A fresh look at the left and right political blogospheres

by Eszter Hargittai on April 28, 2010

It’s exciting to see a paper about blogs across the political spectrum that goes beyond the by-now rather common practice of looking at who talks to whom among bloggers (e.g., whether there are any cross-ideological conversations going on). Yochai Benkler, Aaron Shaw and Victoria Stodden of Harvard’s Berkman Center for Internet & Society have just released “A Tale of Two Blogospheres: Discursive Practices on the Left and Right” showing some significant differences in types of blog platforms used (with different affordances), co-authorships and levels of participation among blogs of different political persuasions. Here is one example of specific findings (based on analyses of 155 top political blogs):

Over 40% of blogs on the left adopt platforms with enhanced user participation features. Only about 13% of blogs on the right do so. While there is substantial overlap, and comments of some level of visibility are used in the vast majority of blogs on both sides of the political divide, the left adopts enabling technologies that make user-generated diaries and blogs more central to the site to a significantly greater degree than does the right. (p. 22.)

There are lots of other interesting results in the paper so I highly recommend reading it [pdf].
It’s very clearly written and summarizes related literature well so in case this is not an area you’ve been following, this is a good piece with which to start to familiarize yourself with related debates. If it is an area that you’ve been following then this is a must-read to see some truly original contributions to the literature.

For more on this elsewhere, Ari Melber has an interview with Yochai Benkler on this research in The Nation.



harry b 04.28.10 at 1:50 pm

Completely frivolous comment, but I was completely misled by the still-to-me-weird use of blue for left and red for right.


Barry 04.28.10 at 1:59 pm

Harry b, this is what’s been used on TV and in public discourse in the USA for – twenty years now?


Gareth Rees 04.28.10 at 2:14 pm

twenty years now?

Ten years, according to Wikipedia, which adds, “This unofficial system of political colors used in the United States is the reverse of that in most other long-established democracies”, hence harry b’s confusion.


kid bitzer 04.28.10 at 2:26 pm

does this new study do anything to distinguish respectful links from point-and-laugh links?

i mean–to say that balloon-juice links to the nro corner (as an instance) is not to say that there is any degree of ideological affiliation between them, or even that there is any dialogue in the sense of discussion likely to moderate or synthesize the views of the participants. john or dougj just link to some stupid shit that some cornerite has said, says “can you believe the stupid shit these cornerites say?”, and that’s as far as it goes.

this kind of linking–fairly prevalent among the blogs i read–has always seemed to pose a problem for these wide-scale studies of “who talks to whom among bloggers”. there’s talking and then there’s ridiculing. yes, the ridiculing does mean that the readers of balloon-juice are at least *exposed* to (some of) the content of the ridiculed blog. but then again, you’ll also hear bill o’reilly ranting about the evilness of dailykos on occasion, so his listeners are “exposed to” the content of dkos in some sense, without that exposure doing much to lessen the echo-chamber effect.

anyhow–how do these studies solve this?


Henry 04.28.10 at 2:31 pm

Eszter herself has done work on this – see “here”:http://www.springerlink.com/content/p7m41t21344130t7/fulltext.pdf.


belle le triste 04.28.10 at 2:34 pm

Didn’t the red-blue thing used to be signify “winner-loser” (or “loser-winner”) in the US, and thus switch sides every time this changed?


Substance McGravitas 04.28.10 at 2:35 pm

Maybe red meant crazy radical and it’s now true.


Harry 04.28.10 at 2:46 pm

Sorry Eszter. Please, everyone, keep the thread on track. …


ScentOfViolets 04.28.10 at 2:49 pm

does this new study do anything to distinguish respectful links from point-and-laugh links?

i mean—to say that balloon-juice links to the nro corner (as an instance) is not to say that there is any degree of ideological affiliation between them, or even that there is any dialogue in the sense of discussion likely to moderate or synthesize the views of the participants. john or dougj just link to some stupid shit that some cornerite has said, says “can you believe the stupid shit these cornerites say?”, and that’s as far as it goes.

On a more abstract note, how do you measure “degree of ideological affiliation”? I think this is the real $64,000 question. I’m not a regular reader, but I will at least look at links to Marginal Revolution or Volokh Conspiracy because they make at a minimum make a pretense of scientific inquiry and will acknowledge the validity of, say, the theory of evolution. But there are other sites that on the surface appear to have the same ideological bent (at least, in terms of policy prescriptives on specific issues) which I will not touch with the metaphorical ten-foot pole. Heck, there are sites whose general opinions are much more in accord with mine that I will not patronize because of their anti-science, anti-reality stance (to my mind, they equate to much the same thing.) For example, any site that actively promotes the belief that vaccination causes autism is one that I don’t care to read, no matter how much they diss torture or the expansion of executive power under Obama.


laura 04.28.10 at 3:21 pm

I’m sorry, Eszter, but I had some real problems with this study.

The authors say that the liberal blogosphere outstrips the right in terms of engagement with the public and with use of new technology. The Daily Kos, for example, has that whole diary thing. Maybe. But the biggest boon for the lefties has been Huffington Post. Its audience dwarfs every other blog. Why? Not because of its comment section certainly. But because of casheroo. It has an army of faceless drones constantly updating the website with new information.

Also, the authors say that Obama’s success with online activism owes much to the practices of the liberal blogosphere. This is a real stretch.

First of all, the Netroots wanted nothing to do with Obama. They were actually annoyed that his campaign never reached out to them. Secondly, Obama brought in top social networking gurus, including the guy who ran Facebook, to help him reach the public. The Netroots are just one part of a huge group of people who were using the Internet effectively.


Ari Melber 04.28.10 at 3:47 pm

To Laura’s point, the study suggests that the practices, habits and infrastructure developed by the liberal blogosphere and netroots helped pave the way for Obama’s success. It does not argue that bloggers actively coordinated with the campaign or were enthused about Obama during the primaries. (Most weren’t, to be sure.)

Here’s how I describe it my article today discussing the findings:

The study proposes a historical theory for the Democrats’ current edge online. (It’s a big one, as the 2008 election revealed, when Obama raised $500 million through the web alone. McCain raised under $200 million from all individual donors combined.)
According to the authors, the netroots’ early embrace of deeper participation platforms, coupled with progressive bloggers interest in mobilizing fundraising and specific actions, helped prime the tactics and habits that supported the Democrats’ later web dominance.

The survey data does show that progressive bloggers were far more demanding of their readers. One out of three liberal sites made direct fundraising pitches, and almost half asked readers to take some political action, according to a section of the study analyzing the top sixty-five blogs. On the right, however, only one out of twenty blogs pushed fundraising, and fewer than one out of five issued “calls to action.”



Russell Arben Fox 04.28.10 at 3:47 pm

I have to agree with Laura, I’m afraid; while the paper provides some interesting data to support their typologies, it still sets them up in, I think, simplistic ways.


Bill Gardner 04.28.10 at 5:40 pm

Digression: Benkler is interviewed by Russ Roberts in an interesting podcast on internet policy issues. Benkler is massively smart and, at least to an outsider, exceptionally well-informed. The interplay between them is also interesting, in that Roberts tries to argue for markets-only solutions on first principles, to which Benkler responds that you actually need to look at the evidence to know what to do. Yes.


clod Levi-Strauss 04.28.10 at 8:03 pm

in re: Benkler, but also the previous thread.
Call me old fashioned think it’s silly to keep referring to proponents of individualism, methodological or philosophical as on the left. Communities are constitutive, not chosen, and tastes including intellectual preferences are developed in childhood as a result of accident and influence. Freedom if it exists is marginal.

The current debate is among technocrats who see themselves as a community of monads, extreme individualists who want to “go Galt” with their social security checks, and a theocratic peasantry.
All three deserve analysis, but none deserve support.


Ben Alpers 04.28.10 at 9:04 pm

First of all, the Netroots wanted nothing to do with Obama. They were actually annoyed that his campaign never reached out to them.

This really isn’t true. There certainly were a range of responses to Obama during the campaign in the “left” blogosphere, and some sites (most notably MyDD) were quite hostile. But DailyKos (which is usually seen as the center of the “Netroots”) was a distinctly pro-Obama site during the primaries. So much so that there was an organized stomp-off of Hillary Clinton supporters during the middle of the 2008 primary campaign.


laura 04.28.10 at 9:37 pm

It’s quite well documented that the Netroots wasn’t that happy with the Obama candidacy. Several members of the Netroots actually have chapters in books about it. I can get sources if needed.

“… the study suggests that the practices, habits and infrastructure developed by the liberal blogosphere and netroots helped pave the way for Obama’s success.” Suggests. Key word. In an off hand comment in their conclusion. Their actual research certainly doesn’t show that. They didn’t do a historical study or interview Obama campaign leaders or anything. Believe me, the liberal blogosphere was not the first place on the Internet to bilk readers for money or to set up a comment section. This was all going on simultaneously all over the Internet.


Matt Austern 04.29.10 at 3:34 am

Red/blue in the US is quite recent, and I don’t think it has any particular symbolism.

The TV networks like to use bright primary colors for their electoral votes maps on presidential election nights. They used to switch every so often — nothing so systematic as winner/loser or incumbent/challenger.

In the 2000 election everyone was staring at these maps for weeks. It just so happened that that year most networks had been using red for Republicans and blue for Democrats. Everyone else pretty quickly converged on it, and people started using “red” and “blue” as shorthand, and it stuck. Purely contingent, as far as I can tell.


Down and Out of Sài Gòn 04.29.10 at 11:57 am

A small pedantic point, but why is Crooked Timber on the list of “top 155 U.S. political blogs” rather than “top 155 international political blogs”? I’ve never thought of this site as particularly American, with its Aussies, Dutch, Irish and British contributors.


jdw 04.29.10 at 12:29 pm

Back on the point of the original post, and stripping away some of the sublime and funky language, I don’t know how exciting or interesting it is to hear that there is some left/right contact in US political blogs, that some of it includes substantive argument (“deliberation”), or even that left-blogs tend to arrange for more access (“affordances”), and mobilization. Isn’t that all sort of obvious? As a non-specialist reader, this looks to me more like a case of taking the statistically low-hanging fruit and then claiming interesting discoveries.

Deliberation, access, mobilization: but on the basis of what actual information and understanding? The invasion of Iraq was facilitated by a considerable level of generally-accepted pig-ignorance, was it not; and the GWOT or whatever it is being called now, is clearly fueled by a considerable level of what you could generically or politely call xenophobia. To put it another way, there is a very dramatic failure to “deliberate” or “pay respectful attention to” the views of certain non-American, non-European people and groups, particularly cases where America is waging, or perhaps threatening to wage, war in their countries, but not only in those cases. The internet, as it happens, provides abundant resources of fighting against this kind of narrowness of view, with news and information available in many languages from many points of view, readable in all kinds of scripts, and even an astonishing level of usable machine-translation. The internet can remind us that it is a big world, and help us understand it. Hasn’t “social science” of the type reported on here shown itself to be sort of irrelevant in this context, perhaps inevitably because of the kind of “science” that it is?


tomslee 04.29.10 at 1:45 pm

I just got to reading the paper and was surprised that the comment thread hardly addresses it at all. It’s not primarily about linking behaviour, or about affiliation between blogs and official campaigns.

I am not a professional, but unlike jdw, I don’t find it at all obvious that left and right-oriented blogs use technology in different ways – at least, it’s not something I would be surprised by, I guess, but it’s not a hypothesis I could back up and that’s pretty much what I mean by “not obvious”. And to suggest that the typologies are simplistic just emphasizes the difficulty of going beyond “looking at who talks to whom” or discussing a single thing called a “blogosphere” while still pursuing something statistical, and staying outside anecdotal studies of individual events and sites.

It will be interesting to see how far studies of this ilk can go before the notion of “blogosphere” crumbles and becomes no longer useful. It will presumably happen with time (as platforms evolve) and with the sophistication of the investigations, and I will be interested to see what replaces it.


chris 04.29.10 at 6:41 pm

@20: It’s funny that you mention technological change over time — I thought one of the most notable omissions of the paper was any mention of Internet discourse *before* blogs, and in particular, USENET. The way those authors write you’d think they thought blogs sprung full-grown from the head of Jupiter — and if they actually do think that, doesn’t that lack of perspective cast some doubt on their conclusions?

The old Internet culture, IMO, depended to a significant extent on the type of people who built the early Internet and spent a lot of time on it — i.e., academics and geeks. Bringing in more mainstream users is changing that, and I would have liked to see more discussion of that. But maybe it was beyond the scope of their paper — you can’t answer all questions at once, after all.

@19: To a certain extent, I guess it’s easy to respond to “right blogs are more comfortable with top-down centralized control structures” with a gigantic “duh”, but there’s a long scholarly tradition that intuition is no substitute for rigorous study, because of an even longer tradition of failures of intuition. So it’s still worth something to see this thesis confirmed, even if it accords with what most people were probably suspecting all along.


va 04.30.10 at 6:43 am

The authors include jonswift.blogspot.com in their list of “right” blogs. This minor oversight may have skewed the results.


Uncle Jeffy 04.30.10 at 5:46 pm

Link to the PDF doesn’t work…


Bob Calder 05.02.10 at 9:44 pm

With regard to measuring sentiment, perhaps one metric that can be used is timestamps on posts and comments to determine genuineness.

When I was actively working against global warming deniers, it often seemed to me that there was a distribution system, or at least a mindless cut-and-paste mind set used to influence statistics and readership. Maybe it’s just my paranoia.

Comments on this entry are closed.