Doctor Science Knows

Monday, March 05, 2007

Left-Right differences in civility

There's now a civility discussion over at Slacktivist. I said:

There is one point on which I've found total agreement between Left & Right Blogistan: each side finds the other shockingly uncivil.

Logically, this means either: the Right is more uncivil, the Left is more uncivil, or both are vile pits of boiling incivility but you only notice the other guy's. I've wondered for quite some time if there could be a way of quantifying incivility, so that we could tell which is true. What would you measure and how would you measure it?

one response included:
I'd probably also measure their tendency to label or categorize the other as a tactic for avoiding the substance of what they say (i.e., replying to an argument with "yeah, but you're just a liberal/neo-con/bleeding heart/bigot/whatever").

to which I said:

Of the various dimensions you suggest, I think the easiest to measure in volume would be labeling: specifically, nick-naming. I've seen this on both Left & Right, so I'm sure there's something *to* measure, and I think it would be a good stand-in for a lot of other factors that are very difficult to determine objectively.

Patrick Ishmael tried to quantify online vulgarity using Google & George Carlin's list of Seven Dirty Words, and found the left to be much, much more vulgar. There are a lot of methodological problems, though, and I don't see that strict linguistic vulgarity has any particular correlation with civility.

I later said:

One interesting set of comparisons would be meta-discussions such as this one. Can anyone point to a similar discussion to this one in Right Blogolandia? Google leads me down a mazy of twisty passages all the same.

Francis had an excellent reply to my "right? left? both?" comment:
Option d: the two groups mean different things by civility. The right wing blogs I've been to have invariably been polite to anyone who is there who is not directly insulting - whereas many of the left wing ones are quite prepared to tell people that they are being idiots. On the other hand, when the insults start flying on the liberal blogs, they will normally be justified and with evidence backing them up whereas the right wing appear to pull things out of thin air far more - but only do this once a more considerable reserve of politeness has been broken.

Also, there is a different approach to attacking groups - when an authoritarian (and here rather than the previous point about discourse is where the authoritarian liberal divide appears), they reflexively assume that any person who is there is not part of the group they are attacking (or they wouldn't be there in the first place or are one of the few "exceptions that prove the rule" (I detest that perversion of a phrase)) whereas when a liberal attacks a group it is much more careful because there is an underlying assumption that anyone who claims to be a member of the group actually is and that there are sensible people who are - and hence liberals tend to pick out specific individual targets (which is a right-wing breach of civilised conduct (see the first half of this comment)) to avoid splash damage - but by doing so, the right wing perceives this as personal.

Francis' observations seem to say that conservative (or at least authoritarian) blogs as have firm group borders, while liberal blogs have permeable group borders. This strikes me as extremely plausible and well worth further research.

Which definition of civility do the libertarian blogs follow, do you think?



  • There were 3 main reasons that the survey was wrong.

    1. The liberal blogs Ishmael tallied had over 10 times as many web pages as the conservative blogs.

    2. A bug in Google caused it to claim that huffingtonpost has almost twice as many pages with foul language, as it has pages - and it has about twice as many pages as all of the conservative websites in the survey combined.

    3. Another bug in Google made in unable to count obscenities in littlegreenfootballs, which accounted for 1/3 of all the conservative webpages.

    The corrected results show liberals cursing twice as often. I don't think this is reliable, though, because Google's results are unreliable, and the small sample size means a single additional error by Google could easily account for the difference. Details, and (profane) source code to conduct the same survey yourself, are on my livejournal blog.

    I explained the problems to Ishmael, but he chose neither to respond, nor to post any of the information on his page.

    By Blogger shagbark, at 1:12 PM  

Post a Comment

<< Home