[I wasn’t happy with the earlier version of this post, so I’ve revised it substantially]

I’ve talked about the role of multivocal signaling in American politics before, particularly in terms of “dog-whistle politics”: the art of sending different signals to different audiences. Henry Farrell’s review of Jacob Hacker’s and Paul Pierson’s Off Center: The Republican Revolution and the Erosion of American Democracy shows that I’m not the only academic who has been “connecting the dots” between John Padgett and Christopher Ansell’s seminal article and the way that politicians increasingly tackle the problem of building coalitions between ideologically incompatible segments of the electorate.

Hacker and Pierson, at least according to Henry, have written a fascinating account of how the Republicans constructed the infrastructure necessary to sustain multivocal action in an information-rich environment. This is no insignificant feat. Multivocal signaling usually requires that audiences do not share information with one another, or at least have some sort of differential access to information (consider the way that many Palestinian leaders, at least in the past, would say moderate things in English but then give red-meat speeches in Arabic). If audiences share information, they may demand that a politician make unambiguous his or her identity and preferences – a phenomenon I’ve called, rather clumsily, “forced clarification.”

It seems that this might be happening to the Bush administation. David D. Kirkpatrick of the New York Times reports:

When Gov. George W. Bush of Texas hit the presidential campaign trail, he seldom brought up his view of abortion. But with conservative Christian crowds, he never missed an opportunity to praise “pregnancy crisis centers.” Abortion opponents, knowing such centers steered women away from the procedures, cheered and took heart.

It was the beginning of a delicate balancing act that, until President Bush picked Harriet E. Miers for the Supreme Court last week, had enabled him to forge an unprecedented bond with social conservatives without unnerving more moderate voters. President Bush may have perfected it during the 2004 presidential debates. He said he would not appoint justices who would approve of the Dred Scott decision – the 19th-century fugitive slave case that abortion foes compare to Roe v. Wade – but also pledged not to make the abortion issue a “litmus test” for judicial nominees.

The nomination of Ms. Miers demonstrated the fragility of a coalition built in part on code. The administration relied on subtle clues about her evangelical faith and confidential conversations with influential conservative Christians to enlist grass-roots support for Ms. Miers.

Instead the Miers nomination has threatened to shatter the coalition that Mr. Bush and his adviser Karl Rove hoped would be the foundation of a durable Republican majority. Social conservatives say that Mr. Bush made them tacit promises to appoint justices who would rule their way on abortion and other social issues. They wanted a nominee with a clear record and Ms. Miers had none.

One of things I like about Kirkpatrick’s article is that he doesn’t restrict his analysis to the “codes” that Bush uses to engage in multivocal signaling, but looks at how a variety of conditions might be eroding Bush’s strategy.

And this is important, in my view. Multivocal signaling is about action, not simply the code words a politician uses. It can be dangerous to focus too much on the inherent properties of political rhetoric. People evaluate the identity – and dispositions – of a speaker by interpreting both what he does and how he justifies it.

Indeed, politicians can use all the code words and ambiguous phrasings they want, but it won’t help them very much unless some of the audiences he’s targeting is predisposed to view them as a member of their ‘in group.’ For a variety of reasons, Bush built up this kind of relationship with the Christian right, which gave him a lot of running room over the last term. By the same token, it isn’t completely surprising that he hasn’t been able to fulfill the rising expectations of his most important constituency (and boy did Christian conservatives have rising expectations after the 2004 election). As their sense of concern mounts, it is entirely possible that they will demand “clarification” of Bush’s preferences and identity, and that this might trigger a cascading failure of Bush’s political strategy.

This is the possibility Kirkpatrick highlights:

In its efforts to quell the revolt from its base, the administration has come increasingly close to characterizing Ms. Miers’s views. This carries its own political risks as well, including energizing liberal opponents.

Thursday afternoon, White House aides enlisted Dr. Dobson, Mr. Land and others in a conference call to explain their support for the nomination to uneasy conservatives around the country. Dr. Dobson assured them he was convinced she was an opponent of abortion.

Jay Sekulow, chief counsel of the American Center for Law and Justice, said he wanted her on the court for an upcoming partial birth abortion case. And Mr. Land asserted that both the president and Ms. Miers would consider it “a deep personal betrayal” if she ruled against his expectations.

Democrats and liberal groups, eager to paint Ms. Miers as a darling of the right, have quickly seized on the administration’s efforts. On Friday, the liberal group People for the American Way began disseminating a transcript of the conservative conference call to its own allies and Democratic aides. And Democratic Senators have started a steady drumbeat of calls for Dr. Dobson to disclose whatever he has been told. Mr. Olasky said the stakes of the Supreme Court nomination limit the effectiveness of what he called Mr. Bush’s “secret language.”

I’m not convinced that such a scenario is in the offing, but all of this holds some important lessons.

First, understanding political signaling, multivocal or otherwise, won’t work unless we keep in mind that rhetoric always involves a “dialogue” between a speaker and her audience(s). The latter are not passive recipients, but active intrepeters.

Second, variation in structural conditions will make multivocal and other forms of action more or less likely to succeed. Again, static discourse analysis or interpretations of rhetoric won’t get us an understanding of these conditions. A lot of other factors, such as interest-group bargaining or the hunger of a constituency to win an election, may explain why an audience won’t seek clarification when their favored politicians engage in ambiguous political action or why they may interpret ambiguous signals as proof positive of a shared political identity and agenda.

It could simply be, for example, that “dog-whistle politics” stop working when there’s no one else to blame.

Filed as: multivocality, Miers, and political rhetoric