Harvard Law Review Harvard Law Review Harvard Law Review

Harvard Law Review Forum

Even a Dog: A Response to Professor Fallon

Responding to Richard H. Fallon, Jr., Constitutionally Forbidden Legislative Intent

In Constitutionally Forbidden Legislative Intent,1×1. 130 Harv. L. Rev. 523 (2016). Professor Richard H. Fallon, Jr., persuasively argues that Supreme Court doctrine offers various, mostly unsatisfactory, answers to the question of how to judge the constitutional significance of forbidden legislative intent. The doctrine concerns a wide range of issues, such as the intent to disadvantage a minority group, to favor religion, or to discriminate against interstate commerce. Case law is inconsistent: sometimes forbidden legislative intent renders an otherwise facially valid law per se invalid, sometimes forbidden intent triggers heightened scrutiny, and sometimes forbidden intent merely triggers an inquiry into whether the legislature would have enacted the legislation even absent the forbidden intent. Sometimes courts look for indications of subjective intent; sometimes they seek so-called objective intent. Fallon shows that there is little rhyme or reason to the choice among these options in the case law.2×2. Id. at 528.

In place of the disarray he finds, Fallon first proposes that most tests of objective intent be replaced by substantive constitutional rules.3×3. Id. at 529. With respect to subjective intent, he makes a normative argument for a general rule with an exception. He first argues that “courts should never invalidate legislation solely because of the subjective intentions of those who enacted it.”4×4. Id. (emphasis omitted). He also proposes, however, that forbidden subjective intent among a majority of legislators should trigger heightened scrutiny while forbidden subjective intent among a minority of legislators should trigger such scrutiny only where it sufficiently colors the social meaning of a law.5×5. See id. at 530.

Because I find Fallon’s critique of current doctrine and his proposed reconceptualization of objective intent both perceptive and persuasive, I shall focus in this Response on his normative proposal for addressing forbidden subjective legislative intent. Part I explains that Fallon does not rely on two familiar arguments against the invalidation of otherwise-valid legislation motivated by forbidden subjective intent: the notion that the intent of a multimember body is unknowable; and the difficulty of fashioning a lasting remedy for legislation infected by forbidden subjective intent. Indeed, Fallon cannot rely on either of these arguments because they would be fatal to the role that subjective legislative intent plays in Fallon’s exceptions.

Accordingly, Fallon’s reform proposal would appear to stand or fall on the basis of one key argument. Yet, as Part II explores, that argument has potentially disruptive implications for various areas of law that Fallon does not consider. Although Fallon brackets questions about forbidden subjective intent of nonlegislative governmental actors and private parties, his core reasoning rests on an argument by philosopher T.M. Scanlon that applies with equal force to such actors.6×6. See id. at 563–64, 564 n.196 (citing T.M. Scanlon, Moral Dimensions (2008)); see also Scanlon, supra, at 62–66 (discussing intent without regard to the actor’s social position). Scanlon’s argument has potential implications for end-of-life care, abortion, the law of war, criminal law, and tort law. Because we may be justifiably reluctant to follow Scanlon’s argument where it leads in these other areas, I suggest that we ought to be cautious about following it more generally, including with respect to forbidden subjective legislative intent.

Thus, with the familiar arguments against subjective legislative intent unavailable to Fallon (who, in any event, does not avail himself of them), and the novel argument that he does offer leading to possibly harmful and unevaluated side effects, my verdict on Fallon’s proposal is, as a Scottish jury might conclude, “not proven.”7×7. Samuel Bray, Comment, Not Proven: Introducing a Third Verdict, 72 U. Chi. L. Rev. 1299, 1299–300 (2005).

I. Arguments Fallon Does Not Make

Fallon gestures at a commonly offered reason for skepticism about legal rules that turn on forbidden subjective legislative intent — the difficulty of aggregating the intentions of multiple legislators, each of whom may have mixed motives, into a single legislative intent.8×8. Fallon, supra note 1, at 537. In explaining his decision to bracket questions of forbidden intent by executive officials and judges, Fallon states that their actions “do not present the main conceptual problem with which I am concerned, involving the aggregation of the mental states of multiple officials into a collective intent of a decisionmaking body.”9×9. Id. at 531.

However, Fallon does not really think that aggregating individual intent presents insuperable difficulties. He distinguishes the problematic use of subjective legislative intent in statutory interpretation — where the task is one of finding a dominant intent among the multiple, somewhat conflicting aims pursued by various legislators — from the relatively simple task of identifying evidence of a single forbidden intent in individual legislators and then applying an admittedly arbitrary (but nonetheless defensible) aggregation rule.10×10. See id. at 537–38. Because he thinks the latter task is manageable, he endorses aggregation as a trigger for heightened scrutiny.11×11. See id. at 541, 575. Accordingly, Fallon’s real reason for treating legislators differently from executive officials and judges (and from private actors, which I discuss further in Part II) is not and cannot be that the aggregation problem besets the inquiry in the legislative context.

Neither does Fallon rely on another difficulty with legal tests that turn on subjective legislative intent — fashioning a remedy. Suppose that a majority of legislators vote to locate a sewage treatment plant in a predominantly African American community because of racial animus. A rule that invalidates the plant-location law solely on that basis gives rise to a puzzle: can the legislature now go back and reenact the same law so long as the legislators are careful not to express their racist intentions out loud? If not, how long must they wait before they pass the same law? A year? A decade? Forever?

This remedial difficulty may explain the outcome in Palmer v. Thompson,12×12. 403 U.S. 217 (1971). in which the Supreme Court upheld Jackson, Mississippi’s decision to close its public swimming pools rather than integrate them.13×13. See id. at 219. Given the timing, the city council’s decision to close the pools undoubtedly was based on forbidden subjective intent,14×14. See id. at 218–19. but judicial invalidation of the pool closings solely on that basis would have created a quandary. Municipalities have no freestanding obligation to operate public swimming pools, so how long could an injunction keeping the pools open be maintained?15×15. See id. at 227. Because there is no good answer to that question, this account of Palmer suggests, courts should not invalidate otherwise permissible legislative decisions solely on the basis of forbidden motives.

In offering this remedial rationale for the result in Palmer, I am not saying that the case was rightly decided. Reasonable minds can differ. My key point here is that Fallon fails to invoke the remedial problem as support for his general rule, presumably because doing so would undermine his proposed exceptions.

Recall that where a majority of the legislators were motivated by a subjective forbidden intent, or where a sufficient minority of legislators were motivated by a subjective forbidden intent so as to make the social meaning of the law presumptively forbidden, Fallon would have the courts apply heightened scrutiny. Some laws that satisfy rational basis scrutiny will fail heightened scrutiny.16×16. For example, a decision to locate a sewage treatment plant here rather than there would be rational if based on a small difference in the cost of land, but a small cost savings typically would not be a sufficient basis for deliberately targeting a racial minority. Likewise, even modest cost savings could justify closing swimming pools under rational basis scrutiny but not under heightened scrutiny. What happens to those laws in cases where heightened scrutiny was triggered by forbidden subjective legislative intent? Can legislators reenact them after a suitable pause, so long as they keep quiet about their intentions? How long a pause? How quiet? We see here the exact same remedial problem that we saw in considering the general rule. Fallon cannot rely on a concern about remedy as the basis for his general disavowal of forbidden subjective legislative intent as a test for validity because the same concern about remedy would render his exception problematic.

II. Potential Implications Fallon Does Not Consider

Having necessarily set aside two common reasons for rejecting forbidden subjective legislative intent as the sole basis for invalidating legislation, Fallon rests his proposal almost entirely on an argument set forth most elaborately by Scanlon.17×17. See Fallon, supra note 1, at 565 (“If Scanlon is correct, his analysis should extend to issues concerning the legal justifiability of statutes.”). Because the case for Fallon’s proposal depends on Scanlon’s argument, that argument warrants careful examination.

Scanlon contends that the doctrine of so-called double effect — under which an actor may sometimes be justified in knowingly bringing about an undesirable end so long as the actor aims at a permissible one — rests on a confusion between the considerations relevant to whether an act is morally permissible and those relevant to how a moral agent should decide what to do.18×18. See Scanlon, supra note 6, at 2–3. Take, for example, a doctor trying to decide whether it is morally permissible to administer a lethal dose of sedative to a patient. The double-effect doctrine asks whether the doctor intends to kill the patient or merely administer a pain-relieving dose. Scanlon would argue that instead of focusing on her own intentions, the doctor should simply ask whether the benefit (pain relief) justifies the cost (death).19×19. See id. at 27.

Fallon is undoubtedly right that Scanlon’s argument, if persuasive, has implications for how courts ought to regard forbidden subjective legislative intent.20×20. See Fallon, supra note 1, at 565. The classic statement on the latter question can be found in Justice Stewart’s pithy summary of equal protection doctrine: whether a facially neutral government action that disproportionately disadvantages a constitutionally protected group has a discriminatory purpose depends on whether that action occurred “‘because of,’ not merely ‘in spite of,’ its adverse effects upon” that group.21×21. Pers. Adm’r of Mass. v. Feeney, 442 U.S. 256, 279 (1979). That test closely parallels the double-effect doctrine, under which a doctor may prescribe a lethal painkilling dose if she intends pain relief in spite of the death that results but not if she intends death to result. If Scanlon is correct that double-effect arguments are mistaken, then Justice Stewart’s test is either wrong or must be justified on some other basis.

Is Scanlon correct? Some scholars think more highly of double-effect arguments than he does.22×22. See, e.g., T.A. Cavanaugh, Double-Effect Reasoning (2006); Ralph Wedgwood, Defending Double Effect, 24 Ratio 384 (2011). For critical reviews of Scanlon’s argument, see Alison Hills, Book Review, 119 Ethics 792 (2009) (reviewing Scanlon, supra note 6); and Hallvard Lillehammer, Scanlon on Intention and Permissibility, 70 Analysis 578 (2010). I confess that neither Fallon nor I have done the hard work of first-order moral philosophy (for which we both lack professional training) that would need to be done in order to reach a conclusive general judgment about double-effect arguments.

Nonetheless, we need not cede the field to professional philosophers because philosophy is not the only discipline relevant to deciding whether to utilize Scanlon’s argument in reforming the law. We can also ask a question that legal scholars are reasonably equipped to answer: what implications does Scanlon’s argument have for other bodies of law?

Scanlon’s argument is relevant to much more than the question of how courts should regard forbidden subjective legislative intent. Unless Fallon has some sound reason for limiting the application of Scanlon’s argument to questions of legislative intent — as opposed to the broader relevance of intent in law — we can test the soundness of using Scanlon’s argument with respect to legislative intent by exploring its implications in other domains. Put differently, the potential undesirable implications of Scanlon’s argument in other contexts may yield a reductio that threatens to discredit (or at least call into question) the argument’s validity with respect to legislative intent.

Scanlon considers when the intentions of individuals are relevant to various moral questions. Legislative intent is not among the subjects he considers. Neither is assisted suicide, which receives only a passing mention in his book.23×23. The word “suicide” appears once in the Preface, see Scanlon, supra note 6, at x, and twice in citations because it is found in the title of an article by Judith Jarvis Thomson, see id. at 218, 242. Fallon, by contrast, does recognize the relevance of Scanlon’s argument for the morality of assisted suicide,24×24. See Fallon, supra note 1, at 564–65. but even he does not consider that issue as a constitutional question.25×25. Fallon quotes one of the leading Supreme Court cases on the issue to explain how the doctrine of double effect works, see id. at 564 n.197 (quoting Vacco v. Quill, 521 U.S. 793, 808 n.11 (1997)), but he does not assess the implications of his endorsement of Scanlon’s rejection of the doctrine of double effect for constitutional law regarding end-of-life medical treatment.

However, the doctrine of double effect plays a potentially crucial role in the constitutional law on assisted suicide. In Washington v. Glucksberg,26×26. 521 U.S. 702 (1997). the Supreme Court unanimously rejected a right to physician-assisted suicide. Yet in separate opinions, five Justices made clear that they did not mean to foreclose all challenges to laws proscribing end-of-life assistance in dying.27×27. See id. at 736–37 (O’Connor, J., concurring); id. at 750 (Stevens, J., concurring in the judgments); id. at 788–89 (Souter, J., concurring in the judgment); id. at 789 (Ginsburg, J., concurring in the judgments); id. at 792 (Breyer, J., concurring in the judgments). Crucially, Justice O’Connor, who provided a fifth vote for the majority opinion, invoked the doctrine of double effect. She explained that the Court’s ruling did not foreclose a constitutional right to end-of-life palliative care, “even to the point of causing unconsciousness and hastening death.”28×28. Id. at 737 (O’Connor, J., concurring).

Acceptance of Scanlon’s argument would, at a minimum, require rethinking the lines currently drawn by most states with respect to end-of-life decisions. If subjective intent is irrelevant, must states that permit lethal narcotic doses when intended to relieve pain now also permit deliberate killing, so long as the patient is also suffering otherwise untreatable pain? Or should states resolve the tension in the other direction and forbid palliative care that has the double effect of killing the patient? Fallon does not say.

A double-effect argument also features in the constitutional law regarding abortion. Since Planned Parenthood of Southeastern Pennsylvania v. Casey,29×29. 505 U.S. 833 (1992). the Court has recognized that the abortion right is partly grounded in bodily integrity.30×30. Id. at 849. A woman has a right not to be forced by the state to be pregnant; where abortion (rather than, say, inducing labor) is the only means of ending a pregnancy, she has a right to abortion despite the regrettable fact that abortion kills the fetus, not because of that fact.

Some scholars think this argument ultimately draws the wrong distinction.31×31. See, e.g., Philippa Foot, The Problem of Abortion and the Doctrine of Double Effect, reprinted in Philippa Foot, Virtues and Vices and Other Essays in Moral Philosophy 19 (1978). Even some scholars (like me) who defend an abortion right note that women often in fact have abortions for the purpose of avoiding parenthood, not merely avoiding the continuation of pregnancy.32×32. Sherry F. Colb & Michael C. Dorf, Beating Hearts: Abortion and Animal Rights 88 (2016). Abandoning double-effect arguments would not necessarily require abandoning a constitutional right to abortion. For example, it might be based on the view that abortion restrictions deny women equal dignity with men.33×33. See, e.g., Reva B. Siegel, Dignity and the Politics of Protection: Abortion Restrictions Under Casey/Carhart, 117 Yale L.J. 1694 (2008).

Nonetheless, the potential consequences of abandoning double-effect arguments or some other shift in rationale are sufficiently important to warrant careful consideration. But Fallon discusses abortion only in the quite different context of inquiring whether a legislature’s intent to make abortions difficult to obtain should result in invalidation of an otherwise permissible abortion restriction. He does not give any consideration at all to the implications of his endorsement of Scanlon’s argument for the right to abortion in the first place.

Moving beyond constitutional law, the doctrine of double effect is part of international humanitarian law. Deliberately targeting civilians is a war crime, but humanitarian law permits lethal attacks on military targets notwithstanding the substantial risk of civilian casualties, so long as the latter are proportional to the military advantage.34×34. See Jens David Ohlin, Targeting and the Concept of Intent, 35 Mich. J. Int’l L. 79, 82 (2013). Scanlon thinks that ex ante intentions are important because of their predictive value but that, as between otherwise similar military operations, one undertaken with the intent to cause civilian casualties is no worse than one that causes those casualties collaterally.35×35. See Scanlon, supra note 6, at 31–32. His view, if implemented, would thus result in a major change in humanitarian law. Fallon does not say whether he agrees.

Perhaps most familiarly, intent plays a substantial role in criminal law, tort law, and antidiscrimination law. “[E]ven a dog,” Oliver Wendell Holmes, Jr., famously wrote, “distinguishes between being stumbled over and being kicked.”36×36. Oliver Wendell Holmes, Jr., The Common Law 3 (1881). A judge or jury in a criminal case could justifiably give a harsher sentence to a defendant who acts out of hatred (under a hate crime enhancer or otherwise) than to one who acts with mere indifference; a judge or jury assessing punitive damages could punish a tortious wrongdoer more severely based on his illicit intent; and antidiscrimination law generally relies on the distinction between actions taken because of some protected characteristic of the putative victim and those taken in spite of that characteristic.37×37. See Wisconsin v. Mitchell, 508 U.S. 476, 487 (1993) (rejecting First Amendment challenge to hate crime statute based on illicit motive and analogizing the challenged law to antidiscrimination law).

I do not mean to suggest that acceptance of Scanlon’s argument would require the complete rewriting of each of these bodies of law, but some method of cabining the implications of Scanlon’s argument is needed. What means are available? Three come to mind.

First, we should note the limited role that the doctrine of double effect currently plays in the relevant bodies of law. For example, the criminal law does not generally distinguish between intending the consequences of an act and intending an act with the knowledge that it will cause, but not with the background purpose of causing, those consequences.38×38. See Norman L. Cantor & George C. Thomas III, Pain Relief, Acceleration of Death, and Criminal Law, 6 Kennedy Inst. Ethics J. 107 (1996). Likewise, in tort, liability can typically rest on mere negligence. Even punitive damages can be awarded based on “evil motive” or “reckless indifference.”39×39. Restatement (Second) of Torts § 908(2) (Am. Law Inst. 1979). Where existing law does not adopt or rest on double effect, abandoning double effect would have no impact. Where extant law does rest on double effect, perhaps alternative rationales can be developed, as noted above with respect to abortion.

Second, Scanlon rejects the doctrine of double effect, but he does not completely reject subjective intent as a moral standard. Scanlon thinks that intent matters to judgments about blameworthiness.40×40. See Scanlon, supra note 6, at 128. Thus, perhaps one could follow Scanlon in rejecting double effect, while preserving a role for subjective intent at the remedial phase of a criminal trial. After all, criminal sentencing concerns blameworthiness. To be sure, tort liability for punitive damages is supposed to be about deterrence, rather than blame as such, but one might say that an especially blameworthy person is more difficult to deter — and thus worthy of suffering a larger punitive damages judgment — than one who is merely indifferent to the harm he inflicts, and thus less blameworthy. More generally, notions of blameworthiness might be substituted for double-effect arguments wherever the latter currently hold sway.

Third, even if no principled reason can be found for distinguishing legislative intent (where Fallon would follow Scanlon) from individual intent (where he might not) there could be practical reasons for drawing the distinction. Fallon invokes just such “consequentialist” reasons for refusing to consider the potential implications of his view regarding legislative intent for individual intent cases.41×41. See Fallon, supra note 1, at 531. If these reasons are persuasive, unpalatable results in other areas of law would not discredit Fallon’s proposal with respect to legislative intent.

Accordingly, it might be possible to base Fallon’s approach to legislative intent on Scanlon’s argument without overturning a great many settled rules or standards. But the key word is “might.” Fallon offers no principled reason for thinking that Scanlon’s argument lacks far-reaching consequences, and he does not fully develop his suggestion that distinctive practical considerations could preserve subjective intent in individual cases despite its abandonment in legislative cases. At the very least, accepting Scanlon’s argument in any domain would open questions that had previously been thought resolved. If one finds the potential for disruption in Scanlon’s argument to be a source of concern, then one ought to proceed very cautiously before accepting it in all domains in which it applies — including legislative intent.


Professor Fallon’s positive analysis alone makes his article important. His normative proposal also has much to recommend it. Nonetheless, caution is warranted. Fallon writes that accepting his proposal “would require the rejection of fewer iconic holdings than one might expect.”42×42. Id. at 529. That is true, but perhaps because Fallon focuses on only those holdings that concern forbidden subjective legislative intent. His argument for his normative proposal rests on a much more sweeping view about the proper role of intent in morality and thus, as Fallon himself would have it, in law.43×43. See id. at 566–67. Until we have assurances that this sweeping view would not lead to unacceptable consequences in other areas of the law, the case for Fallon’s proposal will remain not proven.


* Robert S. Stevens Professor of Law, Cornell Law School. For helpful suggestions, the author thanks Sherry F. Colb and Bianka Takaoka.