Why Does the Academy Tilt Left?
Among the most common (and largely true) complaints that the general public has about university faculty is that it doesn't look like America - not for the usual reasons of race, sex, etc., but because of political views. I work in a university, and I find that a disappointingly large proportion of university professors go to astonishing lengths to refute what is to many people painfully obvious. Is academia leftist? To rephrase Daniel Okrent’s (the former New York Times “public editor”) famous characterization of his newspaper as socially liberal, “of course it is.”
Evidence is not hard to find. People routinely publish survey data indicating that hugely disproportionate numbers of university professors are registered Democrats, or have beliefs that put them well out at the left tail of the American political distribution. For those who prefer to look into it themselves, it is an interesting exercise to go to the campaign-donation website of the Campaign for Responsive Politics. Among other things they allow you to call up campaign donations (of a certain minimum size) by occupation. Go to their site for looking up individual donors. Type “professor” under occupation and “2004” under “election cycle.” The first two pages of donors, which were all I bothered to look at, come back overwhelmingly lefty. Of 100 contributions, 91 were to Democratic candidates or leftist political groups, which is lopsidedly disproportionate. The UCLA Higher Education Research Institute reports that as the country has moved to the right, faculty has actually become somewhat more leftist. 48 percent of faculty surveyed described themselves as “liberal” or "far left" in 2001, versus only 42 percent in 1989. (The number of women faculty who so describe themselves has grown even more.) Eighteen percent are “conservative” or “far right.” Research by Daniel B. Klein and Andrew Western finds that at Berkeley registered Democrats outnumber Republicans by 9.9:1, and at Stanford the ratio is 7.6:1. A core principle of the scientific method is replication, and such findings of bias appear to be common across numerous experimental techniques. Contrary findings of no bias or rightward bias do not exist, to my knowledge. So let us accept the hypothesis as true.
Why is it so? Robert Brandon, chair of the philosophy department at Duke, notoriously claimed that academics are smart, and so the absence of conservatives in academia may indicate a lack of smart folks on the right. But that won’t do. Doctors are smart too, and going to the CRP donor-search site above and entering “doctor” yields a donation pattern almost identical to the population at large. In my scanning of the first two pages, I found 23 donations to the left, 25 to the right and two to nonideological doctors' lobbying groups.
In economic language, the explanation that naturally suggests itself is that academia is a cartel, a group where producers who should compete (universities, here) collude to fix prices or quantities. Certainly the opportunity presents itself. Like any hiring process, faculty hiring requires that the candidate be satisfactory to his employer. But the faculty process is different in at least three ways. First, the hiring decision often requires near-unanimous approval, so that a small minority can shoot down any hire they perceive to be unsatisfactory. Once entrenched, bias can then be difficult to dislodge. In a private firm, in contrast, one manager might make the hiring decision. If he consistently hires poorly his division performs poorly, and he may be fired. Second, posted academic job requirements are often extremely vague, limited to descriptions of the desired field qualifications (and often even those are not included). Apart from those research interests, unhelpful boilerplate such as “potential for major scholarly achievement” or “tremendous potential as a teacher” is often included. This means that within hiring corridors any criteria can be applied. Third, the tenure process itself provides a breakwater for purging undesirable candidates who make it through the screening at the hiring stage. While defenders of tenure (justifiably) note that it promotes freedom of thought for those who receive it, they fail to note that it can also serve to purge undesirable ideas, and make ideologically protectionist universities that might otherwise be vulnerable to competition stronger.
Despite the claims of some that the competition to hire the best scholars mitigates against an ideological cartel, the university market is such that those competitive pressures may not be as strong as those, say, compelling software firms to hire the best programmers. This is all the more true in fields where empirical claims are not easily subject to empirical rejection. (Or in fields where empirical verification of claims about the world is not even an issue.) In such fields false (or at least statistically unusual) knowledge can persist longer in the marketplace of ideas. Stanley Rothman and others find that the greatest leftward bias is found, in descending order, in English literature, performing arts, psychology and fine arts. Three of these fields are in the humanities, and generally do not rely on experimentation to weed out weaker ideas. If the competition metaphor is misplaced, then a cartel could easily persist. Following techniques commonly used in looking for discrimination in labor markets generally, Rothman et al. also find unexplained employment differences even after accounting for professional accomplishments. Note that the usual social-science cautions about a hypothesis whose testing is in its early stages apply.
Another explanation, not necessarily contradictory of the cartel theory, is self-selection. The pool of potential candidates in the academy is disproportionately leftist, this story goes, and so the hiring decisions almost inevitably are as well. (This explanation would be intolerable in a racial-discrimination litigation setting, but let that pass.) Undoubtedly there is some truth to this, as anyone who spends time as a graduate student in a reputable Ph.D. program outside the sciences can attest. But that is no answer either. While the academic life has certain features that might plausibly matter more to the thoughtful lefty than to the hard-charging righty – long vacations, flexible work hours, substantial freedom in one’s work assignments (research topics, classes taught), substantial opportunities for extended foreign travel – it is also possible that potential academics rationally anticipate that leftists dominate the hiring and tenuring process, and so their prospects in academic work will be limited. (This may explain why the right appears to have an advantage in a different intellectual market – the public intellectuals who work in think tanks not just in Washington but throughout the country.) As far as I know, there have been no attempts made to discriminate between the self-selection and cartel hypotheses. But neither is particularly friendly to the idea that competition in the academic marketplace for ideas is so robust that the ideas that professors advocate are likely to be the best ones.
As a final note, in the aforementioned article from Duke, the history chair John Thompson notes that “[t]he interesting thing about the United States is that the political spectrum is very narrow,” comparing it unfavorably to Canada, where there is more support in the public at large for a big welfare state. There are two problems with this argument. First, the welfare state is probably the only sense in which European and Canadian opinion is broader than in the U.S. On many matters debate is far more robust, and the spread of politically permissible views far wider, in the U.S. generally but, conspicuously, not on American campuses. Where in European political-party platforms are campaigns for pro-life causes, gun ownership, traditional marriage and so on? Advocates of single-payer health care, slavery reparations, socialism and the like are not difficult to find on campus (even though, in the case of socialism, one needs an electron microscope to find them in the population at large), but the views that animate the American right are far rarer on campus than in society at large. Finally, what does it say about the American academy and American society at large that the standard of comparison is not the U.S. but other countries? The standard of comparison is not to Europe, or some idealized distribution of true ideas, but to the actual distribution of American views. Making that comparison yields decisive evidence in favor of the proposition that we faculty are simply not like everyone else. Is that a problem? That is a question for another day.
Evidence is not hard to find. People routinely publish survey data indicating that hugely disproportionate numbers of university professors are registered Democrats, or have beliefs that put them well out at the left tail of the American political distribution. For those who prefer to look into it themselves, it is an interesting exercise to go to the campaign-donation website of the Campaign for Responsive Politics. Among other things they allow you to call up campaign donations (of a certain minimum size) by occupation. Go to their site for looking up individual donors. Type “professor” under occupation and “2004” under “election cycle.” The first two pages of donors, which were all I bothered to look at, come back overwhelmingly lefty. Of 100 contributions, 91 were to Democratic candidates or leftist political groups, which is lopsidedly disproportionate. The UCLA Higher Education Research Institute reports that as the country has moved to the right, faculty has actually become somewhat more leftist. 48 percent of faculty surveyed described themselves as “liberal” or "far left" in 2001, versus only 42 percent in 1989. (The number of women faculty who so describe themselves has grown even more.) Eighteen percent are “conservative” or “far right.” Research by Daniel B. Klein and Andrew Western finds that at Berkeley registered Democrats outnumber Republicans by 9.9:1, and at Stanford the ratio is 7.6:1. A core principle of the scientific method is replication, and such findings of bias appear to be common across numerous experimental techniques. Contrary findings of no bias or rightward bias do not exist, to my knowledge. So let us accept the hypothesis as true.
Why is it so? Robert Brandon, chair of the philosophy department at Duke, notoriously claimed that academics are smart, and so the absence of conservatives in academia may indicate a lack of smart folks on the right. But that won’t do. Doctors are smart too, and going to the CRP donor-search site above and entering “doctor” yields a donation pattern almost identical to the population at large. In my scanning of the first two pages, I found 23 donations to the left, 25 to the right and two to nonideological doctors' lobbying groups.
In economic language, the explanation that naturally suggests itself is that academia is a cartel, a group where producers who should compete (universities, here) collude to fix prices or quantities. Certainly the opportunity presents itself. Like any hiring process, faculty hiring requires that the candidate be satisfactory to his employer. But the faculty process is different in at least three ways. First, the hiring decision often requires near-unanimous approval, so that a small minority can shoot down any hire they perceive to be unsatisfactory. Once entrenched, bias can then be difficult to dislodge. In a private firm, in contrast, one manager might make the hiring decision. If he consistently hires poorly his division performs poorly, and he may be fired. Second, posted academic job requirements are often extremely vague, limited to descriptions of the desired field qualifications (and often even those are not included). Apart from those research interests, unhelpful boilerplate such as “potential for major scholarly achievement” or “tremendous potential as a teacher” is often included. This means that within hiring corridors any criteria can be applied. Third, the tenure process itself provides a breakwater for purging undesirable candidates who make it through the screening at the hiring stage. While defenders of tenure (justifiably) note that it promotes freedom of thought for those who receive it, they fail to note that it can also serve to purge undesirable ideas, and make ideologically protectionist universities that might otherwise be vulnerable to competition stronger.
Despite the claims of some that the competition to hire the best scholars mitigates against an ideological cartel, the university market is such that those competitive pressures may not be as strong as those, say, compelling software firms to hire the best programmers. This is all the more true in fields where empirical claims are not easily subject to empirical rejection. (Or in fields where empirical verification of claims about the world is not even an issue.) In such fields false (or at least statistically unusual) knowledge can persist longer in the marketplace of ideas. Stanley Rothman and others find that the greatest leftward bias is found, in descending order, in English literature, performing arts, psychology and fine arts. Three of these fields are in the humanities, and generally do not rely on experimentation to weed out weaker ideas. If the competition metaphor is misplaced, then a cartel could easily persist. Following techniques commonly used in looking for discrimination in labor markets generally, Rothman et al. also find unexplained employment differences even after accounting for professional accomplishments. Note that the usual social-science cautions about a hypothesis whose testing is in its early stages apply.
Another explanation, not necessarily contradictory of the cartel theory, is self-selection. The pool of potential candidates in the academy is disproportionately leftist, this story goes, and so the hiring decisions almost inevitably are as well. (This explanation would be intolerable in a racial-discrimination litigation setting, but let that pass.) Undoubtedly there is some truth to this, as anyone who spends time as a graduate student in a reputable Ph.D. program outside the sciences can attest. But that is no answer either. While the academic life has certain features that might plausibly matter more to the thoughtful lefty than to the hard-charging righty – long vacations, flexible work hours, substantial freedom in one’s work assignments (research topics, classes taught), substantial opportunities for extended foreign travel – it is also possible that potential academics rationally anticipate that leftists dominate the hiring and tenuring process, and so their prospects in academic work will be limited. (This may explain why the right appears to have an advantage in a different intellectual market – the public intellectuals who work in think tanks not just in Washington but throughout the country.) As far as I know, there have been no attempts made to discriminate between the self-selection and cartel hypotheses. But neither is particularly friendly to the idea that competition in the academic marketplace for ideas is so robust that the ideas that professors advocate are likely to be the best ones.
As a final note, in the aforementioned article from Duke, the history chair John Thompson notes that “[t]he interesting thing about the United States is that the political spectrum is very narrow,” comparing it unfavorably to Canada, where there is more support in the public at large for a big welfare state. There are two problems with this argument. First, the welfare state is probably the only sense in which European and Canadian opinion is broader than in the U.S. On many matters debate is far more robust, and the spread of politically permissible views far wider, in the U.S. generally but, conspicuously, not on American campuses. Where in European political-party platforms are campaigns for pro-life causes, gun ownership, traditional marriage and so on? Advocates of single-payer health care, slavery reparations, socialism and the like are not difficult to find on campus (even though, in the case of socialism, one needs an electron microscope to find them in the population at large), but the views that animate the American right are far rarer on campus than in society at large. Finally, what does it say about the American academy and American society at large that the standard of comparison is not the U.S. but other countries? The standard of comparison is not to Europe, or some idealized distribution of true ideas, but to the actual distribution of American views. Making that comparison yields decisive evidence in favor of the proposition that we faculty are simply not like everyone else. Is that a problem? That is a question for another day.
posted by Evan at 11:53 AM
0 Comments:
Post a Comment
<< Home