It’s very confusing that this Journal of …

It’s very confusing that this Journal of Communication study (via Tim Chambers) on the effect of Twitter’s expansion from 140 characters to 280 claims both that the change “increased the prevalence of less uncivil, more formal and more constructive messages” and “decrease[d] … the empathy and respectfulness of messages”. How do conversations simultaneously become both more civil and yet less respectful?

(I’m sort of setting aside for the moment the whole thing where civility is used a cudgel against the voices of people from marginalized or ignored groups, and the fact that therefore conversations can be simultaneously more civil and yet less repectful, because I don’t think that’s actually addressed as a dynamic in this paper.)

“Uncivil behavior,” the authors explain, “was measured in terms of the use of name-calling, profanity, hate speech or invocation of stereotypes of a homophobic, racist, sexist or xenophobic nature.” Empathy (and respect), they write, “[r]eflects the author’s acknowledgment of or sensitivity to others, manifested in positive comments, an empathetic or a respectful response acknowledging other viewpoints”.

While Twitter’s implementation of a character limit change to double the length of tweets led to less uncivil political discussions and more deliberative political discussions, it also decreased empathy and respect among the discussants. These results highlight the importance of the careful design of discussion forums for the quality of political discussion.

[…]

[…] While the character limit improved overall civility in political discussion, the decrease in empathy and respect among compliers is a cause for concern. That is to say, while people were more polite after the character limit change, they were less likely to be empathetic or respectful to other people’s comments.

Somehow, the authors found that the increase in characters led to a decrease in abusive language but also a decrease in sensitivity to others, which is baffling to me, and even having read the entire paper I can’t seem to find (or maybe understand) an explanation for the how or the why of this.