SOME FINAL THOUGHTS ON THE AFFIRMATIVE ACTION RULING,AND RELIANCE IN A CHANGING LEGAL WORLD: Part Three Of A Three-Part Series on Precedent

In Part One and Part Two of this series, my brother, Akhil Reed Amar, and I have been discussing what role precedent should play on the Supreme Court. This is sure to be a key issue this Term, as the Justices revisit two noteworthy cases of the past quarter century, the Bakke affirmative action ruling, and the Bowers v. Hardwick sodomy decision. Akhil is on vacation this week, so I will use our space here to tie up some loose ends.

Our general view is that the Rehnquist Court's articulated theory of stare decisis tends to improperly elevate judicial doctrine over the Constitution itself. It does so by requiring excessive deference to past decisions that themselves may have been misinterpretations of the law of the land.

Nonetheless, in our last column we acknowledged and analyzed how reliance by major societal institutions on Supreme Court decisions could counsel against overruling those decisions. Such a reliance argument, we believe, is quite plausible as to Bakke, but far less plausible as to Bowers In this Part Three of the series, I will discuss in greater depth whether a reliance argument, in the end, can work for Bakke.

Why the Reliance Argument Works Better For Bakke Than For Bowers

Thousands of admissions committees and administrators throughout the country at colleges and universities - both public and private - have for the past 25 years tailored their affirmative action policies around Justice Powell's nuanced opinion in Bakke.

There, Justice Powell expressed the view that race may be used as a factor in admissions -- so long as the program does not mandate racial quotas, and so long as factors other than race are similarly being considered, in an individualistic assessment of how an applicant may contribute to the diversity of the student body. And colleges and universities have largely followed this view in their policies.

If the Supreme Court, in the pending University of Michigan cases, were to throw out these policies today, institutions of higher education would find it hard to turn on a dime. Even if there are ways other than using race as a factor for schools go about integrating their student bodies, such alternative routes take many years to explore and successfully pursue.

Bowers - the 1986 case permitting, over a due process challenge, Georgia to criminalize sodomy - is a different matter. Whereas a vast educational apparatus has accreted around Bakke, there are no major institutions in society that are fundamentally organized around Bowers, and that could not be drastically changed in an instant to accommodate an overruling.

The First Issue In the Bakke Reliance Inquiry: What Did Bakke Really Hold?

To say that the reliance argument in Bakke is credible is not, of course, to say that it is necessarily a winner. A number of specific issues need to be explored before reliance can carry the stare decisis day as to Bakke.

First, we must address the contention that Powell's opinion in Bakke simply expressed the views of a single Justice and said nothing meaningful on behalf of the Court as a whole.

This, for instance, was the view taken in the well-known 1996 Hopwood case, in which the United States Court of Appeals for the Fifth Circuit struck down the University of Texas Law School's race-based affirmative action plan. If Justice Powell wrote only for himself, the Hopwood argument runs, then any reliance by admissions departments and others is unreasonable and thus unworthy of protection.

It is true that much of Powell's writing in Bakke was not joined by any other Justice - a great deal of his discussion of diversity was written for himself alone. But it is also true that five Justices - the Brennan Four (Brennan, White, Marshall, Blackmun) as well as Powell himself - signed onto Part V-C of Powell's opinion. The question, then, is how to interpret the Brennan Four's views, as expressed in the combination of Part V-C of Powell's opinion, and the other opinion the Four joined.

In that separate opinion, the Four advocated an alternative theory of permissible race-based affirmative action grounded not on Powell's rationale - diversity - but rather on the separate rationale of remedying past violations of racial equality. Meanwhile, in Part V-C, the Four signed on to the following part of Powell's language (included here in its entirety):

In enjoining [the Davis Med School] from ever considering the race of any applicant, however, the courts below failed to recognize that the State has a substantial interest that legitimately may be served by a properly devised admissions program involving the competitive consideration of race and ethnic origin. For this reason, so much of the California court's judgment as enjoins [Davis] from any consideration of the race of any applicant must be reversed.

This language was embedded in a larger opinion by Powell that embraced the diversity theory, though obviously, it does not mention "diversity" per se. And that leads to a key question: Did the Brennan Four intend to accept Powell's diversity approach, in addition to their own, more permissive approach, as an allowable justification for affirmative action?

There is good reason to believe that they did. To begin with, note that the Brennan Four, in their own discussion of permissible remedial affirmative action, never required "competitive consideration of race"; for the Brennan Four, quotas as well as plusses were permissible. Indeed, "competitive consideration" seems to be code for a "plus system." After all, only in a plus system do white and minority candidates compete directly; in a quota system, whites cannot compete for a certain spots in the entering class reserved for minorities.

Accordingly, the fact that the Brennan Four signed onto Powell's V-C language requiring "competitive consideration" strongly indicates that they meant to allow affirmative action plans that met Powell's more stringent test as well as those programs that satisfied their own, more lenient, test. Why mention "competitive consideration" if you do not endorse the "plus" system that embodies that very type of consideration?

Moreover, the Brennan Four spoke the language of diversity themselves in parts of their own opinion. They pointed out, for example, that the Davis program "does not establish an exclusive preserve for minority students apart from and exclusive of whites. Rather, its purpose is to overcome the effects of segregation by bringing the races together." In other words, government may permissibly use race to integrate rather than segregate, to create a common citizenship rather than a two-class society. Those, of course, are the very goals of diversity - to bring different races together, so that an exchange of views, perspectives, and experiences can occur.

There is thus good reason to think, as most of America has, that Powell's embrace of the "Harvard Plan" approach - the approach, that is, that uses plusses but not quotas for the goal of accomplishing diversity - did indeed garner the approval of a five-justice majority of the Supreme Court.

Is Reliance Undermined Due to The Court's Later Retreat From Diversity?

Even if that is so, opponents of Bakke point to a number of Supreme Court decisions, issued over the past decade and a half, that seem to call into question affirmative action laws and programs that make use of race. These decisions, Bakke detractors say, undermine reliance on Powell's opinion: They should have functioned as a wake-up call, on this view, to any college or university that thought race-based admissions remained acceptable.

As a general matter, reliance is less justifiable (and thus less worthy of protection) if the Court has sent clear signals that it is planning to overrule an earlier decision. Moreover, to the extent that post-Bakke cases really are inconsistent with Bakke itself, then the stare decisis issue could be said to cut both ways.

That is, the Court cannot respect stare decisis and follow Bakke without disrespecting stare decisis in ignoring more recent rulings. Something must give: Either Bakke must be overruled, or the more recent rulings must be overruled (or, what can amount to the same thing, "confined to their facts.")

To put the point slightly differently, whenever two existing cases conflict, stare decisis has already been ignored (whether explicitly or not) by the Court in the second case. And when a third case comes along requiring a choice between the two earlier decisions, adherence to stare decisis does not by itself easily resolve this choice.

The Contracting Cases: Do They Really Undermine Reliance on Bakke?

In fact, however, the post-Bakke cases do not really undermine reliance on Bakke because they do not necessarily conflict with the essence of Bakke as I have described it.

Take, for starters, the 1989 Croson and 1995 Adarand decisions. In both, the Court spoke disparagingly about contracting set-aside programs in which public dollars are directed towards minority-owned business enterprises (MBEs). Are opinions about set-asides for MBEs really in tension with a decision about affirmative action in education?

To my mind, the answer is no. Indeed, these cases have little to do with educational diversity. Generally, the theory behind MBE programs has been remedial - not diversity-based. Moreover, MBE programs, unlike the educational diversity program at issue in Bakke and the pending University of Michigan cases, do not necessarily promote integration, for several reasons.

First, there is no requirement that MBEs who receive public monies themselves have integrated workforces. Nor is there a requirement that MBEs work closely and exchange ideas and values with non-MBEs. By contrast, diversity-based programs in education by their nature bring people of different races together, and at a time in their lives when they can most benefit from interaction.

Also, MBE programs, because they involve large sums of money, may be more susceptible to corruption than educational diversity programs. Some MBEs may be minority-owned only on the books, but not in reality.

In contrast, opportunities for fraud or sham in educational admissions, while not absent altogether, are greatly reduced by the personal interaction between the applicant and the university. Applicants must submit personal essays, and often must also attend interviews in the course of the admissions process.

Finally, the MBE programs shut down by the Court have themselves been quota programs, which do violence to the core of Justice Powell's opinion.

For all these reasons, claims that cases like Croson and Adarand undermine Bakke's endorsement of the diversity rationale are dubious indeed. To put the point most simply, Croson and Adarand are not about diversity, and so they are not on point.

The Role Model Theory Case: Does It Undermine Reliance on Bakke?

Another decision relied on by Bakke opponents is Wygant, a 1986 case where a school district had laid off white teachers in order to retain teachers of color who could serve as "role models" for students of color.

The Wygant Court rejected the school district's approach. But in doing so, it did not call into question Justice Powell's diversity theory. Indeed, Powell himself wrote the plurality opinion in Wygant, without signaling any retreat from his writing in Bakke.

Moreover, as Justice O'Connor recognized in Wygant, the role model theory and the diversity theory are quite different. Indeed, in some ways, the two approaches are at odds: The role model theory presupposes that minority students need to interact with, and learn from, minority teachers. In contrast, the diversity theory starts from the integrationist premise that people need to interact with, and learn from, people of races other than their own.

Thus, like the MBE program struck down in Croson, the role model theory rejected in Wygant did not necessarily integrate society, a key selling point of Powell's educational diversity theory. So even putting aside the fact that Wygant involved laying off employees rather than declining to admit students (a key distinguishing factor for Powell himself), Wygant tells us very little about educational diversity.

Does the Case on Broadcasting Diversity Undermine Reliance on Bakke?

One additional case - Metro Broadcasting v. FCC - may be quite important. That 1989 case did involve diversity, of sorts - unlike Croson, Adarand, and Wygant.

In Metro Broadcasting, the Court upheld, by a 5-4 vote, the FCC's policy of taking the race of a license applicant into account in order to promote diversity in broadcast programming. Metro Broadcasting is thus a kind of hybrid contracting/diversity case.

At first blush, opponents of Bakke should find little support in a decision that upheld a race-based program designed to promote viewpoint diversity. But Bakke bashers would quickly point out that four dissenting Justices in Metro Broadcasting questioned the FCC's diversity premise altogether. And that group of four Justices - Chief Justice Rehnquist, Justice Scalia, Justice Kennedy, and Justice O'Connor -- joined the next year by Clarence Thomas, who replaced Thurgood Marshall, currently seems to be controlling the direction of race cases at the Court. Now, with Thomas's vote, they have the five needed to convert their dissenting views to majority views.

Metro Broadcasting is particularly important because the dissenters were led there by Justice O'Connor, the crucial swing vote in these matters. In Metro Broadcasting, Justice O'Connor denied that the race of the license owner necessarily dictated the kind of programming he or she would offer. Specifically, she excoriated the FCC policies as a "direct equat[ion of] race with belief and behavior, for they establish race as a necessary and sufficient condition [for] securing the preference."

Was Justice O'Connor, in Metro Broadcasting, clearly signaling a rejection of Powell's approach in Bakke? If so, Bakke would indeed be in trouble and people should rely on it much less. But I don't necessarily read Justice O'Connor's comments, here and elsewhere, that way.

First, she has repeatedly cited favorably to Powell's opinion in Bakke in the past, and has never directly called it into question. Second, and more important, her language in Metro Broadcasting and elsewhere is, when carefully parsed, completely consistent with Powell's opinion in Bakke.

Consider what Justice O'Connor's specific complaint in Metro Broadcasting is: What arguably offends Justice O'Connor is not that government uses race as a factor to predict what kind of broadcaster (or student) someone will be. It is that government may overuse race as the only factor to predict what someone's voice will be like. That is why she targets the "direct equation" of race and belief, as if they were interchangeable. That is also why she objects to making race "a necessary and sufficient condition" - or, in other worlds, effectively the only condition.

Remember, the Powell approach in Bakke permits race to by used, but only as one factor in a "competitive consideration." For Powell, and O'Connor as well perhaps, what government may not do is ignore the rest of the whole person by focusing predominantly on race.

But considering race alongside a host of other factors is a constitutionally different matter altogether. Thus, even O'Connor's seemingly extreme language in Metro Broadcasting does not necessarily undermine the essence of Bakke. The language pretty clearly suggests Justice O'Connor would condemn a quota system - where race is a necessary condition for certain spots in an entering class. But it leaves open what her views might be on a "plus" system designed to enhance diversity.

The Redistricting Cases Suggest Justice O'Connor May Believe in Diversity

Indeed, if I were to analyze Justice O'Connor's attitudes even more deeply, I would also include the so-called racial redistricting cases. There, the Court has placed limits on the ability of state legislatures to use race when drawing voting district lines.

Importantly, there, as elsewhere, Justice O'Connor's vote has been key. And her approach in that arena is remarkably similar to Powell's in Bakke - that race may be used, but not as the "predominant" factor that crowds out all other voter characteristics.

Which way, then, will O'Connor, and the other Justices lean in the University of Michigan cases that will decide Bakke's fate this term? It is difficult to predict. At the end of the day, the Court is going to have quite a bit to sort through, as it decides how much of Bakke remains intact today, and how much of it ought to govern the University of Michigan cases.

The stakes are high and the arguments complicated. And whatever the Court ends up doing, there will be many vocal critics who think it erred. That is the one thing on which everyone can rely.