Proof that the decision by the Village was motivated in part by a racially discriminatory purpose would not necessarily have required invalidation of the challenged decision. Such proof would, however, have shifted to the Village the burden of establishing that the same decision would have resulted even had the impermissible purpose not been considered.
—Village of Arlington Heights v. Metropolitan Housing Development Corp. (1977)
Chapter 6 explained why our most intuitive understanding of animus—as subjective ill will—does not provide a fully workable methodology for courts to uncover it. Chapter 7 introduced a set of commonsense factors, borrowed from a related constitutional law doctrine—the doctrine of discriminatory intent—and suggested that those factors could help courts in the search for the analogous phenomenon of animus.
Chapter 8 noted a limitation to that analogy: Unlike a finding of discriminatory intent, a finding of animus should conclusively decide the case against the constitutionality of the challenged government action. It made this argument as a matter of logic and conformance with the class legislation heritage of animus doctrine. But Chapter 8 also added a troubling note. It identified several of the foundational animus opinions the Court or individual justices have written as having suggested that judicial review should continue even after the court identified animus. Those calls for further scrutiny even after the Court identified animus suggest that animus is not a doctrinal “silver bullet.” In turn, that conclusion poses serious problems for our attempt to connect animus to older notions of class legislation.
Chapter 8 concluded by offering a solution to this conundrum. It may be that a court’s continued search for a legitimate foundation for an action found to have been motivated in part by animus actually reflects a search for the ultimate intention behind that action. In other words, it may be that animus really is a doctrinal silver bullet. But conclusively identifying animus may require searching for other possible explanations for the challenged law. This chapter examines how that search should be undertaken.
This argument requires us to return to our discussion of the intent inquiry in equal protection law. After all, what we are talking about in this chapter is how best to identify animus. Given that we are using the same tools as the intent inquiry (that is, the Arlington Heights factors we introduced in Chapter 7), it makes sense to consider how discriminatory intent jurisprudence actually uses those tools. Once we understand how the intent inquiry uses them, we can consider whether such uses can be directly transferred into the animus inquiry or whether we need to tweak that doctrine to account for our different context.
The intent inquiry involves two stages of argumentation and proof. First, the plaintiff is required to show, using the Arlington Heights factors, that an intent to discriminate on the alleged ground was a motivating factor. Note the italicized word in that last sentence. The plaintiff is not required to show that the alleged intent was the sole, or even the predominant, motivator of the challenged government action. Instead, the plaintiff need only show that that intent was one motivation.
But if the plaintiff makes that showing, that is not the end of the case. Instead, the burden then shifts to the government defendant to show that it would have made that same decision, even in the absence of the discriminatory intent that the plaintiff has just proven constituted part of the government’s motivation. (This part of the Arlington Heights opinion is quoted at the start of this chapter.) If the government can show that, indeed, it would have made that same decision even had its motivations been completely pure, then the court will hold that there was no intent to discriminate on the alleged ground. But if it cannot make that showing, then what we have is a case of intentional discrimination on the basis alleged by the plaintiff.
A simple example will both illustrate this idea and reveal its foundation in common sense. Recall our firefighter hypothetical from Chapter 8, in which a fire department has a rule requiring that firefighters be of a certain height. One might expect that this rule would disqualify more women than men, given the statistical reality that, on average, men are taller than women. Suppose that a woman sues the fire department, alleging that the rule “discriminates” on the basis of sex. “Discriminates” is in quotation marks because I am using that term in its legal sense. In a very basic way, the rule clearly “discriminates” against women, in that it has the effect of disqualifying more women than men. But since the Equal Protection Clause prohibits only intentional discrimination, such disparate effects are usually not enough by themselves to establish “discrimination” in the constitutional sense.1
Under standard equal protection doctrine, the female plaintiff in our hypothetical would shoulder the burden in the first phase of the intent inquiry. As explained above, her burden would consist of showing that the fire department’s height rule was motivated, in part, by an intent to classify based on sex. She would not have to show that it was the only factor motivating the rule, or even the main factor motivating it—she would just have to show that it was one of the motivations. If she succeeded, then the burden would shift to the fire department. At that point it would have been held that the department had sex at least partly in mind when it promulgated the height rule. When confronted with such a finding, the department would face a new hurdle: It would have to show that it would still have promulgated that rule even if—now, contrary to the court’s finding—it never had a sexist bone in its institutional body. In essence, the department would have to prove that height was a consideration that, independent of its correlation with sex, prompted the department to promulgate the regulation.
Consider the common sense of this burden-shifting framework.2 It would be unfairly burdensome on a plaintiff, such as the woman in this case, to prove that sex was the complete or even just the primary motivating factor in a case such as this. Even with the ability to review the department’s records, proving that sex was the only or the primary motivating factor would be difficult. More conceptually, the reality that institutions usually act based on a variety of motives makes it unfair to insist that she prove that the department acted only out of a sex discriminatory motivation. Still, because she is the plaintiff, it is fair to require that she offer at least some proof of the fire department’s intent—at least if we assume, as we are here, that intent is a constitutional requirement.3 But it is also fair to shift the burden to the fire department if the plaintiff shows that sex was at least one of the things the department was thinking about when it promulgated the rule. After all, the department is best situated to prove its own motivations.
Finally, consider the reasonableness of the ultimate conclusions this framework yields. If the fire department cannot prove that it would have made the same decision even absent the sex discrimination intent, then it is fair to say that the woman has in fact been the victim of sex discrimination.4 In that case, she has suffered a burden “because of” the department’s intent to classify on that ground. Conversely, if the department does manage to carry its burden, then the woman cannot in fairness be said to have been the victim of sex discrimination. In that latter case, the woman is simply the victim of a decision motivated by a conclusion that, in fact, fire fighters really do need to be of a certain height. In that case, she would still be the victim of height discrimination (although that would likely not help her much in court, given the constitutionally non-suspect status of height). But not sex discrimination.
Let’s now map this structure onto the animus inquiry. As applied to that inquiry, the discriminatory intent burden-shifting framework would require some explanation from the government if the plaintiff can demonstrate that animus was a factor in the challenged government action. Thus, for example, if the plaintiff can convince a court that a law’s severe disparate impact implies simple dislike or disapproval of the burdened group, or if the plaintiff can show that the government’s decisional process leading to enactment of the challenged law was explicitly infected with animus, then it might be appropriate for a court to seek an explanation from the government. In the intent context, that explanation would take the form of a government showing that, as with the fire department’s height rule, the government would have made the same decision even had it lacked the alleged discriminatory intent the plaintiff succeeded in suggesting. In the context of the animus inquiry, that explanation would take an analogous form. It would require that government demonstrate that legitimate needs motivated the government’s action.
But note something interesting about this parallel. In both situations, the burden rests with the government. In the intent context the government must demonstrate that it would have made the same decision even absent the discriminatory intent. Similarly, in the animus context the government would have to demonstrate that legitimate needs motivated the challenged action. Of course, in the intent context this placement of the burden on the government occurs before any particular level of equal protection scrutiny is performed—indeed, the entire question at this stage is whether the court actually has before it a case of, say, sex discrimination. Only after it decides that it does would the court apply the appropriate level of heightened scrutiny.
The situation is slightly different in the animus context. In the animus context, this burden shifting occurs as part of the application of the relevant scrutiny level. Do not forget that in these cases we are dealing with explicit classifications on the challenged ground—disability, sexual orientation, and so forth. Thus, a court would immediately turn to application of the relevant scrutiny level—usually rational basis, since the very reason courts need to worry about animus is that heightened scrutiny is, for whatever reason, deemed inappropriate. Once we realize this, then we realize that the burden shifting we have been talking about has to occur as part of the rational basis scrutiny itself. And when we realize, as we did a few paragraphs above, that part of that burden shifting involves placing proof burdens on the government, we can start to understand why, even under rational basis review, the Court has sometimes placed the burden of explanation on the government. For example, the Cleburne Court explicitly placed that burden on the government when it noted the lack of evidence in the record justifying the government’s argument that the group home’s population density was inappropriate for the neighborhood. Such a placement of the burden may sound inconsistent with rational basis review—indeed, Justice Marshall lodged exactly this objection in his partial dissent in Cleburne. But it makes sense, once we realize both the parallel between the animus inquiry and the intent inquiry and the lack of complete equivalence between the two.
This understanding of animus reconciles the troubling precedent we identified in Chapter 8. Recall that in a number of animus cases we have studied the Court did not stop when it encountered evidence of animus. In Moreno the Court fretted over the troubling legislative history behind the food stamp law’s definition of households (the statements about “hippies” and “hippie communes”) but nevertheless went on to consider other, more legitimate justifications for the law. In Cleburne the Court noted the city’s argument that it denied the intellectually disabled group a zoning permit because of constituents’ dislike and fear of that group, but, like Moreno, it nevertheless continued on to consider other more constitutionally palatable justifications for that denial. Finally, in Lawrence, Justice O’Connor explicitly stated that “a bare . . . desire to harm a politically unpopular group” triggered, not an immediate strike-down, but instead closer judicial review. When I summarized these facets of those cases in the last chapter, I noted the problem they posed for any theory that considered animus a per se constitutional violation.
But now that we understand how animus is found, those cases fit neatly into our theory. If those cases’ more careful scrutiny and consideration of alternative, more legitimate, justifications for laws constitute part of the search for animus itself, then we can still say that animus does indeed constitute a doctrinal “silver bullet.” But the process of finding animus requires that we give the government a chance to prove that something else was really motivating the government. Just as the intent inquiry gives the government (for example, the fire department in our hypothetical) the chance to prove that it would have taken the challenged action (that is, the height rule) even absent the alleged intent (that is, to discriminate on the basis of sex), so, too, the search for animus gives the government a chance to prove that it really was not motivated by animus. But again, that latter inquiry is part of rational basis review itself, as illustrated by rational basis cases such as Cleburne, rather than an inquiry into “discriminatory intent” per se.
Other scholars have reached similar, but not identical, conclusions. For example, Dale Carpenter has argued that a law should be struck down if animus is found to have “materially influenced” the government’s decision to enact it. As noted in my earlier reference to his argument, there is good wisdom in acknowledging that government bodies, especially multi-member bodies such as legislatures, usually have multiple motivations for their actions. Thus there is likewise good wisdom in recognizing that a smidgen of animus should probably not suffice to fatally infect a statute. The only questions then become, (1) How much animus is enough to cause that fatal infection? and (2) How do we structure the evidentiary inquiry? The “how much” question is addressed in the next chapter. Our discussion up to now concerns the structuring of that inquiry. That discussion suggests that the burden-shifting structure borrowed from other equal protection doctrines provides an approach that is both logical (because it asks the right question: whether the law was enacted “because of”5 animus) and normatively attractive (because it harmonizes the animus line of equal protection doctrine with its more completely worked-out discriminatory intent cousin).
Still, this harmonization is incomplete: As I have noted, in the animus context the burden-shifting structure I have been describing occurs within the framework of rational basis review, rather than as part of a preliminary inquiry into discriminatory intent. This difference may well strike readers as a trivial detail. One might object that, as a practical matter, the animus inquiry I have been sketching out is the exact same as the intent inquiry. Both require the plaintiff to make an initial showing that something troubling is lurking in the challenged law. Neither requires the plaintiff to show that that something is the efficient, or but-for, cause of the challenged law. And finally, if the plaintiff does make that initial showing, both inquiries require the government to prove that it was really motivated by benign purposes.
But there is a difference, and it matters. As Chapter 8 explained, a finding of discriminatory intent is different from a finding of animus. A finding of such intent—even an intent to discriminate on a presumptively unconstitutional ground such as race—simply triggers the relevant standard of review. To be sure, that review may well be quite stringent—as it is, for example, in the case of race itself. But, at least ostensibly and, indeed, even in practice, that review may not lead to the (intentional) race classification being struck down.6 By contrast, the animus conclusion—once the Court reaches it—does in fact lead to the law’s demise.
This distinction reflects the fact that animus is, on its own, a constitutional wrong. Not a presumptive wrong, subject to possible validation if the Court finds a compelling reason for the animus. Instead, as a per se matter, there can be no legitimate reason for an action taken out of animus. By insisting on this rule, we can clarify constitutional law jurisprudence by tying the modern Court’s animus doctrine to prior generations’ focus on class legislation and the framing generation’s focus on “faction” as the primary risk posed by majoritarian legislation. Indeed, it bears noting that the foundation for modern animus doctrine—Moreno—spoke of “a bare . . . desire to harm a politically powerless minority.” When we remember that James Madison expressed concern that democratic governance raised the risk of majoritarian oppression, we can draw a further connection between animus and the idea of faction.
It is worth the effort to draw this connection. Animus doctrine requires a strong, coherent doctrinal foundation in order to avoid serious conceptual and practical problems. First, its seeming imputation of bad subjective motivation raises difficult practical problems about proof—the problems we identified in Chapter 6. In turn, those problems exposed an even deeper practical, or perhaps political, problem that flows from defining animus subjectively: the problem of accusing a legislature, and even voters directly, of harboring and acting out of a subjective dislike of a particular group of their fellow citizens.
To be sure, the “animus” label, whatever the process by which it is affixed, carries with it some stigma of impropriety: however one slices it, labeling an action as motivated by animus necessarily suggests some type of improper motivation. Some scholars have cited that fact to criticize the entire concept of animus, even while favoring the results that doctrine reaches. One has described judicial conclusions of animus as “insulting and disrespectful,” while another scholar places the entire concept of animus in a category of judicial name calling he labels a “jurisprudence of denigration.”7 Some justices have leveled similar criticisms. In particular, recall that Justice Scalia began his dissent in Romer v. Evans (the Colorado Amendment 2 case) by accusing the Court of mistaking a (constitutionally valid) “cultural struggle” with a (constitutionally invalid) “fit of spite.”8 More colorfully, in Windsor (the Defense of Marriage Act case) he accused the majority of viewing members of Congress as “unhinged members of a wild-eyed lynch mob.”9
This objection can be overstated—at any rate, it is not unique to animus. For example, the concern about stigmatizing legislatures and voters has not stopped the Court from insisting that equal protection plaintiffs prove “discriminatory intent.” While this latter concept may not carry the full moral implications of a finding of “animus,” it nevertheless surely tars the challenged government action with some taint of moral impropriety.
Still, we can concede that a finding of “animus” stings—perhaps even more than a finding of “discriminatory intent.” But heeding these calls to abandon reliance on an animus theory threatens to discard a concept that can play a useful role in constitutional law. Despite any reasonable concerns about inflaming the culture wars by accusing one side of acting in ways that suggest a “lynch mob,” the fact of the matter is that animus plays a necessary and historically supported role in American constitutional law.
It is necessary because the enormous diversity of American society and complexity of American government make it more difficult than perhaps it used to be to conclude that any use of a particular classification trait is always necessarily suspect. Today, government makes a vast variety of decisions, at levels ranging from neighborhood councils and local agencies and boards to the Congress of the United States, on issues affecting nearly every type of human activity in which Americans engage. Such decisions necessarily involve classifying persons. The ubiquity of government classification matters because today, across any relevant axis—religion, ethnicity, lifestyle, socioeconomic class—American society is probably more diverse today than it has ever been. That diversity makes it more difficult than ever to judge all governmental classifications according to a rigid, three-tiered structure.
But an even deeper problem lurks. The crumbling of older binaries—black/white, Catholic/Protestant, and today, even male/female—makes it even more difficult for courts to engage in the categorization that serves as the first step in tiered scrutiny analysis. To be sure, classification tools—for example, race, sex, or sexual orientation—may ostensibly remain manageable. But to the extent such tools have historically been wielded with a solid understanding of what they mean and who is affected, such confidence may be eroding. For example, until recent decades it was relatively clear that “sex discrimination” meant discrimination against either women or men. Today, however, the term “sex” is far more shaded, involving, at least potentially, transgender persons (of various transition stages), intersexed persons, and, depending on one’s understanding of “sex discrimination,” even gays and lesbians.10 This is not to say that such developments make government action more likely to be constitutional. But it is to say that decisions about the constitutionality of a particular government action may have to be made at more granular, particularistic levels. The concept of animus can play an important role in making such granular decisions.
Indeed, we know that animus plays such a role because we have always known it. This raises the second point made above, about the historical groundedness of the idea of animus. At least since James Madison’s time, American political thinkers have recognized that democratic self-governance inevitably raises the danger of majoritarian self-dealing. Groups—“factions” in Madison’s terminology—may seek political power, not to promote their conception of the public good, but simply to further their own parochial interests. Of course, as I have noted several times already, the Court eventually realized that it could not easily distinguish such parochial concerns from conceptions of the public good that simply happened to benefit a certain group. A lower tax rate for capital gains income might well promote the interests of a particular class, but such a tax policy might also promote the public good. At the very least, courts have no special competence to decide such questions.
But what courts can do is investigate motivations. Again, discerning subjective motivations sometimes remains beyond a court’s competence. However, sometimes, as in Moreno and Cleburne, such intent might be relatively plainly visible. If not, as in Romer, it might be inferable from the larger context of the challenged law. In cases such as these, it is not only appropriate for courts to strike down the government action, but it is also consistent with Madison’s concern with faction and with nineteenth-century courts’ concern with class legislation. In other words, when courts embrace animus, even in its indirectly proven form, they embrace a principle that has animated American constitutional jurisprudence since the founding.
Let’s take stock. I have now suggested an approach to animus that recognizes both its logical connection to the intent inquiry but also its lack of complete equivalency. The complex relationship between animus doctrine and the intent inquiry—related, but not identical—rests in part on logic: the conceptual incorrectness of concluding that an animus finding, just like a finding of intent to discriminate on a suspect ground, simply triggers heightened judicial scrutiny. But it also rests on history: in particular, the opportunity to connect animus doctrine to firmly grounded constitutional concerns about class legislation and, even earlier in our history, about “factions” in a democracy.
We have now gone even farther. We have recognized that often the Court’s animus cases do in fact treat concerns about animus as simply a trigger for heightened scrutiny, rather than as a reason to stop the analysis immediately and rule for the plaintiffs. We were able to explain that seeming anomaly by noting that those cases’ heightened scrutiny can again be helpfully analogized to standard equal protection intent analysis—in particular, the component of the intent inquiry in which, after the plaintiff shows that the alleged intent played at least some role in the challenged decision, the burden shifts to the government to demonstrate that it would have made the same decision even absent that allegedly invidious intent.
I argued that such burden shifting is analogous to the heightened scrutiny courts apply in animus cases. In both cases, the goal is to figure out what is really motivating the government. But the searches proceed differently. In the intent context, that search is conducted by forcing the government to explain itself when the plaintiff proves that the alleged intent was a motivating factor in the government’s decision. In particular, when the plaintiff makes such a showing the burden shifts to the government to explain that other, non-discriminatory, purposes motivated the action the plaintiff claims is motivated by an intent to classify on the alleged ground (for example, race). Only after intent is established is the proper (heightened) level of scrutiny applied. By contrast, in the animus context, discovery of the government’s real motivation is accomplished through the means of heightened rational basis scrutiny that features a requirement that government affirmatively justify its action. In a very real sense, then, the animus inquiry folds the burden-shifting step into the heightened scrutiny step. If the government fails that combined step, then it loses—just as it loses if, in the standard intent inquiry, at step 1 it fails to convince the court that it would have made the same decision absent the alleged intent and then, at step 2, it also fails the resultant heightened scrutiny. Thus, the inquiries are very similar, but in standard equal protection analysis the inquiry proceeds in two steps, rather than in one as in animus cases.
All good work—at least, I hope that the reader thinks so. But we are left with two immensely practical, and immensely difficult, questions. First, in an animus case, how should courts go about determining when to use more careful scrutiny? In other words, when should courts suspect animus, with the result that they subject challenged laws to more careful scrutiny to confirm its presence? Second, what should that heightened scrutiny look like? Again, these are practical questions. The first one essentially addresses the question of how convincing the plaintiff has to be that animus may lurk behind the challenged law. The second one asks what a court is supposed to do when the plaintiff succeeds in raising that suspicion. Chapter 10 takes up both of these issues, before Chapter 11 applies what we have learned to concrete examples.