Anyone who thought that social media was going to usher in a utopian era of communication without borders is going to have their faith badly shaken by a new study from Cornell University. On Twitter, it seems, there are already at least two walled-off nations: happy people and unhappy people.

And never the twain shall tweet.

The Cornell study, spotted by New Scientist magazine, tracked 102,000 Twitter users and analyzed the 129 million tweets they sent and received during a six-month period. It examined the words they used using what the authors call “standard techniques from psychology” to rate their sense of self-fulfillment — an important measure, in the burgeoning field of happiness studies, known as subjective well-being or SWB. People with a high SWB were significantly less likely to send or receive tweets from someone with a low SWB, and the same was true in reverse.

In other words, the happy people have formed little Happy Twitter clubs. Meanwhile misery loves company in social media — as much as, if not more than, in real life.

“Beyond demographic features such as age, sex and race, even psychological states such as “loneliness” can be assortative in a social network,” writes the study’s lead author, Johan Bollen. But he admits that even he doesn’t know why that should be the case.

So why is it so? Is it simply human nature, or a function peculiar to short-form virtual communication?

The answer depends on who you ask. Users on the “happy” side of the SWB equation will probably tell you they meant to respond to that depressive tweet from their Debbie Downer friend, but it was such a bummer they couldn’t be bothered. Unhappy tweeters will grumble about the insufferably peppy quality of tweets from those shiny cheerleaders over in the happy corner.

If there are any Twitter API developers looking for a Nobel Prize-worthy challenge, now would be the time to devise an app that could bring the two factions together in a state of semi-contentedness.