No End of Ideology: A Partial Response and Addendum
by Scott Winship
I wasn’t going to take the time to respond to the criticism of my piece in the American Prospect from last week, but then two prominent bloggers questioned my honesty and the quality of my work, which I can’t abide. This response is intended to be a defense of my motives and research. The American Prospect
will soon publish has published a companion response piece to substantive criticisms levied at my essay by commentators affiliated with the magazine. I urge you to read both responses if you are interested in this debate.
I do want to emphasize from the outset that the arguments I have made and make below are mine and not the views of The Democratic Strategist. The Strategist takes no stand on this or any other issue, being dedicated to empiricism and to engaging all factions of the Democratic Party in order to build an enduring majority. As managing editor, one of my jobs is to solicit contributions from a diverse pool of political professionals, astute observers, academics, and activists. As blogger, I am responsible for providing my own perspective on the events and issues of the day. As it turns out, I have spent far more time on the issue of the netroots than I ever intended, and I hope that with these responses I can move on to other important topics.
First, I need to make two exculpatory clarifications by way of defending myself. The first is that I was not responsible for the “slug” used to promote my piece (“Netroots members insist that they’re non-ideological pragmatists. They’re wrong.”). While I don’t disagree with the point it conveys, the way it was expressed is more antagonistic than I would have begun my piece. The title that the Prospect’s editors gave to my friend Mark Schmitt’s piece – “Putting Down Netroots” – also had the unfortunate effect of making me look hostile to the netroots. I don’t fault the American Prospect for this – the provocative words the editors used likely brought more readers to my piece, and they need to attract eyeballs themselves.
I tried as best I could to make my piece as dispassionate as possible while still making the points I intended to argue. I hope that whatever the reader thinks of my piece, he or she will agree that it is written in a respectful tone. Indeed, nowhere do I indicate that I think liberalism is bad, and truth be told, I have quite liberal views myself (see this introductory post to my blog).
All of this is a preface to denying Chris Bowers’s assertion that I “clearly [take] sides with” the view that the netroots are “amateurish ideologues whose across-the-board liberalism will drive the party off a cliff.” In hindsight, it’s regrettable that after laying out the two opposing views among the most vocal participants in the Lieberman debate, I preface my analysis with “Who’s right?” which implies that I must choose one or the other side. To clarify, I do take the views that the netroots are almost uniformly liberal, that this liberalism affects their politics, and that it may have implications for the Democratic Party’s electoral success. On the other hand, nothing in my piece argues the netroots are “amateurish”, and I acknowledge in the final paragraph of my piece that even if I am right about the netroots’ ideological predisposition and the role it plays in their decisions, it may not be problematic for the party.
The second clarification is that I would like to have addressed more points than I did in the piece, but I was already well over the word limit I had been given. Space limitations were the reason I did not discuss Chris’s survey, and as will become clear, I don’t believe my omission of any mention of it makes my case “flimsy”. I strongly object to any insinuation that I omitted mention of it out of dishonesty.
On that note, while I have enjoyed my email discussions with Chris and while I admire the value he places on bringing data to netroots discussions, I am unpersuaded that his BlogPac survey of MoveOn.org members is the “best” data available. (Incidentally, I previously had explained this opinion and defended my reasons for using the Pew survey here and here.) Chris is right that my initial objection was to the low response rate in his survey, which may not be low by the standards of internal campaign polls but is very low indeed by the standards of academic research. Chris dismisses the risk low response rates pose to the validity of his results, calling my concern “preposterous”, but low response rates are generally a very big problem.
By way of example, let’s say one had wanted to conduct a survey of the views of Democratic voters in 2004. But to do so, one conducted a survey of delegates at the Democratic National Convention – a very small and unique group of Democratic voters. From this survey, one would have concluded [pdf] that only 12 percent of Democratic voters believed the federal government was too involved in private life. In actuality, the figure was 45 percent. The problem is that convention delegates do not represent Democratic voters as a whole.
When the decision to participate in a survey or not is up to the individual, if only a small number of them agree to participate, then there’s every reason to worry that those people are unrepresentative in the same way as DNC delegates are unrepresentative of Democratic voters. Indeed, this is almost surely the biggest problem with presidential exit polls – the people who agree to stop and take the survey are different from those who do not (in particular, they are more Democratic).
The problem of sample self-selection is potentially worsened in Chris’s case because he has – not inappropriately – been a strong advocate of the view that perceptions of the netroots are inaccurate. If the MoveOn.org members who received a request to participate in the survey were informed that the survey was sponsored by BlogPac or involved Chris, I would worry that this would disproportionately attract those who share Chris’s desire to prove the MSM wrong about netroots demographics and politics. The result would be a very select group of respondents that didn’t represent MoveOn.org members as a whole.
The Pew survey I used in my piece has a response rate twice that of Chris’s, but even so, it falls well short of academic standards too. I previously defended the Pew survey on the grounds that the researchers had conducted checks to see how biased their select sample might be. I have since learned that Chris’s survey involved similar checks, and so my concerns are somewhat alleviated. Nevertheless, for Chris to argue that in principle low response rates couldn’t bias his results is just all kinds of wrong.
My bigger concern about Chris’s results, though, is that he defined the “netroots” in a peculiar way. Chris and I disagree, but my own starting point was that the netroots consist of political blog participants. That is also Markos Moulitsas’s view, as indicated by the quote I included in my piece. Chris disagrees and would extend the definition to include, for instance, anyone who ever signed a MoveOn.org petition or donated money to a campaign online. (He expressed this view to me via email.) The problem is that this definition doesn’t pass the mom-and-dad test. My parents have been quite active with their local MoveOn.org chapter, but until I started writing about them a few weeks ago, they had no idea what the netroots were and only read blogs if I pointed them toward one.
If one agrees with Markos and me that one needs to participate in political blogs to be in the netroots, then Chris’s working definition makes it impossible to interpret his results as representing the netroots. According to the figures on his “50-State Strategy” chart here, those who “regularly” read “blogs such as Daily Kos, Talking Points Memo, or MyDD” amount to just 19 percent of his sample. Nearly 4 in 10 said they never read blogs.
I should say that even if I did not disagree with Chris, I don’t have access to his data, so I have to make do with what is available. Chris clearly has information on blog participation in his survey, so it would be easy for him to re-run his results looking only at that sub-group. Do I know that his results would confirm mine? No, but I’d be more inclined to take them seriously. Heck, if he wants to give me the data, I’ll look at the results and summarize them myself, whether they support or contradict my conclusions. I have no dog in this fight.
That said, beyond Chris’s data and sample, I do dispute a few of his other conclusions. (From here on out, I will take the above data and sample issues as unproblematic.) Consider first his chart showing that 73 percent agree that the party should run inspirational candidates who communicate clearly, while 24 percent agree that it should run candidates who are liberal across the board. I’ll first note that the fact that one-quarter of the netroots – by Chris’s definition – is unambiguously ideological is not to be dismissed lightly. Furthermore, if we compare the percent who strongly agree with each position – and the netroots are nothing if not strong-minded – the split is not 73/24 but 32/18. Restricting the definition of “netroots” to blog participants would surely produce a result showing that a larger share is unambiguously ideological.
Another problem with Chris’s interpretation eludes the basic issue at hand. What is inspirational to the netroots is likely going to elicit their liberalism or populism (see Governor Schweitzer). While it may be, as Chris’s poll finds, that in the abstract the netroots prefer inspiration to (uniformly) liberal positions, in practice they are not likely to find inspirational candidates who do not share the bulk of their views. If netroots members assume that these liberal positions or populist attitudes will inspire other voters, as I claim a substantial share does, then this assumption may affect the fate of the party.
Who should the party run in conservative and swing districts? One in five of Chris’s respondents asserts it should run liberals no matter what the chances of winning. The rest say they would tolerate a moderate if “a liberal or progressive candidate may have little chance of winning.” Put this way, the fundamental problem is again eluded. The issue is when and where a liberal has little chance of winning. And one point that I make in my piece is that two-thirds of the netroots seems to believe either that voters are more liberal than they are or that a candidate’s ideology doesn’t matter. Finally, I’ll again note that the breakdown among those who strongly agree that the party should run moderates in moderate areas is not 78/19, but 36/12.
Regarding Chris’s emphasis on the diverse Democrats supported by the netroots, I’ll first note that of the six most favorably-rated politicians in his data, none voted for the Iraq resolution (either because they voted against it or because they were not in Congress at the time of the vote). Furthermore I think the two most relevant findings are that Russ Feingold leads all potential presidential candidates except Al Gore and Barack Obama – both of whom claim they are not considering running – in the number of people giving them a “very favorable” rating (and Jack Murtha and Barbara Boxer out-poll him), and that less than four in ten of the people in Chris’s sample gave very favorable ratings to any candidate save Gore and Obama.
Before putting Chris’s criticisms aside, I have to emphasize that his post ironically demonstrates one of the central points of my essay: that the ideology of the netroots – masked as pragmatism – serves to stifle critiques from moderates. Chris worries that if (when?) Lieberman loses, my piece will be used to bolster negative views of the netroots. Put aside the fact that I claim nothing about the netroots other than that they are liberal and ideological; my piece begins with the sentence, “Tuesday’s Connecticut primary race between Joe Lieberman and Ned Lamont is not about the netroots.” (emphasis in original)
To be clear: It’s not that I don’t care whether my piece is misused by ideologues on the center or right; it’s that I care more whether the ideology of the netroots will hurt the party. The netroots can’t rule out on principle critiques from the center because they “reinforce Republican talking points.” To do so is to use partisanship as an excuse for opposing an ideological critique that runs counter to their liberalism.
To Stirling Newberry’s charge that “people like [me]” are ideological, I can only say, “Guilty!” Note that nowhere in my piece do I say being ideological is inherently bad – I just raise the issue of whether a particular ideology is helpful or hurtful to the party. The only relevant question about my own ideology is whether it supersedes my objectivity, and if Newberry wants to argue it does, then bring…it…on…
It seems that Newberry needs a guide to rhetoric, which I’m happy to provide here. The belief that his TPMCafe photo is pretentious, for instance, would be an opinion. The major arguments in my essay would be supported factual claims, even if we might disagree about the extent to which they are adequately supported. The bulk of Newberry’s blog post consists of opinion and unsupported factual claims. For example, he writes that I assume that moderates aren’t ideological. This is a factual claim about me, but unsupported by anything I wrote in my piece. About moderates, he writes,
They are more willing [than the netroots] to engage in violence to defend their interests and world view, they are more hostile to outsiders and they are more rigid in their thinking. They have an ideology which they use to force fit everything into a very small view of the world.
For Newberry and the similarly confused: this is a mix of unsupported factual claims and opinion. The claim about violence is also – opinion coming – bizarre.
Newberry has the right to call me “sloppy”, but he could not have picked a more offensive epithet to throw at me. Hopefully it’s clear from this defense and my companion response that he wouldn’t know sloppy if it gave him a wet kiss.
Moderates care about the party as much as liberals do and in fact share most of their views. The question of whose strategic assessment of the electorate is more correct is an empirical one that all of us ought to be working to get to the bottom of. That was the intent of my piece, and it is the intent of everything that I write in this blog. Aside from the side-splitting humor.