Human Language Is Biased Towards Happiness, Say Computational Linguists | Hacker News Reader

Human Language Is Biased Towards Happiness, Say Computational Linguists

medium.com

35 pointssytelus13 comments

Doesn't this only suggest that most people are biased towards happy "personal interpretations" of words? That's a conclusion about people, not "human language," whatever that is.

there's a huge Sapir-Whorf bias in a lot of philosophy, critical theory and social science which I totally don't get.

Interesting how the thesis of the article is the exact opposite of the thesis of this one, which was posted to HN a few days ago:

http://aeon.co/magazine/psychology/humans-are-wired-for-nega...

That article talks about vocabulary. One would naturally assume that vocabulary correlates with frequency of usage, but that may not be the case.

This article talks about frequency of usage. Perhaps someone can dig deeper and figure out where they get their frequency of usage data from?

If they are scanning novels and lyrics in various languages, there is also an interpretation that musicians and writers have in general written more positive words in their work. This can be part of art instead of something inherit to language.

If you're really confused, the aeon one is the better one scientifically, and reviews a much larger array of science from a number of domains, rather than scanning google books.

Peter Dodds, the leader of the team in question, isn't a computational linguist. I just read his CV and he has zero publications in CL journals or proceedings, and his Ph.D. is in mathematics.

Computational Linguists are Biased Toward Computers, Say Happy Humans

If you google search positive term dictionaries vs negative term dictionaries you'll find the negative term dictionary is 5x - 10x larger.

Point being, we have many more terms to describe negative events. Likely, this means we have a large biased towards negativity.

As Kronopath mentioned, there is definitely research out there describing the exact opposite of this article.

I don't think you can backsolve for Sapir-Whorf.

So very interesting!

I'm sure they got sarcasm right, and all the cases where humans damn with faint praise.

A gold star for everyone involved!

Four days ago we saw the submission "Humans are wired for negativity":

https://news.ycombinator.com/item?id=8275688

But language is biased toward happiness but its users are wired for negativity.

What does that mean? We are negative, but through language, are trying to kid ourselves otherwise?

This submission is about counting words on Google Books and other online sources.

The other one reviews a few decades of research in psychology.

I would say that books that are happier sell more, so there are more of them written.

Reply on news.ycombinator.com