Keys To Victory? Constructing Liberal Identity, Values And Narrative For A Political Realignment

by: Paul Rosenberg

Sat Aug 11, 2007 at 13:59

Since then, the tides have turned, creating an opening for us  ("Democratic Party is growing more liberal" [McClatchy/Pew]). But the underlying problem remains--even with a recent shift toward liberalism, far more people self-identify as conservatives.  Here, I want to present the core of an idea about how to radically change the structure of the political universe in a few short years--precisely the sort of restructuring that typically accompanies classic realigning elections such as 1860, 1896 and 1932.

I begin with the observation that liberal/conservative split in candidate support is wildly out of line with attitudes on specific issues, and proceed from there to the idea of conservative identity politics as the core of what we are up against.  I then ask what it would take to create a counter, liberal (or progressive) identity politics, which naturally engages us in diversity issues as well as questions of unifying narratives and defining values.

Because of length, I'm splitting this into two parts.  Part one is the setup, ending with conservative identity politics.  Part two starts with what the challenge of creating a liberal identity politics.

To begin with, in his analysis, Chris observed:

In all of my post-election analysis, when all of the tactical ideas are put aside, I keep coming back to a single, basic idea: conservatives are the enemy, and conservatism as an ideology is our main roadblock to electoral success.

We have long since left the era when the two parties could accurately be considered regional and ethnic coalitions rather than ideological coalitions. There are no longer any more conservative Democrats than there are liberal Republicans. A few of each kind manage to hang on, but the ideological vote in this election was clear:

	Bush	Kerry	Margin
Conservative	84	15	69
Liberal	13	85	72

For that matter, the ideological vote was also clear in 2000:

	Bush	Gore	Margin
Conservative	81	17	64
Liberal	13	80	67

In both cases, the overwhelming majority of liberals voted for the Democrat, while the overwhelming majority of conservatives voted for the Republican. The 2000 and 2004 margins are almost identical, with the reduction of third-party votes accounting for almost the entire difference between the two elections.

Now, the first major point I want to make is that these ideological differences in candidate support represent an extreme divergence not found on individual issues.  To make the comparison, we need to start by flipping the axis of analysis from candidate margins within ideologies to those across ideologies:

2000	Bush	Gore	Total
Conservative	81	17
Liberal	13	80
Margin	68	63	66

2004	Bush	Kerry	Total
Conservative	84	15
Liberal	13	85
Margin	71	70	71

The Liberal/Conservative Issues/Value Overlap

In sharp contrast to the situation seen with candidates, with margins around 85% and overlaps around 15%, the liberal/conservative convergence on issues and values is dramatic.  On spending issues, for example, the situation is virtually the reverse.  On a set of 11 spending issues polled from 1984 through 2004 as part of the General Social Survey (GSS), the average liberal/conservative margin was 13.6%, the overlap was 86.4%.  The margins ranged from 3.9% on spending for space to 22% on military spending.

This overlap is not necessarily a sign of a large consensus (we'll get to consensus in a moment), but it does undermine the movement conservative demonization of liberals in two ways: First, to the extent that conservatives take liberal or moderate positions that agree with liberals, liberal attitudes can't be that terrible.  Second, to the extent that liberals take moderate or conservative positions, liberals themselve can't be that terrible.  The rhetoric of demonization depends on having an other that is really other, not one that agrees with you more than 4 times out of 5.

Although it's not logically necessary, it turns out that contained within this overlap there actually is a fairly strong consensus.  This is because relatively few liberals take a conservative position.  The conensus is not necessarily for more spending on domestic programs, but it is very definitely opposed to less spending. 

Shifting to another set of measures (partially overlaping this set of 11)  that are specifically welfare state measures,  we find majority support for inceasing spending on the following, measured from 1992 to 2000, a period of GOP dominance in Congress:

	Too little	About right	Too Much	Liberalism Index *
Improving nations education system	55.5	22.0	22.5	71.2
Dealing with drug addiction	52.6	29.8	17.7	74.8
Social Security	51.0	37.6	11.4	81.7
Improving & protecting nations health	50.5	28.0	21.6	70.0
* Liberalism index = "too little" [liberal position] / ("too little" + "too much") [liberal position + conservative position]. A useful single number to indicate level of support.

And we find landslide majority support for either increasing or maintaining spending on the following, during the same time period:

	Too little	About right	Too Much	Liberalism Index
Solving problems of big cities	49.8	26.1	24.1	67.4
Assistance to the poor	45.6	27.6	26.7	63.1
Improving & protecting environment	44.2	30.7	25.1	63.8
Highways and bridges	41.4	46.1	12.5	76.8
Mass transportation	34.5	47.2	18.2	65.5
Parks and recreation	29.5	55.3	15.1	66.1

Now, here's the kicker: The above figures are not a measure of public opinion in general.  The public as a whole is more liberal than this.  In fact, the above figures are not even a measure of conservative public opinion in general.  They are a measure of attitudes among the 3% of people who self-identified as the most extreme conservatives-7 on a scale of 7.

This is the worst possible news for rightwing demonizers of liberals: the overlap on issues isn't because even liberals see the folly of the welfare state.  Just the opposite-it's because even extreme conservatives see the wisdom and necessity of the welfare state.  They may not like it ideologically, but they are pragmatic enough to realize that it's necessary, and so they support it in no uncertain terms.

This sort of overwhelming conservative support for the welfare state is one of the main reasons for the "culture war"-understood in its common sense as a war over social issues, rather than economic ones.  Simply put, movement conservatives had no chance whatsoever to win a political war based on destroying the welfare state.  Even their own base of self-identified extreme conservatives would not support them in such a war.

Here's the funny, however.  While there's not a uniformly strong liberal consensus on culture war issues, neither is there dramatically stronger polarization.  On a full battery of 18 GSS questions on abortion, the overall average margin between liberals and conservatives was just 15.9%--not much different than the margin on spending items.  The overlap was 84.1%.  On a battery of 13 GSS affirmative action questions, the overall liberal/conservative margin was 14.6%, while the overlap was 85.4%. On a battery of 9 GSS gay and lesbian questions, the overall liberal/conservative margin was 19.3%, while the overlap was 80.7%.  All this suggests that differences on social issues need not polarize us and prevent agreement on economic policies, policies which can, in turn, reduce the intensity of investment in social issues.  Of course, this is presently just a theoretical possibility.  But it is a very real one, if we can devise a strategy for shifting the nature of our political discourse.

Conservative Identity Politics

I did a series of posts early last year devoted to the notion of conservatism as a form of identity politics.  In it (here), I wrote:

Here's my thesis: Conservatism is a form (indeed the original form) of identity politics. It is expressed through multiple forms of political ideology based on justifying elite rule and the division of the human race into dualized classes (ideal and counter-ideal) in terms of some "natural" moral order.

Conservatism appears in various forms as the rationalizations and dualized classes shift over time, and in three distinct states of realization, reflecting different levels of development of the self. The overt rationalizations commonly mistaken for conservative ideology are, in fact, derivative phenomena-tertiary at best. The primary phenomena is the creation of a conservative identity, the subject of conservative political narratives. The secondary phenomena is the supporting ideology of superior and inferior groups, casting conservative identity as something to be preserved, promoted, and defended against the forces of evil, embodied in its demonized others. The primary and secondary phenomena are relatively constant over time, while the tertiary phenomena vary considerably.

Just to underscore, what I'm saying here is that so-called "principled conservatism" is teriary at best.  And that's why it's so readily abandoned when conservastives actually have the power to do virtually whatever they want.  Their real, core belief is in their own superior virtue, which is why they have the right to do anything to anyone at any time, and still claim to have God on their side.

My second post in the series dealt with hard core data.  I'm just going to focus on one part of that, which is what really first crystalized the idea of conservative identity politics for me.  It comes from a landmark study by  Lloyd Free and Hadley Cantril published in 1967, The Political Beliefs of Americans: A Study of Public Opinion<, based on surveys done by Gallup in 1964. This study marked the original discovery of the basic fact described above that a majority of conservatives support welfare state spending on the sort of social programs that defined New Deal liberalism. Free and Cantril used three different measures of liberalism and conservatism.  One was simple self-identification. The second was an ideological scale composed of responses to five questions. The third was an operation scale based on support for spending on five programs.

What Free and Cantil discovered then remains true to this day: self-identitified and ideological liberals are a relatively small part of the electorate, while operational liberals are a landslide majority.  Conversely, self-identified and ideological conservatives are relatively large groups compared to their liberal counterparts, while operational conservatives are not even a majority among conservatives themselves by either other measure.  It makes sense to regard operational conservatism as the core of conservatism for at least three main reasons: First, because it empirically is the core.  This is the relatively small group of people who are overwhelmingly conservative by all three measures. Second, because it reflects the actual political agenda of the conservative movement over time-its political leadership, that is.  Third, because it fits with a variety of different theoretical constructs.

The findings I want to focus on concern the amount of power that different groups have in society.  People were asked if blacks ("negroes"), Jews, Catholics, and unions should have "more influence," "less influence," or if the present amount of influence was "about right." Amazingly-or perhaps not so much-those saying that blacks should have less influence was just slightly more (31%) than those saying they should have more influence (30%).  Remember, this survey was conducted before the 1965 Voting Rights Act was passed. The majority of blacks in the South could not vote, there were no black mayors and there were just four black representatives in Congress-less than one percent of Congress, and less than 1/10th the number they would have if they were represented proportionately to their numbers.  But what concerns us here is not simply the attitude toward blacks, but operational conservatism and the attitude toward accepting the political power of all these WASP outgroups-including unions, who represent the organized working class.  As the chart below shows, there is a consistent relationship in attitudes toward all four groups: the more operationally conservative, the greater the opposition to power-sharing:

Opposition To Power-Sharing By Operational Spectrum Percent saying each should have "less influence"
Operational Sprectrum	Negroes	Jews	Catholics	Unions
Completely Liberal	21	16	18	33
Predominantly Liberal	35	21	26	50
Middle of the Road	35	24	28	64
Predominantly Conservative	45	24	32	69
Completely Conservative	60	27	41	92

The above chart strongly supports what all of us already know from an abundance of experience and annecdoratal evidence-conservatives don't much trust or play well with others.

This is not necessarily true on a personal level, of course, which is something that conservatives always use to defend themselves.  "Some of my best friends are negroes," was the classic Southern segregationist line back in the 1960s.  But, of course, that "friendship" was entirely defined within lines drawn by the segregationists.  Lines are not nearly so rigidly or hypocritically drawn today, of course.  But the principle still remains: a conservative's definition of personal relationships is used to excuse prejudicial group attitudes.

The chart also implicitly tells us that good conservatives are WASP-identified: they either are WASPs themselves, or else they think that WASPs have the right ideas about things, and should be allowed to run the show.  They either aspire to be WASP-like themselves, or else are happy in a subservient role.

Furthermore, the distrust of others centered among operational conservatives also explains why demonizing liberals as the "other" comes so naturally to conservatives.  They have plenty of practice thinking similar thoughts about lots of other groups of people--all of whom, coincidentally enough, liberals have stood up for when conservatives have attacked them.

The challenge for liberals/progressives, which will bregin Part 2, is how to respond by forging our own identity politics, when (a) we do not have a single cultural identity to cluster around and (b) we have some residual degree of distrust in power-sharing ourselves.