewed positive. In other words they all rate above the median of five or “a usage-invariant positivity bias” as they call it in the study.

The resulting study “Human Language Reveals a Universal Positivity Bias,” appears in the February 9 online edition of the Proceedings of the National Academy of Sciences.

The results of the positivity in language may be a result of what is caused the “optimism bias”. People, even if they understand that bad things are happening in the world, are unlikely to think that they bad things will happen to them and will generally believe that things will get better.

“The belief that the future will be much better than the past and present is known as the optimism bias. It abides in every race, region and socioeconomic bracket. Schoolchildren playing when-I-grow-up are rampant optimists, but so are grownups: a 2005 study found that adults over 60 are just as likely to see the glass half full as young adults,” said Tali Sharot of Time in May, 2011.

The University of Vermont study is part of a larger project to create a “hedonometer” or happiness meter. According to the researchers it can now be used to trace English-language twitter posts on a near-real-time basis. Other languages and sources of data are expected to follow shortly.

The Hedonometer can be viewed at hedonometer.org.