The following commentary was originally written the day after the 2003 U.S. Supreme Court rulings in Grutter v. Bollinger  and Gratz v. Bollinger, and about a month after the author's exclusive investigative op-ed in the Wall Street Journal exposing contractions between a husband and wife scientific team the University of Michigan offered as "expert testimony" claiming that "diversity" had alleged "educational benefit. June 24, 2003. The Amazon advertised book at left, released August 7th, 2007, discusses the scientific questions in detail.

by Chetly Zarko, © 2003.

A New Era for "Social Science"
The New “Academic Freedom” to Violate Civil Rights.

Something went unnoticed in the heat of the day and the debate over racial preferences and the 14th amendment following Monday's Supreme Court decisions, Grutter v. Bollinger  and Gratz v. Bollinger, et al. The decision may have gone much farther than merely permit universities to use racial preference in admissions.

One possible construction of the decision would give universities a special academic "deference" to violate any constitutional right, from the fourth-or-fifth amendment protections against unreasonable search and self-incrimination to the first amendment's "freedom of speech" (think in terms of the "educational benefit" of speech codes and procedural styles of student tribunals). The irony is that this "deference" derives from the "first amendment" itself. All a school now needs to do is publish "social science" suggesting an “educational benefit" (or some other "compelling interest" yet to be defined) for its action, and under the newly expanded doctrine of "academic freedom" (previously only recognized in 1959 to protect individual professors academic expressions from federal government investigations related to 'suspicion of communism') the Court might abandon a concept known as “strict scrutiny.” The victim of a Constitutional rights violation then must do the nearly impossible - prove the university policy or science justifying it is in "bad faith." There is a difference between using “academic freedom” as a shield to protect individual professors from the vagaries of political persecution; and now using it as a license not to strictly scrutinize the science itself used to justify and implement other civil rights violations. That distinction is now blurred. The ramifications for all civil rights in the educational sphere are now in question. Ironically, universities, of all places, may have a special first amendment right to violate other parts of the first amendment (if there is an "educational benefit" or some other "social science" justification). All of the Justices in dissent strongly criticized the move away from “strict scrutiny.”

Another irony of these cases is that the Plaintiffs could have almost proved such “bad faith” in the quality and scope of the social science used in 1999. They just didn’t follow through. Therein lies a story, now a footnote in American history.

Back in the early phases of these cases, the Defendants commissioned several important studies, namely by Professors Patricia Gurin, Steve Raudenbush, and former university presidents Derek Bok and Bill Bowen.  Among these studies, the Plaintiffs sought the primary data these scientists used in formulating their opinions that “diversity” is “educationally beneficial.” Plaintiffs were relying on both discovery rules and a law requiring experts giving testimony to provide all materials used in formulating their opinions.  In some cases, the Plaintiffs received questionnaires used in the studies, but in no known instance did they receive the raw data underlying these conclusions. Defendants issued a letter denying their legal obligation to turn over the data, and thus denied Plaintiffs an informed scientific challenge. Fights over discovery are often expensive and indeterminate, and the Plaintiffs made a strategic decision not to pursue the matter.  When the 1999 District Court ruled accepting the expert’s conclusions on the "educational benefit" of diversity, the error in that strategy became evident.

From 1999 to as late as March 2003, various independent scientists critiqued the methodology in their own amici briefs, but none had access to the a full dataset to do a complete “peer review” or test other hypotheses that the data might have supported. Dr. Stanley Rothman, Dr. Tom Wood, and others lead this effort; quite independently of the Plaintiffs. It should be remembered that the Plaintiffs only represented two people; but upon those few attorneys and the interests of those two people hinged the interests of an entire public. These methodology critiques exposed the multiple mythologies of diversity science (for example, that political support for environmental policies equated to a “democracy outcome,” and that diversity programs increase such support).

A particularly bad piece of U-M's scientific puzzle was that of Patricia Gurin, a psychology professor and even an interim Dean of the LS&A, U-M's largest sub-unit. At the time, even the defense team admitted this particular piece alone held with it the scientific success or failure of their claims.  Gurin's work was never "peer reviewed," and indeed, no one outside U-M's circle has ever received the data to do such. The later critiques, lacking the raw underlying data to make the definitive counterargument, created a “he said, she said” stalemate - a nebulous cloud that supporters could grab onto and that opponents could see through.  Even Justice O’Connor couldn’t “embrace” this science, to arrive at her preferred conclusion. She could only give “deference” to it - since the Plaintiffs could neither prove nor disprove it. It is that “deference” that is so dangerous. The previous standard was that once a Plaintiff proved a constitutional violation, the burden shifted to Defendant, under “strict scrutiny,” to prove a compelling need for the violation.

Too little, too late, the final questions raised about U-M’s diversity science came from the inside. Revealed only after renewed prying, I received a FOIA response from Michigan answering as many questions as it raised. In May, after a month and a half of due diligence and tracking different parts of the story down, the Wall Street Journal published my op-ed analysis revealing that Patricia Gurin’s own husband had a written a little-circulated report in 1994. The 1994 conclusions, written without the bias of pending litigation, revealed a “mixed bag” of positive and negative outcomes from diversity policy.  This “mixed bag” included findings that diversity by itself doesn’t change educational outcomes, that blacks self-segregated (defeating one of diversity’s goals), and that there was increasing racial polarization over time at U-M. Gerald Gurin, a now-retired psychology professor as well, presented an inconsistent picture from the same dataset Patricia cited in her 1999 expert testimony.  In that testimony, she remarked how “unusual” it was for the “pattern of consistency” in the conclusions to she saw in data.  Only after the Journal piece did she admit that she only looked at half of the data; and ignored the “inconsistent” stuff.  This was all perfectly rational though, so she argued. Those "other" parts simply were irrelevant to her particular hypotheses, as she naturally defined and redefined it.  To question the "science" given that obvious fact was missing the point.  To this day, Gurin and U-M officially deny anyone in the public whose goals "clash" (unbelievably, there's actually a document that says this) with their goals access to the 'other half' of the dataset.

Its not “unusual” to have “consistent” conclusions when you pick and choose your data.  This may not “prove” perjury or scientific fraud (Gurin's own subconscious desires may have lead her down this road), but it certainly creates substantial doubt about the manipulation or “good faith” of this science. By late May, there was no practical way to get this information before the Supreme Court, and Michigan was refusing still to produce the raw data under state FOIA anyway (which would have been the only “final answer”).  The ultimate hidden conclusion of (Gerald) Gurin’s first work - that economic differences outweighed racial ones. Despite her past support for the idea, O’Connor foreclosed any incentive U-M might have had to use what appears to a be an internally justified socio-economic race-neutral alternative.

What can be done now? Congress should tighten rules for expert testimony, requiring anyone testifying about scientific data used to arrive at their conclusions to release such data to anyone who asks for it (the public often has different interests than the adversaries in litigation, which is expressed also through our amicus briefing process).  Currently, only Plaintiffs have that right, and in this case Plaintiff either missed the opportunity or misjudged the importance of fighting for the data. Indeed, any social science used in lobbying or other policy change efforts should be subject to the same full disclosure requirements. Such a policy could have a wide-ranging and positive effect on areas of law and public concern as diverse as the effects of tobacco or global warming. Simply, if you use the science, you must disclose the science. Such legislation would not extend to the marketplace, and therefore would not affect private companies wishing to use science for purely private purposes. Congress or the executive could even condition the receipt of federal research money on acceptance of disclosure requirement (for example, OMB could add this to their other research disclosure requirements).  Individual state legislatures should follow the same lead.

O'Connor's majority opinion, in arriving at the conclusion that the University of Michigan Law School may constitutionally use an individualized racial preference program, took a few logical turns that were actually unnecessary to her conclusion.  Aside from the “moderate” alternative of embracing socio-economic alternatives as a compromise, O'Connor's conclusion that "diversity" was a "compelling interest" could have been arrived at purely on political grounds. If she felt any need to rely on "educational benefits," she could simply have embraced the science after “strictly scrutinizing” it (given the poor quality of the science, this certainly would have required some other convolutions we can’t even imagine, and O’Connor obviously wouldn’t go that far). Hopefully, when some bright university attorney tries to use "social science" and the Court's newly fashioned "deference" to justify another form of Constitutional violation, the Court will find in the confusing language of this decision a way to narrow its mistake. By at least continuing ‘lip service’ to “strict scrutiny,” and writing an otherwise horribly convoluted opinion, with its own built-in timer; the precedent may not be firmly set, and this case may just be an aberration.

Chetly Zarko is a freelance writer in Ann Arbor, Michigan.  He has been published in the Wall St. Journal, The Michigan Bar Journal, and a variety of other local and national publications.  His work has been cited by Nat Hentoff, Linda Chavez, Thomas Bray, on the Hannity and Colmes (FOX-TV) show, and by a wide array of news and public interest organizations. He graduated from the University of Michigan with honors in 1993.