This month, the Supreme Court will hear argument in Fisher v. University of Texas at Austin, a case that is widely expected to end race-based affirmative action in higher education. A decade ago, Grutter v. Bollinger upheld that practice, holding that public universities and colleges could lawfully include race as one factor in admissions decisions to foster racial diversity on campus. At the time, the Court speculated that such diversity would be achieved in twenty-five years and that race-based affirmative action thus would no longer be necessary in 2028. The Roberts Court now appears ready to ditch the practice much sooner.
This prospect should come as no surprise. Vulnerable from the start, Grutter was the product of a deeply divided Court and has lacked majority support among the Justices ever since its author, Justice O'Connor, retired in 2005. Since then, the Roberts Court has voiced its hostility to race-based criteria in a host of contexts and has also repeatedly shown its willingness to displace precedent it dislikes. Adding the fact that Fisher contains none of the characteristics that typically justify Supreme Court review and it looks like a safe bet that the Roberts Court did not take the case to affirm the wisdom of diversity-seeking affirmative action. It is far more likely that it will use Fisher as a vehicle to condemn Grutter and the type of decisionmaking it fosters.
What remains to be seen is precisely how the Roberts Court will express that condemnation. Many anticipate the Court to scrap Grutter entirely. Still, overruling the case is not the only means by which the Court might voice its objections to the Grutter framework. In fact, recent decisions show the Roberts Court responding in three very different ways when confronted with precedent it disfavors. Each presents a plausible template for resolving Fisher.
I. Giving Notice: The NAMUDNO Template
Most narrowly, the Court might decide to use Fisher to issue a warning, much like it did in 2009 when it sidestepped a constitutional challenge to the Voting Rights Act (VRA). Northwest Austin Municipal Utility District Number One v. Holder (NAMUDNO) addressed a provision of the VRA that requires jurisdictions with a history of voting discrimination to obtain federal approval prior to changing any aspect of their voting laws. The Supreme Court had repeatedly upheld the provision, but by 2009 questions had arisen as to whether it was still justified. Writing for the Court, Chief Justice Roberts seemed quite skeptical that it was justified and listed reason after reason why the statutory provision appeared constitutionally infirm.
His opinion nevertheless opted not to throw out the statute. Instead, the Chief Justice simply rewrote the statute, holding that the plaintiff could apply for a statutory exemption for which it had previously seemed to be flatly ineligible.
That statutory construction, contrived as it was, not only enabled the Court to avoid a constitutional ruling, but also provided the Justices a mechanism through which to issue a warning. NAMUDNO put Congress on notice, making it clear that the Justices stand ready to scrap the statute in the next case unless something significant about the statutory regime has changed by then. In other words, the NAMUDNO Court's criticism of the VRA, while technically dicta, might be better understood as the operative holding—one that strikes down the statute but stays the order until the next case in which the question is presented.
Were the Court to follow this template in Fisher, it would issue an opinion spelling out why the Justices think the diversity Grutter celebrated no longer provides sufficient justification for the use of race. Fisher would nevertheless stop short of overturning Grutter. Instead, it would find some alternative means to dispose of the case, with mootness being the most likely candidate. True, the University's claim that the case is moot is far from unassailable, but it is no more implausible than the statutory reading the Court constructed as an exit strategy in NAMUDNO.
Fisher might, accordingly, opt to follow NAMUDNO by outlining the Court's objections to Grutter and then disposing of the case on a distinct ground. Doing so would give notice to school administrators at the University of Texas and elsewhere that the Supreme Court presently views reliance on race in the admissions process as deeply flawed. The message would be clear: Change your programs now or be assured that the Court will change them for you in the next case.
School administrators might respond by scaling back the ways in which they use race to promote diversity. Were they to do so, the Court would have avoided the need to issue a divisive constitutional ruling. But even if they opt to stay the course, much like Congress has done with regard to the VRA, the NAMUDNO template might still prove an attractive one to use in Fisher.
First, NAMUDNO postponed resolution of the constitutional question and thus allowed the VRA's regional provisions to remain operational for an additional redistricting cycle. That cycle, in turn, has yielded a host of developments that illuminate the statute's operation in ways that may prove beneficial to the Court when the Justices revisit the issue. Second, NAMUDNO provided a measured first response to a contentious issue that promises to temper the charges of activism that will inevitably follow should the Court ultimately decide to strike down the provision.
Following the NAMUDNO template in Fisher might yield similar benefits even if school administrators leave existing affirmative action plans unaltered. Allowing Grutter more time to operate may expand our understanding about the administration and operation of diversity-promoting programs. Also, reliance on the NAMUDNO template might lessen charges of activism should the Court ultimately decide to displace Grutter as precedent.
Still, critical differences between the two cases make the Roberts Court unlikely to use the NAMUDNO template in Fisher. NAMUDNO presented a constitutional challenge to a highly salient, albeit poorly understood, federal statute; one that the Court had repeatedly sustained and expressly preserved in the very line of cases that presently render the VRA so vulnerable. The Roberts Court no doubt felt inclined to tread lightly, as NAMUDNO itself reflects. The Justices are apt to feel less constrained in Fisher. Diversity-based affirmative action in higher education has a very different pedigree than the VRA, and the Court has already specifically aired and addressed the precise legal arguments supporting and opposing the practice. As a result, the Roberts Court may feel more freedom to rule expansively in Fisher than it did in NAMUDNO.
II. Overruling Grutter: The Citizens United Template
Ruling most expansively, of course, would mean that the Court would do what it is widely expected to do in Fisher: use the case as the vehicle to overrule Grutter. Should the Court follow this course, it would disavow Grutter's core holding that "student body diversity is a compelling state interest that can justify the use of race in university admissions." Fisher would hold that fostering diversity is not a compelling interest and hence not sufficiently weighty to justify the use of race-based criteria to achieve it. That is, Fisher would bar school administrators from considering a student's race when making admissions decisions by rejecting the very interest the Court upheld in 2003 as sufficient to justify the admissions policy under challenge.
The Roberts Court has issued a decision of this structure previously. Two years ago in Citizens United v. Federal Election Commission, the Court disavowed a governmental interest it had only recently upheld as sufficient justification for a challenged regulation. At issue in Citizens United was the application of a federal statute that limited certain forms of corporate speech in advance of an election. Back in 2003, McConnell v. FEC upheld the statute, finding that Congress's interest in preventing both corruption and the more inchoate "appearance of corruption" provided adequate justification for it. Citizens United, however, announced that Congress's legitimate regulatory interest was far more limited and extended only to the prevention of so-called "quid pro quo corruption." Finding that the disputed statute did nothing to advance this more circumscribed interest, Citizens United struck down the statute as unconstitutional and overturned portions of McConnell that held otherwise.
By renouncing an interest that was recently deemed sufficient to justify a challenged regulation, Citizens United offers a template for resolving Fisher should the Roberts Court now be inclined to disavow diversity. Just as Citizens United deemed the governmental interest in curbing influence—as opposed to outright bribery—insufficient to justify the challenged limits on corporate speech, the Fisher Court might similarly discard diversity as a governmental interest sufficient to warrant the race-based decisionmaking Grutter sanctioned.
Admittedly, the structural connection between what Citizens United did and what Fisher might do is obscured by the distinct subject matter addressed in the two cases. But the connection is a real one. Indeed, the widespread belief that Fisher will scrap Grutter necessarily posits—albeit implicitly—that Fisher will follow the Citizens United template. By virtue of existing doctrine, any ruling that bars school administrators from considering a student's race when making admissions decisions would also need to discard the interest that presently justifies such reliance. And if the Roberts Court wants to jettison diversity, it knows how to do so because it did something structurally similar in Citizens United.
Still, the Roberts Court may be more reluctant to follow the Citizens United template in Fisher than those predicting Grutter's demise generally allow. True, the personnel changes that enabled the Citizens United Court to overrule McConnell so quickly also suggest a Court prone to scrap Grutter in a similar fashion. And yet, the extensive and ongoing criticism Citizens United generated may have tempered the Justices' enthusiasm for projects of this sort.
Even if it remains undeterred, the Fisher Court may reject the Citizens United template simply because it lacks the votes to overturn Grutter using it. In particular, Justice Kennedy's support for doing so is far from certain. He has, to be sure, repeatedly objected to the ways in which public decisionmakers use race in a host of contexts, including his dissent in Grutter. But even as he has done so, Justice Kennedy has been careful to avoid condemning the goals that underlie such projects. Thus, in his Grutter dissent, Justice Kennedy called for more exacting scrutiny of the process by which "a university's compelling interest in a diverse student body [may] be achieved." Four years later, he emphasized that, "[d]iversity, depending on its meaning and definition, is a compelling educational goal a school district may pursue," even as he voted to strike down the race-based policies two school districts used to assign students to particular schools.
Needless to say, such statements do not preclude Justice Kennedy from now voting to overturn Grutter using the Citizens United template. He may now be convinced that promoting diversity is a flawed project and doomed to fail regardless of the means selected to promote it. But were Justice Kennedy to so hold, he would need to disavow not only Grutter's recognition that diversity constitutes a compelling interest, but also his own statements to that effect and the distinct doctrinal path his own opinions have charted.
III. Policing Means, Not Ends: The Kennedy Template
In a series of opinions, Justice Kennedy has mapped out a third template the Court might follow in Fisher. His dissent in Grutter v. Bollinger, his concurring opinion in Parents Involved in Community Schools v. Seattle School District No. 1, and his recent opinion for the Court in Ricci v. DeStefano all demand that the Court closely police the means by which public officials use race in decisionmaking processes. These opinions nevertheless either explicitly affirm or simply leave undisturbed the recognized legitimacy of the specific goal advanced by the challenged race-based criteria.
In 2003, Justice Kennedy's dissent in Grutter made it clear that he viewed the flaw in the Michigan program to be one of means, rather than ends. While he joined Chief Justice Rehnquist's dissent—and thus agreed that the law school's program involved more rigid racial balancing than the majority had allowed—Justice Kennedy separately acknowledged the legitimacy of the underlying goal, noting both "a university's compelling interest in a diverse student body" and the established bases for "the Court's acceptance of a university's considered judgment that racial diversity among students can further its educational task." 
Five years later, Justice Kennedy again parsed means from ends in Parents Involved in Community Schools v. Seattle School District No. 1. In dispute were programs implemented by two school districts to increase racial diversity and avoid what school administrators called racial isolation. In a 5–4 ruling, the Court held that the programs, which assigned students to particular elementary and secondary schools based on their race, amounted to unconstitutional racial balancing. Justice Kennedy provided the critical fifth vote for this ruling, but he did not join Chief Justice Roberts's majority opinion in full. Instead, he wrote separately to explain why he believed the student assignment plans were flawed. One plan, he wrote, had been implemented haphazardly with what he saw as nebulous standards and inadequate oversight, while the other plan relied on racial categories so broad that it threatened to undermine the very diversity it was seeking to achieve. In other words, Justice Kennedy objected to the means by which school administrators sought to achieve their goals, rather than to the goals themselves.
Justice Kennedy, in fact, explicitly stated that he believed those goals were lawful. He wrote: "Diversity, depending on its meaning and definition, is a compelling educational goal a school district may pursue." He added that school administrators could permissibly use race-based criteria in carefully tailored ways to address problems like "de facto resegregation in schooling" and "racial isolation in schools." For Justice Kennedy, the problem in Parents Involved was not that school districts were pursuing flimsy or even destructive goals, but instead that they failed to tailor their programs adequately to achieve the worthwhile ends they sought.
Justice Kennedy again invoked this distinction between means and ends in his opinion for the Court in Ricci v. DeStefano. In this case, city officials in New Haven had claimed that Title VII's disparate impact provision required that they scrap a promotional exam for firefighters after test results revealed that no African-American firefighters would have been promoted. Justice Kennedy's majority opinion disagreed with this claim, holding that New Haven lacked a sufficiently sound basis upon which to conclude disparate impact liability would have followed had it certified the test results.
Justices Alito, Thomas, and Scalia agreed, but added that no amount of evidence would have sufficed to justify the city's decision because city officials had been pursuing an unlawful end. Justice Alito believed city officials simply wanted to placate a vocal racial constituency and he condemned as impermissible the racial pandering he understood to be animating the city's decision. For his part, Justice Scalia objected categorically to Title VII's disparate impact inquiry, suggesting that the race consciousness that unavoidably inheres in it is irreconcilable with statutory and constitutional commands regarding equal treatment. Justice Kennedy, however, held the Court to a narrower ruling by finding error in the means by which the city made its decision (namely the insufficiency of the evidence) but decidedly not in the project it was pursuing.
Taken together, Justice Kennedy's opinions in Grutter, Parents Involved, and Ricci suggest a template the Court might use in Fisher. The Kennedy template would require the Court to examine with rigor the distinct ways administrators at the University of Texas (UT) presently use racial criteria to pursue their goal of racial diversity on campus. It would nevertheless leave undisturbed Grutter's recognition that the goal of racial diversity is a compelling objective that school administrators may lawfully pursue.
Under the Kennedy template, two aspects of the Texas program are particularly vulnerable. The first is the targeted way UT administrators use racial criteria to promote racial diversity not only in the entering class, but also at the classroom and program level. The second is the decision to use Grutter's "holistic review" in conjunction with its 10% plan, under which UT guarantees admissions to all students who graduate in the top 10% of their high school class. Both aspects of the Texas program raised concerns in the lower courts.
These components of the UT program look vulnerable under the Kennedy template. While a strong argument exists that the benefits of diversity identified in Grutter require that it extend to the classroom and program level, the means by which such diversity is achieved involves targeted racial moves of a sort likely to displease anyone employing the Kennedy template. Recall Justice Kennedy's discomfort with what he believed was an unduly rigid use of race in Grutter, and again in Parents Involved and it is easy to envision the charge of racial balancing in an opinion striking down UT's effort to achieve classroom-level diversity.
So too, under the Kennedy template, the Court would likely look skeptically at UT's decision to employ Grutter-style review in conjunction with its 10% plan. The likely objection here would not be the arguable rigidity with which UT is using racial criteria, but instead the claim it needs to use race at all. True, UT has a solid argument that, standing alone, the 10% plan offers an inadequate substitute for the flexibility and individualized attentions promised by Grutter's holistic rule. And yet, the fact that UT admits a greater proportion of minority students under the 10% plan than outside of it will fuel concern that the use of race is not necessary to achieve the diversity of the sort Grutter protects.
Back in Parents Involved, Justice Kennedy complained that the school districts' race-based policies did little to foster the diversity school administrators sought. In Fisher, one can imagine him raising a similar complaint. If the facially race-neutral 10% plan more effectively fosters diversity than a race-conscious one, why not rely on, or expand if necessary, the 10% plan rather than employ less effective racial criteria to achieve the same purpose? Or so an opinion relying on the Kennedy template might posit.
Such an opinion would overrule the lower court and strike down UT's program. But it would leave standing Grutter's recognition that diversity is a compelling interest and the specific means Grutter held could be used to achieve that diversity. Individualized holistic review of the sort once used at the University of Michigan Law School would remain a constitutionally permissible means to admit a racially diverse incoming class. Nothing in Fisher would be to the contrary, as the decision would be limited to the ways in which the UT program exceeds the bounds of what Grutter had approved.
It is worth remembering that Abigail Fisher's petition to the Supreme Court made precisely this point. While ultimately calling for Grutter to be overruled, the petition focused more on the ways in which UT's program goes beyond the boundaries of what Grutter itself deemed permissible. Reliance on the Kennedy template would likely lead the Court to agree.
Undeniably, such a decision would be far-reaching, both in Texas and beyond. The Kennedy template is a sweeping one, and appears a moderate stance only when compared with the Court's approach in cases like Citizens United. Application of the Kennedy template in Parents Involved and Ricci left considerable destruction in its wake, sharply circumscribing local discretion to consider race in a host of critical decisions. And yet reliance on the Kennedy template in Fisher would yield a ruling that would be less transformative than most anticipate. It would not eliminate affirmative action entirely, which Fisher is widely expected to do. Indeed, school administrators would remain free to consider race in admissions precisely so long as they do so the way administrators at the University of Michigan Law School once did and avoid the distinct practices used at the University of Texas.
The Roberts Court will likely use Fisher as a vehicle to criticize affirmative action. It may, as many expect, take the opportunity to overrule Grutter entirely. Citizens United provides the template for doing so—instructions, so to speak, to disavow a governmental interest only recently upheld as sufficient justification for a challenged regulation. But overruling Grutter is not the only way to resolve Fisher. The NAMUDNO template offers a more measured response to a contentious issue and a forum in which to voice criticism and provide guidance. So too, the Kennedy template's focus on means rather than ends allows for a more targeted response than does Citizen United and provides the means to excise what the Court may find most objectionable about UT's practice.
As is true with many cases, the outcome in Fisher appears to lie with Justice Kennedy's vote. To be sure, the same was said about the fate of the Affordable Care Act, which, it turns out, survived constitutional scrutiny despite Justice Kennedy's vehement opposition to the statute. That Fisher might similarly surprise is possible, but that result is highly unlikely. Far more probable is that Fisher will find fault with the affirmative action program at UT with Justice Kennedy providing the deciding vote as to why. What proves to be the fatal flaw remains to be seen.
Prior precedent makes clear that Justice Kennedy is no fan of affirmative action and he may decide the time has come to scrap the practice entirely. But that outcome is hardly certain, for reasons Justice Kennedy himself has articulated. His opinions evince no enthusiasm for race-based decisionmaking of any sort, but neither do they suggest a willingness to clear the deck entirely.
And it is this caution, displayed most decisively in Parents Involved, that may wind up restraining the Court now. In that case, it was Justice Kennedy who insisted that the plurality's mantra that "[t]he way to stop discrimination on the basis of race is to stop discriminating on the basis of race," was "not sufficient to decide these cases." And it was Justice Kennedy who described rigid color-blindness as too unyielding a practice. The Government, Justice Kennedy wrote, has a "legitimate interest . . . in ensuring all people have equal opportunity regardless of their race." Time will tell whether he still thinks so.
* Professor of Law, University of Michigan Law School.
 Id. at 343.
 See infra Parts II & III.
 See Adam Liptak, Justices Take Up Race as a Factor in College Entry, N.Y. Times, Feb. 21, 2012, http://www.nytimes.com/2012/02/22/us/justices-to-hear-case-on-affirmative-action-in-higher-education.html?pagewanted=all&_moc.semityn.www (reporting that "[b]oth supporters and opponents of affirmative action" viewed the Court's decision to review Fisher "as a signal that the court's five more conservative members might be prepared to do away with racial preferences in higher education.") (link); Jeffrey Toobin, The Other Big Supreme Court Case, New Yorker, May 1, 2012, http://www.newyorker.com/online/blogs/comment/2012/05/the-other-big-supreme-court-case.html (link).
 NAMUDNO, 557 U.S. at 201–04 (noting the "substantial 'federalism costs'" Section 5 exacts, its broad application to all electoral changes "however innocuous," the fact that "[t]hings have changed in the South," that the racial gap in voter registration and turnout rates is diminished and in places nonexistent, that minority candidates hold elected office "at unprecedented levels," that "[b]latantly discriminatory evasions of federal decrees are rare," the dated character of the coverage formula, its weak relation to current conditions, and the fact that the distinct burdens imposed on covered jurisdictions "may no longer" be warranted).
 The Voting Rights Act allows a "political subdivision" to seek this exemption, but defines a "political subdivision" in terms that facially exclude the plaintiff in NAMUDNO. See 42 U.S.C. § 1973b(a) (link). The Austin water district was neither a county nor a state subdivision "which conducts registration for voting" when the county does not. See 42 U.S.C. § 1973l(c)(2) (defining "political subdivision" under the VRA) (link).
 See Ellen D. Katz, From Bush v. Gore to NAMUDNO: A Response to Professor Amar, 61 Fla. L. Rev. 991, 998 (2009) (link); Ellen D. Katz, Leave It Up to Congress, Nat'l L.J. (Apr. 13, 2009), http://www.law.com/jsp/nlj/PubArticleNLJ.jsp?id=1202429774549.
 As they are sure to do. See, e.g., Shelby Cnty. v. Holder, 679 F.3d 848 (D.C. Cir. 2012), petition for cert. filed, 81 U.S.L.W. 3064 (U.S. July 20, 2012) (No. 12-96).
 See, e.g., City of Boerne v. Flores, 521 U.S. 507 (1997) (distinguishing the statute under challenge from the regional provisions of the VRA).
 Grutter v. Bollinger, 539 U.S. 306 (2003).
 Id. at 325.
 See 2 U.S.C. § 441b (2006) (barring corporations and unions from using their general treasury funds to make independent expenditures for speech that is an "electioneering communication" or for speech that expressly advocates the election or defeat of a candidate) (link); see also id. § 434(f)(3) (defining electioneering communication as "any broadcast, cable, or satellite communication" that "refers to a clearly identified candidate for Federal office" and is made within thirty days of a primary election) (link).
 130 S. Ct. at 910.
 Id. at 910–11.
 See, e.g., Nick Baumann, Grayson: Court's Campaign Finance Decision "Worst Since Dred Scott", Mother Jones (Jan. 22, 2010, 5:00 AM PST), http://motherjones.com/mojo/2010/01/grayson-courts-campaign-finance-decision-worst-dredd-scott (link); Bob Edgar, Op-Ed, Supreme Court's Campaign Ruling: A Bad Day for Democracy, Christian Sci. Monitor (Jan. 22, 2010), http://www.csmonitor.com/Commentary/Opinion/2010/0122/Supreme-Court-s-campaign-ruling-a-bad-day-for-democracy (link); Press Release, Org. for Security and Co-operation in Europe, Head of OSCE Election Body Concerned About U.S. Supreme Court Ruling on Election Spending (Jan. 22, 2010), http://www.osce.org/odihr/elections/51838 (link).
 See Grutter v. Bollinger, 539 U.S. 306, 387–95 (2003) (Kennedy, J., dissenting); see also Rice v. Cayetano, 528 U.S. 495, 517 (2000) ("One of the principal reasons race is treated as a forbidden classification is that it demeans the dignity and worth of a person to be judged by ancestry instead of by his or her own merit and essential qualities.") (link).
 Grutter, 539 U.S. at 392 (Kennedy, J., dissenting).
 Grutter, 539 U.S. at 387–95 (Kennedy, J., dissenting)
 551 U.S. at 782–98 (Kennedy. J., concurring).
 Grutter, 539 U.S. at 392 (Kennedy, J., dissenting).
 Id. at 387–88.
 Parents Involved, 551 U.S. at 732–35.
 Id. at 784–87 (Kennedy, J., concurring).
 Id. at 783.
 Id. at 788.
 557 U.S. 557 (2009).
 Id. at 585.
 Id. at 598–605 (Alito, J., concurring).
 Id. at 594–96 (Scalia, J., concurring).
 Id. at 585.
 See Fisher v. Univ. of Tex. at Austin, 644 F.3d 301 (5th Cir. 2011) (Jones, J., dissenting from the denial of the petition for rehearing en banc) (link); Fisher v. Univ. of Tex. at Austin, 631 F.3d 213 (5th Cir. 2011) (Garza, J., specially concurring) (link).
 See, e.g., Fisher, 631 F.3d at 239.
 See Parents Involved in Cmty. Sch. v. Seattle Sch. Dist. No. 1, 551 U.S. 701, 788 (2007) (Kennedy, J., concurring).
 Id. at 787–88.
Copyright 2012 Northwestern University
Cite as: 107 Nw. U. L. Rev. Colloquy 140 (2012), http://www.law.northwestern.edu/lawreview/colloquy/2012/15/LRColl2012n15Katz.pdf.
Persistent URL: http://www.law.northwestern.edu/lawreview/colloquy/2012/15/