Presidential Motive


Faced with a dire emergency, the President acts. A military plane sneaks into American airspace ready to drop a nuclear weapon on New York City. The President orders the plane shot down, preventing the bomb from detonating, and saving hundreds of thousands of lives. There is no doubt the order to shoot the plane down is in the national interest. But what if the President did not give the order for that reason? What if the President gave that order because he thought it would make him look decisive and his political enemy look weak? What if the only reason he acted was to help himself? Should that matter to whether the order was legal?

Consider another hypothetical. Pursuant to a congressional statute, the President has discretion to allocate military aid to any country so long as its use would further the “national security interest of the United States.” Pursuant to that authority, the President provides military aid to an ally, stipulating that the ally must investigate past allegations of corruption to receive the funds. The ally has historically suffered from rampant corruption and the military aid will help bolster the ally’s defenses against a mutual rival. The grant of aid is thus, in fact, in the “national security interest of the United States.” But what if that was not the reason the President acted? What if what motivated the President was the belief that the corruption investigation would damage his political rival and help his reelection campaign? Should that matter to whether the order was legal?

Consider one more example. A hurricane causes tremendous damage to a southern state, and the governor asks the President for disaster relief funds. The President has discretion to provide such funds if doing so is in the national interest. The President refuses to grant the funds, but he makes this decision not because the aid would not benefit the country, but because the state did not vote for him in the last election. Does it matter that the President’s reasons for acting were politically vindictive, or should we only ask whether the aid would further the national interest?

The question of whether the President’s actions in these examples are legal may seem intuitive to people. But that intuition may vary. It may be intuitively easy, hard, or painstaking—and in either direction. Some might think that the President’s conduct is clearly lawful so long as his actions, in fact, serve the interest of the United States and satisfy any statutory conditions. What more can you ask of the President? Others might argue that whether an act serves the national interest is not the end of the inquiry—rather it matters what internally drives the President, that is, the President’s motive matters. On this view, the President cannot act for any reason at all—he can only act for some reasons and not others.

What, if anything, does the Constitution have to say about this question? Remarkably, we lack a direct answer to this question. This is not for any lack of interest in the President’s motive. To the contrary, discussion of the President’s motive has seemed ubiquitous in public discourse in recent years,1 and has played a prominent role in three of the last four impeachment controversies going back to Richard Nixon.2 Presidential motive has also featured prominently in scholarship on the President, arising in discussions of what constitutes an impeachable act,3 how to assess allegations that the President acted based on impermissible animus in violation of individual rights provisions of the Constitution,4 as well as how statutory obstruction of justice ought to apply to the President.5 But despite routine debates relating to the President’s motive and sporadic extremely high-profile public controversy about it, we lack any sort of comprehensive answer to the question of whether and how the President’s motive might matter to the constitutionality of his conduct writ large. It now seems generally accepted that the President cannot act based on impermissible animus in violation of people’s individual rights.6 But what if the President acts, not based on animus, but based on more commonplace personal or political motives? What if the President wishes to go to war for political purposes? Or build a border wall to fulfill a campaign promise? Or seek a corruption investigation to tar his political rival? Or require vaccinations to bolster his image as a leader?7 What does the Constitution have to say about that? This Article takes on such questions directly by providing the first full-length scholarly treatment of the relevance of the President’s motive to the legal permissibility of her conduct under Article II.8 That is, does the President’s motive matter whenever she exercises power?

The Article begins with the legal question: Does Article II limit which motives can drive presidential conduct? Part I sets the stage by showing that there is, in fact, a remarkable level of consensus on this notion, shared even by prominent defenders of expansive presidential power like former Attorney General William Barr. Such defenders of expansive presidential power seem to accept that Article II limits which motives can drive presidential conduct, their objection is to whether and how Congress might enforce such limits. The crux of the debate, then, is not about whether motive is relevant to legal permissibility, but which motives are relevant and how to enforce such a motive requirement. Part I then turns directly to the text of the Constitution, arguing that the “faithful execution” requirements of Article II require the President to act motivated by the public interest, rather than her personal interest. This argument has support in originalist arguments but is also grounded in a more pluralist argument that takes into account basic features of the structure of our government to interpret the term “faithful” execution. At bottom, to be a democratic representative means to act for the benefit of whomever is being represented, rather than one’s own personal benefit. Because the President is such a representative, to “faithfully” execute her role, she must be motivated by serving the public interest, rather than her own interest. This is Article II’s motive requirement.

Even if we accept that Article II requires the President to act for some reasons and not others, this does not resolve many of the most interesting or complex questions. The first is normative. Can such a requirement be normatively justified? Why should the President be prohibited from taking action that is, in fact, in the country’s interest, simply because she is not motivated by that reason? Could the President’s order to stop a nuclear attack on New York City really be rendered unlawful if the President did it for the wrong reasons?

Drawing on insights from broader debates about the relevance of motive to the permissibility of conduct,9 Part II concludes that inquiry into the President’s motive can be normatively justified on both consequentialist and nonconsequentialist grounds, but that it is more justifiable in some areas than in others. To explain where and why, Part II develops a novel framework for analyzing presidential power dividing delegations of such power into either discretionary authorities or duties that are premised on conditions that are either objectively verifiable or objectively unverifiable. The President’s motive requirement is most justifiable in areas of presidential discretionary authority where power is premised on nonverifiable conditions. It is in such areas where motive is most likely to serve as a useful proxy for the national interest or other condition that cannot be reliably “objectively” evaluated and where the President is exercising discretion rendering motive more intrinsically relevant to the permissibility of conduct. Conversely, looking to motive is least justified in areas of duty premised on verifiable conditions or where the national interest is obvious. In such areas, the President simply “must act”—it is, after all, her duty to—rendering motive less intrinsically relevant. And, because the national interest is obvious or the conditions are otherwise objectively verifiable in such contexts, motive will serve as a less useful proxy for their satisfaction.

Applying this framework can resolve conflicting assessments in some of the cases above, explaining why we might think motive ought not matter in the context of stopping the nuclear attack, but ought to in the context of deciding whether to grant discretionary disaster relief or call for foreign corruption investigations.

Part II discusses further reasons for looking to motive—including that doing so can help prevent pathological presidents and that motive can change the meaning of presidential action—and grapples with the downsides of making motive relevant to legal permissibility. It ultimately concludes that Article II’s motive limit is normatively justified writ large, particularly given that most of the president’s important exercises of power lie in areas where motive inquiry is most justified—areas of discretionary authority contingent on nonverifiable conditions. This normative analysis is important for its own sake, but also because it is necessary to understand which types of presidential power should be the focus of enforcement efforts.

Before turning to issues of enforcement, however, we must know what precisely qualifies as a permissible or impermissible motive. It is easy to identify some impermissibly personal motives—like taking action for pecuniary gain—but the question gets more complicated when we consider the President’s political interest in light of her role as a democratic representative. For example, when the President acts in order to make herself more popular, should that be considered personally interested or public-interested behavior? 

Part III provides a straightforward answer to this question: Furthering the President’s political fortune, like her financial fortune, is a personal, rather than public-interested, reason for acting and is therefore impermissible. This answer might seem too straightforward. After all, on some conceptions of political representation, the representative ought to be motivated by what is likely to be popular. Others might object that prohibiting politically motivated conduct sets the bar too high in practice, because Presidents appear to act purely for political reasons all the time. This section responds to these objections, noting that the theoretical objections do not pose a problem on most contemporary conceptions of representation—or those of the Founding era—which require the President to exercise independent judgment as to what is good for the country, rather than simply pursuing popularity. Meanwhile, the practical objections are grounded in empirical assumptions that are deeply contestable and unproven. Although neither objection proves strong enough to rebut the straightforward conclusion that acting to serve the President’s political interest is an impermissible personal motive, they do raise legitimate concerns. Part III thus assesses an alternative approach to the straightforward answer, which would permit more routine politically motivated conduct but prohibit egregious politically motivated conduct that serves to undermine the democratic process. Such a democratic “Rules of the Game” approach has some intuitive appeal but ultimately proves unworkable as a means to distinguish between permissible and impermissible political interest for purposes of Article II. The most straightforward answer remains the best one we have: The President’s political interest is an impermissible motive.

Of course, sometimes the President will be motivated by multiple motives—some permissible and some impermissible. Part III.B. addresses such “mixed motive” cases by providing a menu of options and proposing a “Meaningful Sole Motive Standard” for presidential mixed motives. This standard would permit the President to act so long as she is meaningfully motivated by the public interest, even if the President is also or even mostly motivated by her personal political interest. Part III explains why such a standard can be justified against more-or-less stringent standards but concedes that the optimal mixed motive standard is contestable—dependent on one’s normative and empirical conceptions. It thus concludes with a framework for adjusting the mixed motive standard based on one’s views of: (1) the empirical frequency of the impermissible motive; (2) the egregiousness of the impermissible motive; and (3) the normative justifiability of motive inquiry in the particular context.

Having established that Article II imposes requirements on the President’s motive, where such requirements are most and least justified, and which precise motives are permitted and which are not, Part IV then takes on questions of enforcement. It first argues that the President, himself,10 could help enforce the motive limit by requiring that a public-interested reason be announced before taking important action. Although such a requirement would not fully prevent personally motivated conduct, it would make it more difficult to pursue self-interested policies and push the President’s action—and those working for him—toward public-interested policy. In keeping with the normative analysis above, a more robust explanation should be required in areas of discretion based on nonverifiable conditions, where it is most important to abide by the President’s motive requirement. Part IV then explains how Congress can help enforce this requirement through impeachment or by requiring public-interested reasons for action and changing how it delegates power to the President. 

Part IV concludes by examining how courts could best enforce the President’s motive requirement. In theory, courts could apply a form of heightened or nondeferential review to the President’s action where it is found to be impermissibly motivated. Such nondeferential review might work for powers premised on verifiable conditions, but it is unlikely to work in areas of discretionary authority premised on nonverifiable conditions where motive inquiry is most justifiable. Courts are simply not equipped to assess whether conduct is, for example, in the “national interest” in a nondeferential way. Because this form of review would not work where the motive requirement is most justified, Part IV proposes an alternative approach. Under this approach, courts could require the President to give a public-interested explanation for action and potentially review such reasons for pretext in limited circumstances. This form of review would help enforce the President’s motive requirement—including in areas where it is most justified—and have additional salutary public accountability benefits. That said, as Part IV explains, the potential for error in judicial review in this context is sufficiently high that judicial overturning of presidential conduct based on the President’s motive ought to occur in only the most egregious circumstances.

None of these enforcement mechanisms will fully enforce the President’s motive requirement. And there is nothing unusual about such underenforcement of constitutional requirements.11 Nonetheless, we can do better. We can better ensure that the President complies with the requirements imposed on her by Article II, and we can better ensure enforcement focuses on where such motive requirements are most important. Doing so requires us to grapple with complex questions that have thus far lurked in the background: Does the Constitution impose requirements on the President’s motive at all? When are such motive requirements most and least justified? Which motives are permitted and which are not? And which enforcement mechanisms are realistically available? To date, we have lacked direct answers to these questions. This Article seeks to change that.

Before beginning, a few clarifications are in order. First, when I use the term “motive” or “intent,” I mean it to describe “the reasons for which agents understand themselves to be acting.”12 As Micah Schwartzman has stated: “For any given action, we can ask a person: ‘Why did you take that action?’ Someone who responds sincerely will state her intention by offering her reason (or reasons) for acting in the way that she did.”13 I use the terms “motive” and “intent” interchangeably in the same way: to signify the reasons the President has for acting.14 Second, the President is one person but has an entire institutional apparatus that works under him. Here, I focus on the President as a person with individual motives, recognizing that most presidential exercises of power rely on numerous actors within the White House complex and executive branch to be enacted. I do this to simplify the analysis and because Article II’s motive limitation most squarely applies to the President. Assessing how Article II’s requirements translate to those who work for the President is a sufficiently complex task as to require separate treatment.15 Third, some might raise threshold epistemic questions about whether we can ever truly know even the individual President’s subjective motivations. I will largely bracket these questions for purposes of this Article. Inquiry into individual motive is commonplace in everyday life and a common feature of ordinary civil and criminal litigation.16 Although it is not always easy to assess the President’s motive, and we may not always be able to do it with full confidence, identifying it does not seem uniquely challenging.17

Some might object on deeper grounds, viewing the inquiry into Article II limits on the President’s motive as a fundamentally misconceived attempt to impose a legal solution to a political problem. On this view, law cannot supersede politics, and, if the problem is that Presidents act self-interestedly, we need to look to politics, not Article II, to address it. Here, it is important to clarify that the focus on law in this Article does not presume any particular relationship between constitutional law and politics. Public law and politics inevitably interact, and politics often prevail. Constitutional law relies on political actors for its enforcement, and if such political actors do not wish to enforce the law, identifying formal legal requirements will not inevitably constrain official conduct.18 Thus if the upshot of this project is to identify a legal requirement that is sometimes honored and sometimes not, that is not particularly shocking.

Even if Presidents are not intrinsically motivated to follow their legal obligations—and, indeed, perhaps especially if that is true—that is not sufficient reason to act as if no legal obligations exist. Law matters even if its violations are not always remedied, and there is value in identifying legal requirements, even if one is unsure of whether they will be—or ought to be—perfectly enforced. Among other things, identifying the law can help encourage better compliance, coordinate conduct, and inform the public, as well as government lawyers, about what might be expected of elected officials—even if it does not result in all violations of the law being invalidated. In other words, Presidents ought to know what standard to hold themselves to, the people who work for Presidents should know what standard to hold them to, Congress and the courts should know what standard to hold Presidents to, and, perhaps most importantly, the public should know what standard to hold Presidents to. We can both believe the President will not always live up to these standards and that identifying them has value. Providing an ethic of presidential legality—even if it is not always lived up to—thus strikes me as a worthwhile endeavor. Only if we understand what the President ought to do can we truly hold the President to account. We might still choose not to punish Presidents for failing to abide by these legal requirements, but for that to be a choice it requires knowing what those legal requirements are. The central task of this Article is to inform that choice.

Article II and Presidential Motive

This section addresses the question of whether Article II makes motive relevant to legality and, if so, which motives are permitted and which are not. It concludes that Article II does impose limits on which motives can drive presidential conduct. In particular, the president must act motivated by serving the public interest and cannot act based on serving personal interests. This limit comes first and foremost from Article II’s requirements of “faithful execution” present both in the Take Care Clause, which requires that the President “shall take Care that the Laws be faithfully executed,” and in the President’s oath to “solemnly swear (or affirm) that I will faithfully execute the Office of the President of the United States.”19 Below I lay out two arguments for why this faithful execution requirement imposes a limit on which motives are permissible and which are not. The first relies on recent originalist work on the meaning of faithful execution, and the second is more pluralist in methodology, looking to basic background structural principles to inform a straightforward textual analysis.

The Consensus Regarding Presidential Motive

Before turning to the constitutional text, it is worth highlighting how little disagreement there is about whether the President’s motive is relevant to the legality of his conduct. Despite frequent disagreement on the nature and scope of the President’s power, there appears to be general consensus among commentators on presidential power that the President’s motive can matter to whether the President’s conduct is legal. This apparent consensus is not definitive proof that motive matters to the legality of the President’s conduct, but it helps set the stage for the doctrinal discussion. If even prominent defenders of expansive presidential power seem to agree that the President’s motive is relevant to the legality of her conduct at least in some instances, this might give us comfort that the seemingly straightforward doctrinal argument for looking to motive is correct.

To understand this consensus, it is important to disaggregate questions about the President’s internal legal obligations—by which I mean the legal obligations required by the Constitution or statutes—and how such obligations might be externally enforced—typically by Congress or courts. Our focus for now is on the first question—what obligations the Constitution imposes on the President with respect to which motives can drive her conduct. How to enforce such limitations—if at all—is a conceptually distinct question that we will return to in Part IV.20 With the distinction between these questions in mind, the general agreement that the President has internal obligations to act based on certain motives comes into view.

Take, for example, former Attorney General William Barr, a famous proponent of extremely expansive views of Presidential power.21 Before he was appointed Attorney General by President Trump, Barr wrote an unsolicited memorandum to senior Justice Department officials arguing against the constitutionality of Robert Mueller’s investigation of President Trump for obstruction of justice.22 The memorandum has its fair share of imprecise rhetoric and tone,23 and has been read as supporting the proposition that Article II does not limit the President’s motive.24 Barr’s memo, however, seems to accept that the President is obligated to act only for certain reasons by virtue of his obligation to “faithful[ly] exercise” his powers.25 Barr’s objection is about whether and how Congress can enforce such limits, not whether they exist at all.26

Barr’s primary objection was to a position he ascribes to Mueller that if the President interferes with a proceeding where his own conduct is being investigated, such influence is per se self-interested criminal “corruption.”27 Barr argues that

it cannot be presumed that any decision the President reaches in a case in which he is interested is “improperly” affected by that personal interest. Implicit in the Constitution’s grant of authority over such cases . . . is the recognition that Presidents have the capacity to decide such matters based on the public’s long-term interest.28

Here, Barr does not claim that the President can act in such cases for any reason at all. Rather, his objection is to the notion that Congress can criminalize all conduct that appears self-interested. On Barr’s view, Congress cannot do this because this would, in effect, remove the President’s ability to act in such cases for the right reasons.29

Barr’s claim thus seems to accept that the President can only act for certain reasons—i.e., “the public’s long-term interest.”30 This requirement seems grounded in repeated references by Barr to the President’s obligation to “faithful[ly] exercise” his powers.31 Barr does not believe Congress can enforce the President’s duty to “faithfully exercise” his power through the criminal law, but he does think such a duty can be enforced by the public or through impeachment, stating “that the proper mechanism for policing the President’s faithful exercise of [his] discretion is the political process that is, the People, acting either directly, or through their elected representatives in Congress [through impeachment].”32 In short, Barr seems to accept that the President’s obligation to “faithfully exercise” his powers limits which motives can drive his conduct—his disagreement is about how such limits can be enforced.

And Barr is not unique in this regard. Although defenders of broad presidential power often object to congressional or judicial enforcement of such limits, they seem to generally agree that there are some limits on the President’s permissible motives.33 The fact that these commentators seem to accept that the President can only act for certain reasons helps clarify where the crux of the debate is. It is not about whether motive is relevant, but which motives are relevant and how to enforce such obligations.

“Faithful Execution” and Presidential Motive

With this in mind, we can turn to the text. Does Article II’s faithful execution language limit which motives can drive presidential conduct? Recent groundbreaking originalist work suggests it does. In their influential article on “Faithful Execution and Article II,” Andrew Kent, Ethan Leib, and Jed Shugerman conclude that “Article II of the Constitution . . . limit[s] Presidents to exercise their power only when it is motivated in the public interest rather than in their private self-interest, consistent with fiduciary obligation in the private law.”34 On this view, if the President’s conduct is, in fact, in the public interest, it can still be unlawful if it was not undertaken for that reason.35 These authors draw this conclusion from exhaustive work on the original public meaning of the term “faithfully,” which, in their view, incorporated what might now be called “fiduciary” duties to act in the best interest of the fiduciary’s beneficiary—here the American people.36 

The fiduciary basis of this work has recently drawn criticism. Several scholars have suggested that it is anachronistic to interpret the Constitution in light of fiduciary principles, because such principles were unsettled at the time or were not incorporated into the document’s meaning.37 But, even critics of the fiduciary turn do not seem to contest that the President must be motivated by the public interest—they simply do not think this requirement stems from fiduciary principles. For example, Richard Primus has argued that there is little reason to think that the Constitution incorporates or meaningfully mimics fiduciary law.38 But Primus seems to accept that government officials like the President must be motivated by the public interest in exercising power—this is, in fact, one of the similarities between the Constitution and fiduciary instruments that Primus highlights.39 Primus simply does not view this requirement as deriving from—rather than overlapping with—fiduciary principles. Other critiques of the fiduciary turn in constitutional law have not been as explicit about adopting this motive requirement, but they have not argued against it either.40 In short, while critics of fiduciary constitutionalism dispute that the Constitution ought to be interpreted as a fiduciary instrument—they do not seem to dispute that the President must be motivated by serving the public, rather than his own interests. They just do not think this duty applies by virtue of the incorporation of fiduciary principles.

Whatever one thinks of this fiduciary critique, some might not find Kent, Leib, and Shugerman’s work dispositive for other reasons. The authors themselves focus exclusively on original public meaning and acknowledge other modalities might be necessary to fill in the content of constitutional law.41 Below I provide an alternative account for the claim that Article II imposes a requirement to act motivated by the public interest based on what I view as a straightforward textual interpretation of the President’s “faithful” execution requirements informed by basic structural features of the Constitution.

Begin with the simple structure of our constitutional government. The Constitution created the federal government as a public body to serve public ends.42 The particular form of government created was a republican, rather than direct, democracy.43 This form of government consists of representative officials tasked with acting for the benefit of the public.44 As Joseph Story put it in his influential treatise on the Constitution, “[i]t should never be forgotten, that in a republican government offices are established, . . . not to gratify private interests and private attachments; not as a means of corrupt influence, or individual profit . . . but for purposes of the highest public good.”45 In short, as democratic representatives, public officials exercise public power on behalf of the public, not on behalf of themselves. This is not a claim reliant on fiduciary principles, but more basic conceptions of what it means to be a democratic representative.

Although the proper conception of representation in a democracy is hotly contested,46 it is generally agreed that, to act as a representative at all means that one is required to act for the benefit of whomever the representative represents, rather than for the representative’s own benefit.47 This approach is exemplified in the classic work on the topic by Hanna Pitkin, who explains “that the representative must pursue his constituents’ interest, in a manner at least potentially responsive to their wishes, and that conflict between them must be justifiable in terms of that interest.”48 As Pitkin makes clear, “[i]t will not do for a representative to assert that he did what he did for his own private interest; after all, he is not there for himself.”49 In short, while there will inevitably be contestation about the extent to which the President should be responsive to the public’s wishes rather than use her own judgment to assess what is good for the public, fundamental to the conception of being a representative at all is a requirement that the President aim at the benefit of whom she represents, rather than her personal benefit.50

With this background in mind, we can look to the text of the Constitution. The President is obligated to “faithfully” execute the “Laws” and “Office” of the presidency.51 What does it mean to execute such laws or offices “faithfully”?52 Dictionaries at the time defined “faithfully” as “With strict adherence to duty. . . Without failure of performance . . . Sincerely; with strong promises . . . Honestly; without fraud . . . Confidently; steadily.”53 The core of executing “faithfully” is thus to “adhere to [the] duty” of execution, to execute “sincerely,” “honestly,” and “without fraud.” If we accept that the government and its representatives are meant to further public ends—that they work on behalf of the public—then executing the laws and the office of the President “sincerely” or “honestly” and “without fraud” requires doing so for public ends.54 

Does this mean that the President must be motivated by pursuing public ends when he acts? I think it does. “Faithfully” executing the laws and office of the President is a requirement that speaks to the subjective mindset of the President—she must act “sincerely” or “honestly,” which are subjective characteristics. If being a representative means acting for the benefit of others, then to execute this job “faithfully”—to do it “honestly” or “sincerely”—is to do so by genuinely seeking to further the interests of those the President acts for—the public.55 If this is correct, the President’s reasons for acting must be to further the public interest, not her own interest.56

This does not mean that Presidents have always abided by this obligation or that identifying where the President’s personal and official interest begin and end will be easy.57 But acting motivated by serving the public seems to follow rather straightforwardly from the Constitution’s requirements. Nothing forced the Framers to set up a representative form of government or to explicitly require the President to act “faithfully” in doing her job. By constructing such a government and imposing such an obligation on the President, the Framers made the President’s motive relevant to the legality of her conduct under Article II. For the President to act consistent with her constitutional obligations, she must do so motivated by serving the public interest—however defined—rather than her personal interest.

One counterargument to this claim is worth addressing briefly. In his important work on Constitutional Bad Faith, David Pozen argues that Madison’s conceit that “[a]mbition must be made to counteract ambition,” suggested that “[o]pportunistic” or “self-interested” behavior was expected, rather than forbidden, by the Framers.58 Pozen’s claim is not directly focused on our inquiry, but it raises an intriguing counterargument: If Madison expected “ambition” to replace the need for “better motives,”59 does this mean that proper motives are not required?

I do not think so. Although Madison was clearly worried that representatives’ self-interest might supplant their duty to the public good,60 this does not mean he did not view them as bound by such duty. To the contrary, Madison worked hard to construct a government that would select for representatives least likely to succumb to such self-interested behavior,61 and remained committed to a conception of “representatives . . . ideally pursu[ing] the ‘public good’ instead of the narrow interests of their constituents—or themselves.”62

Madison’s reference to “ambition” counteracting “ambition,” is consistent with this notion. The “ambition” Madison spoke of did not signify conduct motivated by the personal interest of the official, but rather the interest of the branch of government in which the official sat. Madison sought to align “[t]he interest of the man . . . with the constitutional rights of the place.”63 As Daryl Levinson and Richard Pildes have put it, “[i]n the Madisonian simulacrum of democratic politics . . . the branches of government are personified as political actors with interests and wills of their own, entirely disconnected from the interests and wills of the officials who populate them.”64 Madison’s theory thus envisioned “government officials who care more about . . . their departments than their personal interests.”65 To be clear, Madison’s conception of democratic officials incorporating their branch’s interest was famously misguided.66 But it was entirely consistent with the notion set forth above that Article II imposes a requirement on the President to act motivated by serving the public interest.67

The Normative Defensibility of a Presidential
Motive Requirement

Above I have argued that the President has a duty to be motivated by acting in the public interest. But, even if we accept the Constitution imposes such a duty, it is worth pausing. Does it make sense to render presidential conduct illegal based on the President’s motive? Can this be normatively justified? If the President wishes to exercise power that is, in fact, in the national interest, why should it matter if she does so to help herself, rather than the country? Why should what goes on in the President’s head matter more than what goes on in the world?

This argument against looking to motive is a serious one, echoing prominent debates in both the philosophy and public law literature about the relevance of motive to the permissibility of conduct writ large.68 Drawing on insights from these debates, this section tackles these questions head on, concluding that the President’s motive requirement is normatively justified writ large, but that it is more justified in some areas than in others.

The Main Objection

In the moral philosophy literature, there has long been a claim, sometimes referred to as the “Doctrine of Double Effect,” that otherwise permissible conduct can be rendered impermissible by virtue of the motives behind it. A famous example imagines a bomber tasked with bombing a munitions factory. Pursuant to the Doctrine of Double Effect, the bomber can bomb the factory knowing it will inflict civilian harm, but not if he intends to kill civilians to terrorize the population.69 The effect in terms of civilians killed is the same, but the intuition is that intentionally killing the civilians as a means to an end is wrong, while killing them as a foreseeable, but regrettable, side-effect is permissible.

This notion that the decisionmaker’s internal motivation can render conduct morally impermissible has received significant pushback of late by prominent philosophers.70 For example, T.M. Scanlon concludes that “what makes an action wrong is the consideration or considerations that count decisively against it, not the agent’s failure to give these considerations the proper weight.”71 Such claims against the relevance of motive to permissibility have recently been incorporated into the public law literature by Richard Fallon in his recent work arguing that legislative acts should never be invalidated solely because of the legislature’s motive, rather than based on more objective considerations.72

Many of these accounts draw on hypotheticals that serve to highlight the oddness of concluding that subjective motives ought to trump good effects in the world. For example, Scanlon notes that if a person sees someone drowning in a lake, she should save the person regardless of whether or not she does it for the right or wrong reasons.73 Fallon, meanwhile, uses a running example of a legislature that has criminalized murder, but done so for impermissible religious reasons of enforcing the Sixth Commandment.74 Fallon finds it inconceivable that a court would render such a law invalid just because of the legislature’s impermissible intent.75 These points are intuitive and can be easily translated to the Presidential context. If the President orders an airstrike that prevents a nuclear bomb from going off in New York, could it really be that such conduct would be illegal if the President did it to serve his personal interests? Intuitively, it seems hard to accept such a claim.

Apart from these primarily nonconsequentialist arguments, there are also consequentialist reasons to ignore motive. On consequentialist terms, the question of whether particular conduct should be permissible is a question about effects out in the world, rather than thoughts inside someone’s head. The consequences that matter, in other words, are consequences external to the President’s mind, and so in evaluating whether conduct is justifiable, we should just ask about those consequences.

These claims are serious and worth examining. Below, I give several reasons why, notwithstanding these objections, motive inquiry is justifiable in the context of presidential power, although I conclude it is more justifiable in some areas than others.

Motive as Proxy

One reason to look to presidential motive as relevant to the legality of her action is that motive is likely to serve as a useful proxy for the national interest. Recall that the critique of looking to whether the President is motivated by the national interest is that it would be better to just ask if the relevant conduct was in fact in the national interest, instead of looking to whether the President is motivated by the national interest. On this view, we ought to care about the objective justification for the action, not what happens subjectively inside the President’s head.

The problem with this critique is that for many important exercises of presidential power, it will be exceedingly difficult, if not impossible, to determine whether conduct is in fact in the national interest. This is because whether presidential action is, all things considered, good or bad for the country is often essentially impenetrable to objective inquiry.76

For example, when President Obama decided to use military force in Libya, he stated it was based on several important national interests.77 But was it true that such military engagement was good or bad for the country, all things considered? It is not clear how one can evaluate such a claim “objectively” ex ante with any confidence. When President Trump sought to divert funds to a border wall, was that objectively in the “national interest?” How about President Trump’s attempt to instigate an investigation of then-candidate Joe Biden’s son for his business dealings in Ukraine? On the one hand, the request seemed clearly personally motivated. On the other, many concluded that it would be good for the country to know whether the son of a major political figure engaged in foreign corruption.78 Was such a request good or bad for the country? It is not clear how one could answer this question objectively ex ante. In short, when the President acts, it is frequently impossible to objectively assess whether the action is, all things considered, in the national interest. The inquiry is more a question of judgment than objective fact.79

Looking to motive, on the other hand, is a question of fact that—although not simple—is often more easily ascertainable than the purportedly objective national interest. Further, once identified, motive can serve as a useful proxy for the national interest. We might assume that conduct motivated by personal interest is unlikely to be in the national interest. And the fact that conduct is motivated by public-interested reasons might serve as the best proxy we have for the objective national interest. These proxies will not be one hundred percent accurate. Some personally interested conduct might happen to be in the national interest and publicly motivated conduct might be bad for the country. But if we cannot evaluate the national interest objectively, looking to the President’s motive as proxy is defensible.80

Consider the Ukraine example above. Although we might not be able to assess whether investigating Hunter Biden’s dealings in Ukraine was good or bad for the country in some objective sense, looking to the President’s motive might tell us more. If we believe the President acted solely to further his own ends, then we might think this serves as a reasonable proxy for the conclusion that the conduct was not, in fact, in the national interest. Therefore, if we are reasonably confident in the President’s private-interested motive, we might also be reasonably confident the conduct is not in the national interest.

Other times, resort to motive will not be necessary because it will be obvious what the national interest is. Take the order to strike down a plane en route to drop a nuclear bomb on New York City. In such a situation, there is no need to examine motive as a proxy for the national interest, because the national interest is obvious. The line between what is obvious and what is not will sometimes be difficult to draw, but much will fall easily on one side or the other.81

Understanding this justification for looking to motive helps identify a normative distinction to guide our future analysis: Motive inquiry is more justified when dealing with instances where objective evaluation is difficult, and less justified when such evaluation is easy.

This point also has implications for statutory conditions. Many statutory delegations are conditional on the President making vague findings like determining that conduct is in the “national security interest of the United States” or “paramount interest of the United States.”82 Like the national interest prong itself, these findings will often be essentially impenetrable to objective inquiry. In such situations, looking to the President’s motive is likely to serve as a proxy not only for her Article II duties to act in the national interest, but also for whether the statutory conditions triggering the exercise of authority has been met.

In short, where conditions are not objectively verifiable, looking to motive can serve as a useful proxy for whether the condition has in fact been met.83 Looking to motive can thus be justified, even if we are concerned primarily with consequences in the world.

The converse of this point is that where conditions are objectively verifiable, looking to motive is less justified. In such situations, if one is concerned with effects on the world, one can simply ask if the condition has been met objectively. For example, the President is tasked, by statute, with sanctioning individuals if they “ha[ve] knowingly . . . contributed” to the efforts of certain countries to use or acquire chemical weapons.84 In such situations, we might be better off asking about whether the objective condition has been met—whether the person, in fact, knowingly contributed to the chemical weapons program—than looking to the President’s motive behind making that finding.

In sum, at least in areas where it is hard to objectively verify the national interest or necessary statutory conditions, inquiry into motive can be justified as a proxy for objective inquiry. This presents an interesting inversion of then-Professor Elena Kagan’s famous argument that the objective tests in First Amendment law operate as a proxy for illicit motive.85 As Kagan argued, one underlying reason the Court was interested in identifying motive at all was its view that bad motive was a proxy for bad effects.86 Adapting Kagan’s language here, “[t]he focus on motive . . . provides an indirect way of identifying actions with untoward effects on [the] public [interest]. This identification mechanism is necessarily imprecise—both over- and underinclusive. But given the difficulties of inquiring directly into effects, it may be the best such instrument that [we] can find.”87

Discretion vs. Duty

Apart from serving as a proxy for the national interest, motive might also play a unique role in areas of discretion. In short, looking to motive might be most justified when the President exercises discretionary authority, i.e., where she “May Act,” while looking to motive may be less justified when the President is acting pursuant to a duty, i.e., where she “Must Act.”

Areas of Discretion: “Must Act” vs. “May Act”

Recent work on the relevance of motive has suggested that motive might be least relevant in areas where someone “must act,” but most relevant when dealing with discretionary decisions that “involve actions by repeat players who have time to deliberate about their decisions and who exercise discretion in distributing benefits and burdens.”88 This distinction is important for understanding presidential motive, because the President operates largely in such areas of discretion.

To explain why it might matter whether the President is acting in an area of discretion rather than duty, it is helpful to recall some of the nonconsequentialist arguments against looking to motive. Such arguments frequently look to hypotheticals that play on intuitions that suggest internal motivation ought not matter to whether someone should engage in particular conduct. For example, if someone sees a child drowning in a pond, the right thing to do is save the child, regardless of whether it is done for altruistic, selfish, or even hateful reasons.89 Similarly, as Fallon suggests, we might share the intuition that murder ought to be prohibited—even if it is prohibited for impermissibly religious reasons.90

But relying on such examples to show that motive is irrelevant has been subject to a compelling critique that such examples exclusively focus on situations where someone “must act.”91 It is clear the passerby must save the child (morally speaking), and that the legislature must prohibit murder (morally speaking). But, as S. Matthew Liao has argued, when we pivot from “must act” to “may act” situations, our intuitions often change.92

Liao makes this point with a variation of the famous “Trolley Problem,” where a runaway trolly is coming toward five innocent people who will be killed if nothing is done.93 Liao posits that someone can push a button to direct the trolley onto a separate track where it will kill one innocent bystander, and suggests that pushing the button might be permissible if done to save the innocent people, but impermissible if done because the person “hates the innocent bystander on the side track, . . . and he sees this as his opportunity to kill” them.94 If done for that reason, the person has impermissibly committed “murder[] [of] the bystander.”95 Micah Schwartzman provides another example, posing a diplomat who has the power to give a visa to one of two people. If he gives the visa to person X and not person Y because of person Y’s religion, that act is wrong, even if he otherwise could have given it to X for valid reasons.96 Similarly, if a firm hires new people via a staff vote, and someone votes not to hire someone because of their race, that vote is wrong, even if the person could have voted against the person for valid reasons and even if the person is hired nonetheless.97 In short, in areas of discretion, motive may make otherwise permissible action impermissible.

On this view, if the President is exercising discretionary authority where she has a choice about whether and when to use power, then her motive can be relevant, even if it would not be relevant in contexts where she must act to serve the country. This has important implications for the President. Much of the President’s power lies in areas where she has discretion to exercise government resources, rather than in areas where she is obligated to do so. For example, according to the Office of Legal Counsel, the President has discretion to use military force abroad so long as it furthers sufficiently important national interests, like regional stability or humanitarian concerns.98 There are seemingly endless places where American military force might be thought to serve such vague goals.99 When deciding whether and when to use force for this purpose, the President is exercising discretion—it is typically a “may act” not “must act” situation.100 In this context, then, it seems justifiable to ask whether or not the President seeks to use military power to further her own personal interest—say to win an election—rather than the national interest.

War powers are not unique in this regard. Many areas of presidential power rest on discretionary authority, subject to extremely vague conditions.101 Given how much presidential power lies in such areas of discretionary power, there is strong ground for Article II’s motive requirement.

Enforcement Prioritization

Apart from these primarily nonconsequentialist reasons, there are also consequentialist reasons to look to motive in areas of discretion. Given how much discretionary power the President has, much of what the President does is set enforcement priorities.102 The question for the President, then, is often not which exercise of power is in the national interest, but, Among exercises of power ostensibly in the national interest, which should I prioritize? The President has limited resources, and using power in one area will often mean not using it another. So even if we cabin our view only to exercises of power that are, in fact, in the national interest to some extent, which nationally interested exercise of power to pursue can have consequences. Publicly motivated conduct is likely to be more beneficial for the country than personally motivated conduct, even if both are beneficial for the country by some degree. And, given the President’s limited resources, personally interested conduct is likely to trade off against publicly interested conduct. This provides another justification for looking to motive. We should encourage the President to focus her efforts on action motivated by furthering the public, rather than her self-interest, because publicly motivated conduct is likely to be better for the country than privately interested conduct. This is an empirical claim, but a highly plausible one.

In short, given that much of what the President does is choose between exercises of power ostensibly in the national interest, how the President chooses priorities is deeply important. Setting priorities based on public- rather than private-interested motive is likely to benefit the country and provides another justification for motive inquiry.


Above I have argued that inquiry into motive is more justifiable in “may act” situations of discretionary authority. The corollary is that inquiry into motive seems least justified when dealing with “must act” situations—such as saving the drowning child in the lake or criminalizing murder. What does this mean for the President?

We might think of “must act” situations for the President as involving what we conventionally call “duties.” In some situations, the use of the authority is so obviously necessary as to require it for the President to fulfill her oath to “preserve, protect and defend the Constitution of the United States” or “take Care that the Laws be faithfully executed.”103 Take the order to shoot down the plane carrying the nuclear weapon. In such instances, it is harder to justify looking to the President’s motive. The President should order the strike, regardless of his reasons for it.

Beyond such instances, Congress sometimes imposes statutory duties on the President to act in certain prescribed instances. For example, the President is required to sanction certain individuals if presented with sufficient evidence that they have knowingly and materially aided certain chemical weapons programs.104 As the Office of Legal Counsel put it, “[t]he language and purpose of [this] Act demonstrate that the President has a duty to make determinations, not merely the discretion to do so.”105 In such situations, it makes sense to think of motive as less relevant. Congress has set forward the conditions that require the President to act—in these situations she “must act.” Thus, where the President has a duty to act, looking to motive is less justifiable than when she has discretion to act (or not).106

Expanding the Frame

The focus of the discussion so far has been largely concerned with how to think about particular exercises of power.107 However, such an approach might be too myopic. Rather than looking at discrete uses of power, we might expand the frame beyond one-off transactions. Doing so provides further justification to look to motive.

Pathological Presidents

One danger of permitting personally motivated conduct that happens to be in the national interest is that doing so might serve to benefit a President inclined to act based on self-, rather than national, interest. In other words, a President motivated to act by self-interest in one instance is likely to do so in others. Thus, even if such conduct happens to be in the public interest in one scenario, we might not wish to permit a self-interested President to take such conduct, particularly if it is likely to strengthen her power. The basic assumption here is that a President who uses public power for personally interested reasons in one instance, is likely to do so in other instances, and focusing on a one-off situation where the conduct happens to be in the national interest sidelines the broader goal of avoiding empowering a President inclined to use public power for private ends.108

In short, we might be better off making all personally interested conduct illegal—even when it helps the country—if doing so will avoid facilitating the agenda of a President inclined to act based on their own personal interest, rather than the public interest.

Dynamic Effects

Beyond conduct helping a President who is pathological, permitting conduct motivated by personal interest might also create bad dynamic effects. If people know that the President is inclined to further her own interest, they might start engaging in conduct they would not otherwise have taken aimed at the President’s personal interest. For example, private actors might try to curry favor with the President’s personal interests in order to get the President to use public power for their own purposes. Thus, even if personally interested conduct is in the national interest in a particular instance, it might lead to worse outcomes if we incorporate dynamic effects triggered by permitting such conduct. These dynamic effects provide an additional reason to prohibit personally motivated conduct, even if it happens to be in the national interest.

Changing the Meaning of Presidential Action

Another way in which the President’s personal motive might render conduct wrongful even if it happens to be in the national interest is if the President’s personally interested motive changes the meaning of that act based on its expressive effects.109 For example, even if the President could refuse to grant disaster relief funds to a state for permissible grounds, if he refuses to grant the aid because the state’s citizens did not vote for him, this can change the meaning of the decision not to grant funds.110 If made for such politically vindictive reasons, the decision expresses the notion that the President only seeks to serve the people who vote for him—that the government is not for all the people but only for certain people.111 The reason for the action thus creates an independent harm by changing its meaning.

Changing the meaning of presidential action is no small thing. People do not just care about bottom line policies: They care about the reasons why they were enacted.112 If the population experiences the President’s actions as signaling that the President is not working for the public, but working for himself, this can have serious consequences for the health of our democratic system. If people believe the President is personally interested—that elected officials use public power as a means toward achieving personal ends—this will make accepting the rough-and-tumble, sometimes-you-win-sometimes-you-lose nature of democratic politics all the harder to accept. Motive’s ability to change the meaning of presidential exercises of power thus provides another reason to inquire into the President’s motive.113

Downsides of Motive Inquiry—Looking Inward

The reasons above provide justifications for looking to presidential motive as relevant to the legality of presidential conduct. However, one concern about such motive inquiry is that, when decisionmakers are deciding whether to act, motive inquiry calls on them to look internally—to what is happening in their head—rather than externally—to what is happening in the world.114 This is thought to divert the decisionmaker’s attention away from what really matters.

Scanlon poses a hypothetical to make this point, asking us to imagine that a Prime Minister has asked whether it is morally permissible to authorize a military raid to destroy a weapons factory that is likely to kill a number of innocent civilians. Scanlon suggests we would not respond by stating “[w]ell, that depends on what your intentions would be in carrying it out. Would you be intending to kill the civilians, or would their deaths be merely an unintended but foreseeable . . . side effect of the destruction of the plant?”115 As Scanlon notes, there is something odd about asking the Prime Minister—or the President—to look inward to their motives, rather than outward to the world, when deciding whether to bomb a munitions plant.

But even if looking exclusively inward would seem odd, that does not mean motive inquiry is unjustified. For the reasons given above, a good-faith President might well want to ask herself about her own personal motive before exercising power. If self-interested conduct is likely to be a good proxy for conduct not in the national interest, then reminding the President of this seems wise. If prioritizing exercises of power based on self-interest rather than the national interest is likely to be worse for the country, then reminding the President of this seems wise. If exercising power based on personal interest can create expressive harms or lead to bad dynamic effects, then asking the President to look inwardly seems wise. This is not to suggest the President ought only to look inward—but looking inwardly in addition to outwardly is justifiable for these reasons.116

Of course, in normal circumstances, the White House Counsel or Office of Legal Counsel would not tell the President that a proposed strike’s legality depended on her motive, rather than on whether she viewed the strike as furthering particular national interests. But that is because they would likely not suspect the President would take such action based on personal interest. If they had reason to believe the President was personally motivated, it would be perfectly appropriate to remind the President that she can take the action if she is motivated by the national interest, but that if her purpose in ordering the strike is to further personal interests, then it would be unlawful.

Sum—A Practical Inquiry

In my view, the reasons above provide persuasive justification for Article II’s requirement that the President must be motivated by the public interest, rather than her personal interest. But the justifications are stronger in some areas than in others. Looking to motive is most justifiable in areas of presidential discretion where power is premised on nonverifiable conditions. It is in such areas where motive is most likely to serve as a useful proxy for the national interest or other nonverifiable conditions and where the President is exercising discretion rendering motive more relevant to the permissibility of conduct. Conversely, looking to motive is least justified in areas of duty premised on verifiable conditions or where the national interest is obvious. In such areas, motive will serve as less useful of a proxy and is less relevant to permissibility if the President simply “must act.” The image below seeks to provide a visual spectrum showing where motive inquiry is most and least justified.

The above image is pitched in relative terms. One might still wonder whether the general limit that Article II imposes on what motives can drive presidential conduct can be normatively justified in absolute terms. This requires a practical inquiry taking into account when motive inquiry is most and least justified in general and applying them to the peculiar space of presidential power. Given the reasons above and the fact that presidential power is mostly allocated in areas of discretion premised on nonverifiable conditions, requiring the President to be motivated by the public interest when acting strikes me as normatively justified.

Before moving on, it is worth making explicit the empirical assumptions my analysis depends on. If these assumptions are false, this could change the case for motive’s normative justifiability. First, I assume personal motive would serve as a proxy for conduct not in the national interest. Second, I assume most important exercises of presidential power lie in areas of discretion, rather than duty. Third, I assume that presidential enforcement or discretionary priorities trade off against each other due to limited resources, such that enforcement in one domain is likely to mean less enforcement in another. Fourth, I assume that a president who takes action for self-interested reasons in one instance is likely to do so in others and that private actors or foreign governments might recognize this and seek to capitalize on the President’s personal interested behavior. Fifth, I assume that if the President takes action for personal interest, people will recognize this, and it can change the meaning of his or her conduct. And, finally, I assume that, overall, the benefits of looking to motive (such as preventing self-interested behavior and propping up a bad president) outweigh the costs (of potentially prohibiting conduct that is in fact in the public interest). Some of these claims might be more empirically falsifiable than others, but it is important to clarify what they are.

Undoubtedly, limiting which motives can drive presidential conduct will have costs. It means that, at least in theory, some exercises of power that are in the national interest would be prohibited simply because of the internal reasons driving them. Conversely, there will be instances where conduct is public-interested in motivation, but bad for the country in fact. The rule is thus both over- and under-inclusive. But such is the way when constructing rules of governance.117 Notwithstanding these concerns, there is good reason to defend—indeed laud—the Constitution’s requirement that the President must be motivated by the public, rather than private interest. Although some hypotheticals may give us pause, given the landscape of presidential power as it exists today, the President’s motive requirement is justified.

Permissible and Impermissible Presidential Motives

Above I have argued that Article II requires that the President must have a public-regarding motive in exercising power and that such a requirement is normatively defensible. But, even if we accept both these claims, many difficult questions remain. First, if acting motivated by the public interest is required, what precisely constitutes a public-interested rather than private-interested motive? In the realm of political officials, this is not an easy question. Second, however we define personal- or public-interested motive, what do we do in cases of “mixed motive,” where both permissible and impermissible motives are present? This Section takes each of these questions in turn.

What Is Personal Motive?

What does it mean for a president to act for personal rather than public-interested motivation? Some cases are easy. A business owner gives the President ten million dollars to order an investigation into the business owner’s rival. If the President orders the investigation for the purpose of receiving the money, that is an impermissible personally interested motive because it is driven by a desire to bolster the President’s personal fortune. It is not altogether difficult to construct such easy cases.118

But what about the President’s personal political interest? What if the President enacts a policy because she thinks it will make her more popular nationally or with a certain constituency? What if she does so despite thinking the policy is not good for the country? Has the President just acted impermissibly? Is her motive personally or publicly interested? On this question, there is disagreement.

Below I provide a straightforward answer to this question: Acting to pursue the President’s political interest is a personal reason, not a public-regarding reason, and is therefore impermissible. I then examine theoretical and practical objections to this claim, concluding that these objections are ultimately unpersuasive. The Section then explores the viability of a narrower test geared at permitting more run-of-the-mill political motives but prohibiting more egregious motives that seek to undermine electoral competition, but ultimately concludes that finding a coherent dividing line between permissible and impermissible political motives on this ground proves too difficult. At bottom, the straightforward answer is the best one we have.

This Section begins by treating the President as acting for a single motive, rather than with mixed motives. This approach enables us to focus first on precisely which motives are permissible and which are not. With that understanding, we can then examine how best to address “mixed motive” cases in the section that follows.

The Straightforward Answer: Political Interest Is Personal Interest

Does the President act in a personally interested or public-interested way when she exercises power in order to help her political fortunes—e.g., to make herself more popular or bolster her chances for reelection? I propose a straightforward answer: If the President acts for her political interest, such action is personally, rather than publicly, motivated. To qualify as public-interested action, the reason for action must be to do what is good for the country.

Recall that Article II imposes a duty to act motivated by the public interest. Thus, if the President acts solely for a reason other than the public interest, then she has violated her Article II duty in this respect.119 This is obvious when we think of financially motivated action. If the President enacts policy X because a company has paid her to do so, we can see this was motivated by the President’s personal financial interest. This motive is distinct from acting for the public interest and is clearly impermissible. Now imagine the President enacts the same policy because it will help her get elected: The policy is motivated by the President’s personal political interest. It is not clear why replacing the word “financial” with the word “political” ought to do any work. Neither the President’s financial nor political fortune are equivalent to the national interest and therefore they cannot be the sole reason for acting. Acting for a personal financial interest might be worse than acting for a personal political interest in some sense, but neither is equivalent to acting for the reason of serving the country, and thus both are impermissible. Enacting policy X might happen to be otherwise in the national interest, but if it is not done for that reason, Article II prohibits it. Under this straightforward answer, then, conduct taken in order to help the President’s political fortunes is personally, not publicly, interested.

This straightforward answer makes some hypotheticals easy. Take one created by Hemel and Posner (and discussed at greater length below): Imagine that the President is considering adopting a “soft on pot” policy. Whether the President can do so depends on whether he does so because he thinks the policy will serve the national interest. But what if he, instead, adopts the policy solely to “win votes,” thinking it is in fact a bad policy for the country?120 Hemel and Posner conclude this would present a “closer call” but that such a decision would be permissible in the context of criminal obstruction of justice.121 If we accept the straightforward answer above, this is not a close call for purposes of Article II. If the President adopts the policy solely in order to “win votes,” he has done so for a reason distinct from pursuing the national interest and the action is impermissible.122

Again, this is a straightforward answer. But is it too simplistic? Below I take on some potential objections.

Theoretical Objections: Delegates and Trustees

One objection to the claim that political interest is a personal interest is that we ought not conflate political interest with personal interest in the context of democratic representatives. On this view, because the President is an elected official, she is meant to be motivated by gaining political popularity, and it would be bizarre to conclude the Constitution prohibits such a motivation.

This objection relates to a longstanding debate in political theory about the proper conception of political representation. As noted above, the standard conception of a political representative requires the representative to act for the benefit of her constituents.123 But how precisely the representative is meant to act for the benefit of her constituents is deeply contested.124 The standard debate is between the “delegate” and “trustee” conceptions of representation.125 To put it simply, under the “delegate” model, representatives are supposed to do whatever their constituents want or would want, while, under the “trustee” model, representatives are supposed to do what the representative thinks is in the best interests of their constituents, albeit perhaps informed by what her constituents want.126 Thus, under the delegate model, the representative defers to her constituents about what is good for them, and under the trustee model, the representative is the ultimate decisionmaker of what is good for her constituents.127

For the trustee model, the objection has little force. It is clear that the representative is not supposed to do whatever is likely to make her popular among her constituents. Instead, she is supposed to reflect on what is in her constituents’ interests and act accordingly. Distinguishing between political and national interest will not be difficult on this conception. Conduct taken in the public interest will be taken because of her own judgment of what is good for her constituents, rather than simply what might be popular among them.

Some versions of the delegate conception, however, present challenges to the straightforward answer given above. Under at least some conceptions of the delegate model, the representative is supposed to do whatever her constituents would want—i.e., what would be popular among her constituents—and ought to be largely motivated by fear of electoral sanction—i.e., of being kicked out of office for failing to do what her constituents want.128 Under such a conception, before the President takes action she might legitimately ask herself: What is likely to make me popular, and what is likely to result in my loss of popularity? And this would be such a representative’s method of acting in the national interest. Thus, on some “delegate” conceptions of representation, being motivated by “winning votes” might be synonymous with being motivated by “the national interest.”129 Such a representative might feel required to take action that in her view is bad for the country, because it is not her view that matters; it is that of her constituents, and she is supposed to be motivated by winning elections. For a representative that adopts this view, the straightforward answer will fail to reliably distinguish between personal and public-interested reasons for acting.

But, importantly, this complication only arises if the President views herself as a delegate in these terms.130 It is only where the President believes that the public interest is discovered by asking how the electorate would respond to her decisions that the President’s personal political and public-interested motive collapse together. But Presidents are unlikely to conceive of themselves in this role, and under most contemporary theoretical accounts and those of the Founding era, it does not appear that they ought to.

First, it seems unlikely that Presidents, in fact, adopt such a pure “Delegate” conception of their role as a representative. Although I am not aware of survey evidence of past Presidents on this issue, surveys of members of Congress have revealed they are not committed delegates,131 and Presidents certainly do not speak as if they are bound by what is popular rather than by doing what is good for the country.132 Moreover, even with the rise of conceptions of the “plebiscitary president,” much recent work on the President has argued that the President is not motivated primarily by what is popular or even reelection, but by other considerations, such as exhibiting “leadership” or enhancing their “legacy.”133 Although blanket characterizations about what motivates all Presidents all the time are hard to sustain, there is little evidence that Presidents view themselves as devout adherents of the delegate model of representation in political theory terms. They do not seem to act that way and they do not seem to speak that way.

Moreover, it seems clear that the Framers did not envision the President as operating as a true delegate in this sense,134 and most contemporary theorists have rejected such a pure delegate theory on normative grounds.135 As Hanna Pitkin put it in her classic work on the topic, a representative

must act in [the constituents’] interest, period. [The constituents’] view of their [own] interest may or may not be definitive, depending on the issue and the situation; but if [a representative] follows it, it should be because the action really accords with their interest, not because they merely think it does.136

In short, under most contemporary views of representation, there is a role for the President’s judgment as to what is, in fact, good for the public that goes beyond simply asking what is likely to make them popular or win them votes.137 And, if that is the case, one can distinguish between acting for the reason of serving one’s political interest and acting for the reason of serving the public’s interest.138

None of this is to say that the President never can or ought to consider what is likely to be popular in making decisions. The President will often want to know whether something is going to be popular in order to help determine the public interest. The President is a representative of the people, and the people will often be an important source of information about what is in their interest, even if they are not always the final authority.139 And for many important policies, considering the popularity of an action will often be required to assess whether the action will be in the national interest. For example, a use of military force could in theory be in the national interest because it could achieve certain national security goals. But if the public will not support the war effort long enough to accomplish these goals, then using force would not, in fact, be in the national interest. The straightforward distinction thus does not require the President to act completely oblivious to popular opinion. It simply prevents the President from acting motivated solely by her political fortunes, rather than by doing what is in the national interest.

In sum, there may be some theoretical conceptions of representation that render the straightforward distinction problematic.140 But it does not appear the President holds such views or that she ought to under most prevailing theories of representation today or those of the Founding era. So long as the President believes that acting in the public interest often requires her to exercise independent judgment as to what is good for the country, rather than simply asking what will help her political campaign, then she can distinguish between conduct motivated by her political interest and conduct motivated by what is good for the country. The straightforward distinction thus remains viable—the President’s political interest is a personal, not public, interest.

Practical Objections

Beyond theoretical objections, there are also practical objections to concluding that the President’s political interest is an impermissible personal interest. If Presidents, in fact, act all the time based on winning reelection, rather than the public interest, this straightforward distinction would mean that Presidents are constantly acting unconstitutionally.141 If true, the distinction might be thought to set the bar too high, creating a utopian standard that fails to correspond with the world we live in.142

At its core, this objection rests on an empirical claim about the frequency of certain motives driving presidential conduct.143 I doubt anyone could rigorously test this claim empirically, but I confess that I am not sure it is true.144

To be sure, Presidents frequently take actions that, in fact, help them politically. But it is hard to know if such action is motivated by political benefit, rather than by their conception of the public interest, which is likely influenced by their own political ideologies and thus rewarded by their political allies. In their work on criminal obstruction of justice, Hemel and Posner state that “it is accepted that [the President] can use the powers of his office to advance his party’s interests as well as his own political interest in winning reelection or having a member of his party succeed him in office.”145 Although it is likely true that the public might not view such conduct as qualifying as criminal, I have not seen any studies establishing that the public believes that the President can legitimately use official power motivated by serving her political interest, rather than the public interest.146 Nor do Presidents seem to explain their conduct in this way explicitly. When President Trump came close to doing so, he was typically criticized for it, as evidenced by the significant backlash to, for example, his using the White House grounds for a campaign speech or clearing the crowd in Lafayette Square for a photo opportunity.147 Supporters of any President will often engage in motivated reasoning to support beliefs that a President they like is engaged in permissible conduct (or that a President they do not like is engaged in impermissible conduct), but using public office for the purpose of personal political gain does not appear generally accepted.148

If empirical intuitions or public understanding do not adequately tell us how often the President acts solely motivated by his political interest, we might look to the President’s incentives. But the President’s incentives do not clearly push her to act primarily for her political interest either. Although Presidents surely want to be popular, scholars of the Presidency have argued that presidents have incentives other than their short-term political interest driving their conduct, such as attempts to bolster their historical legacy.149 And, of course, Presidents in their second term are not up for reelection at all. In short, Presidents are likely to have multiple incentives driving their conduct: a desire to be popular, a desire to bolster their place in history, a desire to do what is good for the country. I see little reason to assume that the President’s electoral incentive is the exclusive or even primary driver of presidential conduct.150

Although Presidents likely engage in conduct motivated by their political interest some of the time, it does not appear that conduct motivated solely by such considerations is so common as to make it untenable to conclude the President’s political interest is an impermissible personal motive.151 Given that this objection rests on an empirical claim that has not been proven and is not, in my view, obvious, it does not strike me as persuasive. Moreover, in my view, it is better to set the bar high to encourage a better ethic of presidential decision-making, rather than treating the more cynical interpretation of events as unproblematic.

The “Rules of the Game” Alternative

Even if the theoretical and practical objections above do not require abandoning this answer on their own, they do raise valid concerns about it. Adopting this standard might call into question a lot of presidential conduct that appears politically motivated. Moreover, it bundles together all sorts of politically interested conduct, some of which is more egregious than others. These concerns might push toward finding a narrower distinction that focuses on preventing the most egregious politically motivated behavior—that which interferes with the fundamentals of our democratic system—while permitting more routine, and perhaps less problematic, politically motivated behavior. Such a test would focus on ensuring the democratic “Rules of the Game” are complied with, while permitting political motives that do not interfere with those rules.152 In this section, I examine the most developed variation of this approach to see if it provides a more desirable way of distinguishing between personal and public-interested motivations than the straightforward answer. Ultimately, I conclude that it fails to provide a more workable or coherent approach than the straightforward answer above.

Professors Daniel Hemel and Eric Posner have proposed the most developed variation of the “Rules of the Game” approach to identifying permissible and impermissible political motive in their work on presidential obstruction of justice. To be clear, Hemel and Posner set forth their test in the more limited context of how to assess which motives render conduct criminally liable as obstruction of justice. They never claim their approach should apply more broadly. I use their test as the standard to engage with for broader application, because it is the most thoughtful and developed form of this distinction in the literature.

According to Hemel and Posner, “[a] president commits obstruction of justice when he significantly interferes with an investigation, prosecution, or other law enforcement action to advance narrowly personal, pecuniary, or partisan interests.”153 With respect to what constitutes an impermissible politically partisan motive, Hemel and Posner conclude that the President is permitted to “use the powers of his office to advance his party’s interests as well as his own political interest in winning reelection or having a member of his party succeed him in office.”154 But, they conclude that not all personal political or partisan motive is permitted. Hemel and Posner argue that the key dividing line between what is permitted and what is not is

between actions that are consistent with the ideal of political competition and those that are not. The former include actions that benefit the president or his party politically because they advance a policy agenda of which the public approves. The latter include actions that benefit the president or his party by making it difficult for political opponents to make their case to the public.155

This is their variation of the “Rules of the Game” approach. Does this dividing line provide a superior approach to the more straightforward answer given above? For the reasons below, I conclude it does not.

The first thing to note is that Hemel and Posner’s test is not phrased as depending on motive, rather than conduct. The key is distinguishing between “actions that are consistent with the ideal of political competition and those that are not.”156 This could be read as depending on consequences, not motives, i.e., by asking whether the actions, in fact, benefit the president by advancing a popular policy agenda or by making it difficult for the President’s “opponent[] to make their case to the public,” rather than asking whether this is what motivated the actions.157 We can make the test more explicitly motive-based by stating that the President can act motivated by making himself more popular through popular policies, but not motivated by making his opponent’s ability “to make their case to the public” more difficult.158 But, either way, this sets up an odd juxtaposition. It seems to permit the President to act motivated by furthering her own political fortune, but not by harming the political fortunes of her opponent. But it is not clear why it would be permitted for the President to use official power to increase her popularity but not take action motivated by decreasing her opponents’ popularity. After all, in many instances, making herself more popular will make her opponent less popular, and vice versa. A politician can typically further her electoral interest by increasing her votes or by decreasing the vote share of her opponent.159 The result is the same—a political benefit for the President. It is not clear why one should be categorically permitted and the other excluded.

This exemplifies what I see as the fundamental challenge in constructing a motive test based on the democratic “rules of the game” or, in Hemel’s and Posner’s formulation, the “ideal of political competition.”160 It is not clear precisely what it would mean to be motivated by acting within those rules or ideals. This would seem to require us to define permissible motives based on some sort of conception of being a good sport trying to play by the rules. But what are the rules? If the President can take public action to make himself more popular, why can’t he take public action to make his opponent less popular? If his motive is to serve his party’s ends, why can’t his reason be to harm the opposing party’s fortunes? If the dividing line is based on the “rules” or “ideal[s]” of political competition, then we need to know what those rules or ideals are, and I am not sure there is any agreement on this.

Hemel and Posner admirably try to flesh out this distinction by way of several hypotheticals. But despite their valiant efforts to provide a principled and clear line based on actions’ consistency with the “ideals of political competition,” the analysis strikes me as more contestable, and less coherent, than concluding that acting to further the President’s political interest is an impermissible, personally interested motive.

Hemel and Posner pose three hypotheticals to flesh out their distinction: In one, the President orders the Justice Department to stop prosecuting marijuana violations because he thinks such prosecutions are a poor use of government resources; in the second, the President does so because he thinks such a policy will “draw younger voters to his party”; and, in the third, he orders the Justice Department to drop a case involving a particular Senator of his own party standing for reelection next month.161 They conclude the first and third scenarios provide easy answers: The first is permissibly publicly motivated and the third is not.

Why the third hypothetical is clearly impermissible on their view is worth examining. Although Hemel and Posner do not specify the President’s motive in this scenario, we might assume that the motive is to prevent a politically harmful prosecution from hurting the Senator’s chances of winning the election the following month. Under the straightforward answer given above, the President’s conduct would be impermissible because the motive is not public-interested, it is personally interested—motivated by serving the political fortunes of the President’s party, rather than the country.

Hemel and Posner do not rely on this distinction as what drives the impermissibility—after all, they have concluded that the President can act to “advance his party’s interests . . . in winning reelection.”162 Instead, they conclude that the issue is that the President acts “corruptly[] when he uses prosecutorial power to harass his political enemies while sparing his friends. . . . [T]he president cannot . . . abuse his position of power to distort electoral outcomes by enforcing generally applicable laws only against political enemies.”163 Hemel and Posner add several facts here. Not only is the President acting for partisan gain, but he is: (1) enforcing generally applicable laws only against political enemies; (2) sparing his friends; and thereby, (3) distorting electoral outcomes. Perhaps it is these additional facts that establish a violation of the “ideal of political competition.”

But it is not clear why these additional facts are necessary to make the conduct wrongful. Imagine we remove condition (1) and the President was not generally applying this law against political enemies—the only act he took was to (2) spare his political ally from prosecution. No other political actors were involved in any related prosecution. Would that change whether this ought to be permissible? It is hard to see why it would. The President is still using government power to protect a political ally from prosecution in order to help that ally win an election. The condition that the President is otherwise harming political enemies thus does not seem necessary to conclude the President is acting impermissibly here. What about the fact that the President is (3) “distort[ing] electoral outcomes”?164 Imagine that the Senator would win election even if the prosecution went forward—therefore stopping the investigation does not “distort” the outcome, because he wins regardless (or imagine he would lose regardless). Does that mean the conduct was permissibly motivated? Or in line with the “ideal of political competition”? Again, it is hard to see why it would.

In short, it is not clear that the wrongfulness of the conduct comes from violating the “ideal of political competition,” nor is it entirely clear what that ideal would require here. It is far simpler to characterize the wrong here as the President’s use of official power to serve political ends—he sought to stop an investigation in order to serve his and his party’s political ends rather than those of the country. It is true that it would be worse if it were combined with using the same power to target political opponents and change the result of an election, but those two factors do not seem necessary to conclude the conduct was impermissibly motivated. In my view, the straightforward distinction thus does a better job of identifying the violation here than a distinction relying on differentiating motives consistent or inconsistent with the “ideal of political competition.”

Hemel and Posner’s second hypothetical is also illuminating. There, the President adopts a “soft on pot policy” exclusively “to win votes—he thinks it is otherwise a bad policy.”165 The President also observes that adopting “the policy would throw the opposing party into turmoil, destroying its electoral prospects for years to come.”166 Hemel and Posner conclude this action is a “closer question” but ultimately permissible, i.e., in line with the “ideal of political competition.”

But why is that so? Here the President has used his official enforcement power to further his personal political agenda. As is common, not only does the act help him, but it hurts his opponent. In fact, it will throw his political opponents into disarray and “destroy[] [their] electoral prospects for years to come.”167 It is not clear why this is thought to be consistent with the ideal of political competition. One might think that an electoral contest between an incumbent President and her political opponent ought to be fought in the political realm using political, rather than governmental, tools. Indeed, one could even make the case that this hypothetical provides an example of the President abusing his official power in a way likely to “distort the electoral outcome”—falling within Hemel and Posner’s definition of prohibited conduct. The President has used his public power to alter the political landscape—to win votes for himself and to throw his opponent into disarray.

Hemel and Posner note that the electoral outcome here is not distorted through the particular mechanism of “enforcing generally applicable laws only against political enemies.”168 They state that the President cannot

single out targets of law enforcement for harassment or immunity based on their partisan leanings. This type of partisan or political discrimination undermines political competition by forcing the party out of power to devote resources to fend off prosecutions . . . based on behavior that is no different from that of the president’s supporters . . . .169

But, again, it is not clear why the ideal of political competition permits distortion of the electoral outcome through manipulation of public policy so long as it does not operate by way of applying enforcement resources only against political enemies. Singling out opponents for harassment certainly seems inconsistent with the “ideal of political competition,” but such harassment—while sufficient—does not seem necessary to violate such ideals. The core wrong seems to be the use of official power for political ends—the wrong the straightforward answer identifies—rather than the violation of the “ideal of political competition.”

A final example of the difficulty in drawing the line based on the “ideal of political competition” might help. Hemel and Posner conclude that President Obama’s decision not to prosecute former Bush administration officials for torture was permissible, even “if his real motive was to avoid partisan attacks that might have jeopardized his legislative priorities and threatened his presidency.”170 On the other hand, they conclude that it would be impermissible for a President to stop an investigation into a friend or aide because he believed if the investigation came to light the President “would not be able to obtain the votes for a health care reform bill.”171 This action would be impermissible because “[m]anipulating the conduct of criminal investigations in order to sway the outcome of congressional votes is flatly inconsistent with the norms of political competition and persuasion that undergird a constitutional democracy.”172 But why would a motive “to avoid partisan attacks that might jeopardize legislative priorities” be permissible, but a motive to ensure the President was “able to obtain the votes for a health care reform bill” be impermissible? In both cases, the conduct was (1) to prevent prosecutions (2) in order to gain votes for preferred legislation. It is hard to see why one is “consistent with norms of political competition” and the other is not.

Of course, one can easily differentiate the cases. If President Obama was motivated by avoiding “criminalizing political differences—an important norm in democratic politics,”173 then this is a permissible public-interested motive. Whereas if the President stopped a prosecution just to avoid a political scandal, this would be impermissible. But this distinction is premised on the straightforward answer given above: One is motivated by serving the public, the other by serving the political interests of the President. The former is permitted, and the latter is not.

To be clear, as noted above, Hemel and Posner never claim that their proposal ought to apply as a general matter for Article II and their distinction might be the best one for their area of focus, criminal obstruction of justice, where it seems extremely important to be under rather than overinclusive. But, as a broader attempt to distinguish personal from public-interested behavior in the context of political motive, a test based on the “Rules of the Game” or the “ideal of political competition,” strikes me as unworkable. Perhaps there is another way to construct a variant of the Rules of the Game that works better than the straightforward answer above.174 But, for now, the straightforward answer remains the best one we have. If the President acts motivated by her political interest, this is an impermissible reason for acting. She must act motivated by doing what is good for the country, not what is good for herself.

Mixed Motive

The section above provides a dividing line between permissible and impermissible motives. Permissible motives include acting to further the national interest. Impermissible motives include acting to further the President’s personal interest, including her or her allies’ political interest. But what about situations where the President has “mixed” motives—when she acts for both public-interested and personally interested reasons?

The Constitution does not tell us how to deal with this question; it must be resolved by weighing normative and practical considerations. This part first identifies prominent options for mixed motive standards, then provides a proposed mixed motive test for the President. It concludes by providing a framework to think about how to adjust the proposed standard depending on one’s views of the egregiousness of the impermissible motives, empirical assumptions about their frequency, and the normative justifiability of looking to motive at all. 

The Mixed Motive Menu

In some cases, the President will act based both on public-interested and personally interested reasons. Sometimes the public-interested reason will be greater than the personally interested reason, and sometimes it will be the other way around. Sometimes the President would have taken action based solely on the public interest, and, in others, she would not have acted if not for her personal motive. There is nothing new about mixed motive cases like these. As Andrew Verstein has shown, numerous areas of doctrine require courts to assess such mixed motive cases.175 Courts have developed an array of tests, most of which can be put into one of four buckets.176

First, under a “Primary Motive” test, conduct is permitted if the permissible motive is greater than the impermissible motive.177 Under this test, if the President’s public-interested reason for acting is stronger than her private-interested reason, then the conduct would be permissible. If the private-interested reason were stronger, it would not be. It would not matter if the President’s personal or public-interested reasons were sufficient to take action—all that matters is which motive is greater.178 Second, under a “But-for Motive” standard, conduct is permitted if it would have been taken for the permissible reason alone, but not if the person would not have acted but for the impermissible motive.179 For the President, this would mean that if she would have acted motivated by the public interest regardless of her personal interested reason, then the conduct is permissible. If, on the other hand, the President would not have taken the action without her personal motive, the action would be impermissible. Here, which motive is larger or smaller is not dispositive. Third, under a “Sole Motive” standard, conduct is permissible if there is any permissible motive, but impermissible if the sole motive is impermissible.180 Under this standard, if the President acted solely motivated by her private interest, her conduct would be impermissible, but if she had any publicly interested motive, her conduct would be permissible.181 Finally, under an “Any Motive” standard, conduct is impermissible if there is any impermissible motive. Under this standard, no matter how strong the president’s public-interested motivation, if she had any personally interested motive, the conduct would be unlawful.182

A Proposed Test

With these options in mind, I propose a slight variant of a common standard for the President: a “Material Sole Motive” standard. To understand the proposal, it is helpful to clarify that in some mixed motive cases, the person will have both permissible and impermissible motives, but the strength of one of the motives might be so small as to be de minimis—it exists, but is inconsequential; it is not there in a meaningful way.183 Under my proposed “Material Sole Motive” test, the President’s conduct would be permitted so long as she has any material or meaningful public-spirited reason for acting. If, on the other hand, the President’s motive is entirely or almost entirely privately interested—the only public-interest is inconsequential or de minimis—then the conduct would be impermissible.184

The Constitution requires the President to act motivated by the public interest—it does not specify how strong that motive must be. If the President has a significant public-interest in acting, that strikes me as sufficient to render the conduct permissible. This is so even if the President’s privately interested motivation is the primary or even the but-for motive. So long as the conduct is substantially public-interested, I am comfortable concluding that it satisfies the Constitution’s requirement that it be motivated by the public interest.

Whether this standard is optimal is of course disputable. A Material Sole Motive standard is more forgiving than, for example, a “But-For Motive” standard, because it allows for actions that would not have been taken without the President’s personal reasons.185 One can certainly argue that such conduct ought not be permitted. Indeed, a but-for standard has been proposed for presidential obstruction of justice and for claims of animus in the immigration context.186 And such a standard might be appropriate in those domains. But, because personal political motive is less egregious than the motives posited in the obstruction of justice or animus context, a more permissive standard seems justified. Moreover, in this context, some of the practical objections to defining political motive as personally interested strike me as more persuasive. The President frequently acts in situations where political motive appears to be a but-for cause, such that prohibiting such conduct would be rather disruptive. Although such objections do not compel me to abandon the straightforward distinction I have given above for defining what motives are permissible and impermissible, they do push toward a more forgiving mixed motive standard.

Finally, it strikes me that, internally speaking, the President might genuinely view her own motive as public-interested so long as it is substantially so, even if she would not have taken the action without the private-interested conduct. One downside of a but-for motive standard is that it would prohibit conduct where the public-interested motive is significant, or even larger than the private-interested motive, but not sufficient to act.187 Such a scenario can reasonably be described as public-interested and therefore there is a legitimate argument for its being permitted.

The key to the standard I propose is that the public-interested reason must be substantial—that it not be inconsequential, an afterthought. This might help differentiate examples where the President is almost exclusively privately motivated, but we cannot rule out the notion that she also has some small amount of public-interested motivation. Such a tiny amount of public-interested reason cannot save the conduct. For example, if we take President Trump’s solicitation of an investigation into Hunter Biden by the Ukrainian government, the conduct seems to be entirely—or almost entirely—privately interested in motivation. Even if President Trump could genuinely say that some small part of him was motivated by doing what was good for the country, if it was not a meaningful reason for action, then the conduct cannot be said to be public-interested.

As this example suggests, the standard requires the application of judgment as to when reasons for action are “meaningful,” and there will likely be disagreement about when the standard is or is not satisfied. But there is nothing novel about evaluating mixed motive cases, and this standard does not seem meaningfully harder to apply than the common approaches used frequently by courts.

Adjusting the Mixed Motive Standard

Above I have proposed a test for the President that I think satisfyingly balances the normative and practical stakes in dealing with mixed motives. But the choice of which mixed motive test is optimal is a normative one. Acknowledging the difficulty in crafting an ideal mixed motive standard here, I propose a framework for those unsatisfied with my standard, which could be adjusted depending on one’s views of the following three key variables: (1) empirical assumptions about the prevalence of impermissible motives; (2) the egregiousness of the relevant impermissible motive; and (3) the justification for looking at motive at all in the particular exercise of power.

Empirical Assumptions. The test one chooses is likely to vary based on one’s empirical assumptions about how often the President is motivated by the relevant impermissible motive. The more often one thinks the President is, in fact, acting based in part or substantially on privately interested motive, the more inclined one might be to tolerate such motives. The less such motive occurs, the less tolerant one might be. This might help explain why there are stricter mixed motive test for areas of animus, where we might think—and certainly hope—that the prevalence of the impermissible motive is low.188 So, if we assume the President frequently acts based on impermissible political motive, we might be inclined toward a more forgiving mixed motive standard than if we think such conduct is rare.

Egregiousness of Impermissible Motive. We may wish to have stricter mixed motive standards for some impermissible motives than others. The worse the impermissible motive, the more stringent the mixed motive standard we might prefer. For example, we might prefer a more restrictive standard for presidential animus, because it is a narrow and particularly egregious form of impermissible motive.189 Similarly, we might wish to have a stricter standard for personal pecuniary motives than for more commonplace politically motivated conduct, because we view the former as more egregious. This makes sense of Hemel and Posner’s and Ray’s proposed but-for mixed motive standards in the context of criminal obstruction of justice and animus in immigration, respectively.190 Because their definition of impermissible motive is narrower and more egregious, a more restrictive standard seems desirable.191

In short, the egregiousness of one’s definition of impermissible motive might trade off against the stringency of the proper mixed motive test. The less egregious the motive, the more forgiving the standard and vice versa.

Duties v. Authorities. One might also wish to adjust the mixed motive standard based on how justifiable the inquiry into motive is at all. If motive inquiry is most justifiable with respect to authorities premised on nonverifiable conditions and least justified with respect to duties premised on verifiable conditions, then we could adjust our mixed motive standard accordingly. For duties, we might employ a more forgiving standard—because we are more comfortable with impermissible motive in that space—than for authorities, where we would employ a more restrictive standard—because we are less comfortable with impermissible motive in that space.

*    *    *

There is no definitive answer to the right mixed motive standard. This section lays out relevant options, provides a tentative proposal, and seeks to clarify thinking on how the proposal could be adjusted. Even if one is not convinced by the standard I propose, this framework can help structure thinking as to how to improve it going forward.


Above I have argued that Article II limits the permissible motives that can drive presidential conduct, explained that such a limit can be normatively justified—although more so in some areas than others—sought to clarify precisely which presidential motives are permitted and which are not, and how to deal with “mixed motive” cases. This section turns to more concrete explanations of how Article II’s motive limitation could be enforced by our major governmental institutions—the executive, Congress, and the courts.

At the start it is worth emphasizing that the President’s motive limit will never be fully enforced. Like many other areas of constitutional law, this will be an underenforced constitutional rule.192 That said, armed with the normative framework developed above, we might be more accepting of the lack of enforcement in areas where the President’s motive limit is least justified and focus our enforcement efforts on areas where the President’s motive limit is most justified.

Executive Branch

Most presidential power is not robustly reviewed by any external institution—be it Congress or courts.193 The most impactful way to limit violations of the President’s duties would be internally, because internal constraints can operate on all exercises of presidential power, rather than the limited subset subject to robust external review.194 Of course, to develop internal mechanisms to police this limitation, the President would need an incentive to impose such constraints on herself. But it is not hard to envision a President wishing to make a statement that they wish to prevent themselves—and future Presidents—from acting to further their own personal interest.195 And, once installed, such procedural requirements tend to be quite sticky, tending to stay in place in future administrations.196

One way for the President to limit future self-interested behavior would be to impose a reason-giving requirement on presidential directives that would require the President to give a public-regarding reason for any important action.197 This could be done relatively straightforwardly by modifying the existing executive order governing how formal executive orders and proclamations are issued to require that, before such an order can be approved, the president must sign off on a public-regarding reason for the action. The order governing how such directives are issued currently requires certain procedural steps but does not require such a public-interested reason for the directive.198

Although bad-faith presidents could try to manufacture pretextual public-interested reasons in response to this requirement, a formal requirement that actors within the executive branch go through the process of memorializing a public-interested rationale for important presidential conduct is likely to stop at least some exercises of self-interested power and, by forcing discussion of such a rationale, lead to more public-interested action in itself.199 And, for particularly hard-to-justify exercises of power, the requirement to write down the public-interested rationale will make enactment of the policy more difficult, by, among other things, focusing members of the executive branch on the weaknesses of the policy itself, and serving as a “fire alarm” for whistleblowers in the executive branch to the presence of hard-to-justify presidential conduct.200

Of course, process is never free, and this requirement would make it somewhat more costly to enact policies that are genuinely publicly interested. But giving a public-interested rationale is not particularly onerous if the action is, in fact, motivated by the public interest.201 And if the concern is that publicizing such a statement would have bad effects for national security or other reasons, the statement could be purely internal to the executive branch and still provide many of these salutary benefits.202

Following on the normative framework developed above, the need for such a requirement is particularly glaring in areas of discretion premised on nonverifiable conditions. Taking this lesson to heart, the President should require fuller statements in such scenarios and permit more spare reasons in areas of duty premised on objectively verifiable conditions where motive is less important to assessing the validity of conduct.

Although such a reason-giving requirement will not fully enforce the President’s motive requirement, it would limit the ability to engage in self-interested behavior and push the President toward public-interested action. Moreover, the longer such a requirement is in place, the more internal procedural norms in the executive branch will be structured around it, giving it greater force over time.203


Congress could try to prevent self-interested behavior by the President both through ex post and ex ante mechanisms. Ex post, Congress can impeach the President for alleged self-interested behavior, as it recently did in President Trump’s first impeachment. How precisely motive relates to whether conduct qualifies as a “high crime or misdemeanor” is a topic that will have to remain outside the scope of this Article.204 But the dominant, although not exclusive, sentiment among scholars of impeachment is that motive is relevant to whether conduct is impeachable.205 Whether conduct qualifies as a “high crime or misdemeanor” is likely to depend both on motive and the particular conduct in question. Using public resources to purchase a hot dog might be impermissibly motivated but is unlikely to be impeachable. Going to war to further one’s personal political campaign, on the other hand, could be. As Michael Gerhardt has stated, “an impeachable offense requires bad intent and a bad act.”206 Motive can make some actions particularly problematic, and some actions might exhibit extremely poor judgment even without ill-motive.207

Some scholars and advocates have argued that the President’s motive ought to be irrelevant to impeachment largely on pragmatic grounds. For example, President Trump’s first impeachment defense made the argument that permitting impeachment based on subjective motive would render it too prone to abuse.208

Although there are obvious dangers in permitting politicians to impeach based on impermissible motive, those dangers do not seem distinct from the dangers of politicians impeaching based on other pretextual considerations. If good faith will not restrain members of Congress from overly politicizing their impeachment power, it is hard to see why imposing a formal limit on the ability to impeach based on motive would be effective. If, on the other hand, we expect members of Congress to act in good faith, then there do not seem to be particularly distinct problems in allowing the President to be impeached based on a corrupt motive.

At bottom, the strongest limits on the potential for the politicization of impeachment are likely to be structural and political. Structurally, there are high hurdles to removing a President via impeachment. The Constitution requires a two-thirds majority of the Senate to remove the President from office,209 which, with anything approaching current levels of partisanship, means that it will be nearly impossible to remove a President on purely politicized charges of improper motive. There remains the worry that permitting impeachment based on motive will increase the number of impeachments, even if not removals, which will distract members of Congress as well as the Presidents from the task of governing. However, the cost of impeachment is unlikely to be static. If politicized impeachments really do become commonplace, the public is unlikely to impose the same political cost for impeachment. In other words, as the frequency of impeachments increases, their cost is likely to decrease.

Although there is obvious danger in Congress politicizing its impeachment power, its ability to do so based on motive does not seem to raise unique concerns. It strikes me as perfectly appropriate to impeach a President based on improper motive, assuming the impeachable act otherwise qualifies as a “high crime and misdemeanor.” And, in keeping with the normative analysis above, impeachment based on motive might be most justified in areas of discretion premised on nonverifiable conditions—such as in President Trump’s first impeachment—and least justified when dealing with duties premised on verifiable conditions.

But impeachments are rare and operate ex post. Congress could also do more ex ante to help ensure the President abides by his duty. First, Congress could impose a public explanation requirement along the lines discussed above. It already does so in some instances,210 and it could pass a statute specifying that, before acting pursuant to a statutory delegation, the President must offer a public-interested reason for doing so, perhaps absent an emergency.211 In keeping with the normative analysis above, the statute could impose a more robust requirement in areas of discretionary authority where the conditions are nonverifiable.

Second, to the extent that Congress is concerned about the President acting for self-interested, rather than public-interested reasons, it could change how it delegates power. Rather than providing broad delegations of discretionary authority premised on nonverifiable conditions, it could delegate power more frequently in the form of duties premised on verifiable conditions, rendering the President’s motive less relevant and objective assessment of legality easier. This would require Congress to know more clearly when it wants power exercised and be willing to take more responsibility for such exercises of power—both of which make this suggestion unlikely to be taken up in any robust sense. But changing the delegations themselves is another tool Congress could use to limit the President’s self-interested behavior at least in discrete circumstances.

As above, none of these approaches will fully prevent the President from acting based on impermissible motive. But there is no perfect enforcement of the limits on presidential power, and that fact ought not prevent us from improving the current regime.


Courts too have a potentially important role in enforcing the President’s motive requirement. Below I examine two approaches to judicially enforce the President’s motive requirement: nondeferential review and a public explanation requirement. I ultimately conclude that a public explanation requirement is the more attractive option, because it is better suited to enforce the President’s motive requirement where it is most justified—in areas of discretion premised on nonverifiable conditions. That said, in my view, judicial enforcement of presidential motive ought to be extremely rare (as it has been historically).212 Judges are not immune to conscious or subconscious political bias, and permitting courts to review the President’s motive means that they might do so erroneously. Given the potential for error and the relatively high transaction costs involved in motive inquiries, my view is that courts ought to overrule presidential conduct based on the President’s motive only in the most egregious circumstances.

Ultimately, the optimal scope or form of judicial review is likely to depend on one’s empirical assumptions about how likely courts are to err—in good faith or bad—in assessing the President’s motive.213 Although the optimal level of judicial review is contestable, the approaches discussed below can be implemented in a way sensitive to such concerns.214

Perhaps the most familiar way to address findings of impermissible motive would be to apply a form of nondeferential or heightened judicial scrutiny. Under this approach, if the President were found to have acted based on an impermissible motive, the action in question would be subjected to nondeferential judicial scrutiny.215 This approach seems workable for exercises of power premised on verifiable conditions. In such areas, courts might be able to competently assess whether the condition triggering authority has been met nondeferentially. For example, if the President has sanctioned an individual for “knowingly . . . contribut[ing]” to a chemical weapons program, but a court determines the President did so based on an impermissible personally regarding motive, a court could assess whether the evidence shows the person, in fact, so contributed without giving deference to the President.216 Similarly, if the court is tasked with determining whether a border wall is a “military construction project[],” it could do so nondeferentially.217

However, this approach is unlikely to work in areas where motive is most relevant—discretionary authorities premised on nonverifiable conditions. In such cases, where the President’s power is premised on finding conduct is, for example, in the national interest or “national security interest of the United States,” it is not clear how a court could possibly make this determination nondeferentially. Take the Travel Ban as an example. In addition to the Equal Protection challenge in the case, the plaintiffs challenged whether the President had statutory authority to issue the directive, which required him to find that the entry of certain classes of noncitizens would be “detrimental to the interests of the United States.”218 It is not clear how a court could determine if that condition had been satisfied in an objective way without deference to some other actor. This is ultimately a policy judgment that courts cannot make in any “objective” sense of the word.219 In short, nondeferential review strikes me as potentially workable for duties based on verifiable conditions but impractical for authorities based on nonverifiable conditions.220

Because heightened review does not strike me as workable for exercises of power where motive inquiry is most justified, I propose an alternative approach drawing on recent developments in administrative law. This approach would impose a judicial requirement that the President give a public-interested reason before taking major action. The exercise of power would be subject to a strong presumption of motivational regularity,221 but, if that presumption were overcome, the court could then examine whether the public-interested justification was pretextual and, if so, remand for the President to give her real reason for acting.222 This would have the benefit of helping enforce the motive requirement with the additional benefit of improving public accountability.223 

In the Supreme Court’s recent cases overturning the Trump Administration’s attempt to rescind DACA and put a citizenship question on the census, the Court invalidated the administration’s policies based on the reasons for—or motivations behind—them.224 In the DACA case, the Court overruled the Trump Administration’s attempt to claim that its reason for action was that it had no discretion other than to rescind the entire policy.225 Meanwhile, in the census case, the Court found that the reason given by the Secretary of Commerce for including a citizenship question on the census was pretextual—it was not actually the reason the policy was enacted—and therefore could not be sustained.226 As Benjamin Eidelson explains, in both cases, the Court appeared to permit the administration to undertake its preferred policy, but only if it publicly acknowledged its actual reasons for doing so.227 As Eidelson persuasively argues, requiring the actual reasons for conduct to be publicly announced contributes to important political accountability values: “By insisting that major policy decisions be justified in a manner that does not obscure the decisionmakers’ reasons, courts can help citizens to judge their leaders accurately and to modify their own attitudes toward those leaders accordingly.”228

Although these cases involved action by administrative agencies, the same reasoning can easily apply to the President.229 If the Court requires the President to give reasons before she undertakes action and subjects such reasons to pretext review, this could enable public accountability by informing the public of the actual reasons that drive presidential conduct. This would also help enforce the motive requirement. If the President must publicly give formal reasons for acting, she will surely give public-interested reasons for doing so. If those reasons are not genuine, then they could, in appropriate circumstances, be reviewed for pretext. This could help serve as a check on, at least egregious, instances of self-interested behavior.

That said, such a form of review has the potential to impose serious costs on the President’s ability to exercise power.230 Litigation would ensue along with invasive discovery requests.231 Moreover, there is a significant risk that courts will erroneously find motives pretextual. Given the potential import of these cases, the difficulty in disentangling political and public-interested motives, and the strong pull toward politically motivated reasoning even for judges, the potential for danger here cannot easily be dismissed. A sufficiently high pleading standard might address some of these concerns, but, at bottom, such review should be limited to only the most egregious cases, for example, where the improper motivation is essentially impossible to ignore.232

Others might argue that such limited review is precisely the problem and that such an explanation requirement would be too permissive. On this view, the President will often be able to publish boiler-plate public-interested reasons to satisfy judicial review and thus only the most obvious forms of pretext would ever be overruled. But even that would be progress. Although an explanation requirement can be gamed, it will make it harder to engage in impermissibly motivated conduct.233 It would push the President toward public-interested behavior and away from private-interested behavior. Not perfectly, but somewhat. And that is no small thing.

Indeed, even if presidential action is only rendered invalid in extreme circumstances, there are important benefits to the potential of judicial process itself. Even if courts are likely to ultimately uphold the President’s action, the specter of judicial review can help ensure better governmental conduct by the President and those who work for him in the executive branch. Meanwhile, even when the President’s conduct is ultimately upheld, judicial process can have important benefits through creating public information and disciplining various actors within the executive branch.234 Thus whatever form it takes, having some form of judicial review might be desirable, even if the President’s conduct ought to be overturned only in extreme circumstances.

We ought to be cautious about judicial review of the President’s motive given the potential for judicial error—especially in the highly politicized areas of presidential power. It thus seems to me a reasonable position to avoid judicial review of motive entirely. Nonetheless, it strikes me as justifiable that courts play some role in checking the President in egregious cases.235 This form of review provides a reasonable approach to achieve the benefits of the prospect of judicial review while limiting the potential dangers of politicized judicial decision-making. Such a form of review will not perfectly enforce the President’s obligations, but it would improve the current state of affairs. It would help push the President toward acting in the public interest, force her to give the public reasons for her actions, and better ensure that the government is acting on behalf of the people, rather than on behalf of the President. Given the importance of the President in our system of government, these are serious benefits.


It may seem naïve to expect the President to act motivated by the public interest, rather than her personal interest. But this is what the Constitution requires. It is also consonant with familiar conceptions of how our government is supposed to work—for us. But even if we accept that the President’s motive is limited by the Constitution, many hard questions remain. Should we really care about the President’s reasons for acting, rather than the effects her actions have on the world? If such motive inquiry is justifiable, which motives are permitted and which are not? How do we differentiate permissible public-regarding from impermissible private-regarding reasons in the context of political motivations? What do we do when the President has both permissible and impermissible motives? How can we realistically enforce these limits on the President’s motive? These are all complex questions. This Article has sought to address them in a rigorous and nuanced way. There is no doubt that issues of presidential motive will become highly salient again. With this analysis in mind, we all will be better equipped to address the inevitable questions and scandals to come.

  1. [1]. It is well-known that President Trump’s motive was a frequent topic of conversation during his administration and after. See, e.g., Benjamin Wittes, Gordon Sondland Accuses the President of Bribery, Lawfare (Nov. 20, 2019, 2:43 PM), [] (arguing President Trump’s “corrupt” political motives in crafting foreign policy with Ukraine impeachable); Peter Baker, Radical Break From Tradition: Trump Stages Part of His Convention from the White House, N.Y. Times (Aug. 26, 2020), []; Daniel Hemel, Why Trump’s Lawyer Is Dead Wrong on Obstruction of Justice, Just Security (Dec. 4, 2017), []; see also Robert Draper, ‘This Was Trump Pulling a Putin’, N.Y. Times Mag. (Apr. 11, 2022),
    /11/magazine/trump-putin-ukraine-fiona-hill.html [] (quoting Trump Administration National Security Adviser John Bolton as stating, “[President Trump] is a complete aberration in the American system. We’ve had good and bad presidents, competent and incompetent presidents. But none of them was as centered on their own interest, as opposed to the national interest, except Trump. And his concept of what the national interest was really changed from day to day and had a lot more to do with what his political fortunes were”); Maggie Haberman, Trump Proposed Launching Missiles into Mexico to ‘Destroy the Drug Labs,’ Esper Says, N.Y. Times (May 5, 2022), [] (quoting Trump Administration Secretary of Defense Esper as stating, “[President Trump] is an unprincipled person who, given his self-interest, should not be in the position of public service”). But allegations of impermissible motive span presidents of both parties and continue to this day. See, e.g., Susan Baer, Clinton’s Airstrike Motives Questioned Many Wonder if Attack Was Meant to Distract from Lewinsky Matter, Balt. Sun (Aug. 23, 1998, 12:00 AM),
    .html [] (“Pundits, politicians, the media and a sizable chunk of the public questioned whether the president was acting [by ordering missile strikes against alleged terrorist targets in Afghanistan and Sudan], at least in part, with hopes of shifting attention from his travails in the Monica Lewinsky matter.”); id. (“President Ronald Reagan’s invasion of Grenada in 1983 ... was met with similar skepticism by those who thought the president was trying to divert attention from the deadly bombing of the U.S. Marine barracks in Lebanon ... days earlier”); Matt Bowman, Biden’s Illegal Vaccine Mandate Is About Politics, Not Science, The Hill (Nov. 11, 2021, 3:00 PM), [] (“This is a political motive, not a scientific one.”); Josh Blackman, President Biden’s Choice to Decline Executive Privilege Before Trump v. Thompson, Reason: The Volokh Conspiracy (Dec. 10, 2021, 1:25 AM),
    ecutive-privilege-before-trump-v-thompson []; Julia Preston & John H. Cushman Jr., Obama to Permit Young Migrants to Remain in U.S., N.Y. Times (June 15, 2012),
    ts.html [] (quoting Senator Charles E. Grassley of Iowa as stating, “It seems the president has put election-year politics above responsible policies”).

  2. [2]. The only recent impeachment controversy where motive was not centrally involved was President Trump’s second impeachment for his role in the attack on the capitol on January 6. Motive was central to President Trump’s first impeachment and to Richard Nixon’s potential impeachment. It also played a role in Bill Clinton’s impeachment for obstruction of justice.

  3. [3]. Most impeachment scholars believe motive can be relevant to whether conduct is impeachable. See, e.g., Charles L. Black, Jr. & Philip Bobbitt, Impeachment: A Handbook 25 (2d ed. 2018) (noting that “motive or intent” are “crucial” to impeachments for bribery); id. at 35 (noting that it would be impeachable if “a president were shown by convincing evidence to have used the federal tax system consistently and massively as a means of harassing and punishing his political opponents”); id. at 39 (motive to undermine law is inconsistent with “faithful[]” execution and likely impeachable but finding this a “gray area” (emphasis omitted)); Cass R. Sunstein, Impeachment: A Citizen’s Guide 121 (2017) (noting that the president’s use of IRS to investigate an enemy would be impermissible if “his desire to punish a political opponent is really what motivates him to exercise what he sees as his authority”); id. at 119 (concluding clearly impeachable if President revealed classified information to an adversary “with the clear intention of strengthening it and of weakening his own country”); Michael J. Gerhardt, The Federal Impeachment Process: A Constitutional and Historical Analysis 219 (3d ed. 2019) (“[A]n impeachable offense requires bad intent and a bad act.”). Berger is the outlier in that he concludes that motive is not generally relevant to whether conduct is impeachable. Raoul Berger, Impeachment: The Constitutional Problems 294–95 (1973) (“Given lawful action, ‘motive’ is of no moment. When Congress ... legislates ‘within the reach of its lawful power,’ the courts will not inquire into the motives which induced it to enact the statute. That principle is equally applicable to executive action within statutory or constitutional limits.” (quoting Magnano Co. v. Hamilton, 292 U.S. 40, 44 (1934) (citation omitted)).

  4. [4]. See Shalini Bhargava Ray, Plenary Power and Animus in Immigration Law, 80 Ohio St. L.J. 13, 16–20 (2019); Katherine Shaw, Speech, Intent, and the President, 104 Cornell L. Rev. 1337, 1338–45 (2019); see also Aziz Z. Huq, Article II and Antidiscrimination Norms, 118 Mich. L. Rev. 47, 54–56 (2019).

  5. [5]. See Daniel J. Hemel & Eric A. Posner, Presidential Obstruction of Justice, 106 Calif. L. Rev. 1277, 1278–82 (2018).

  6. [6]. See supra note 4 (collecting sources). Indeed, the analysis of the Equal Protection claim in Trump v. Hawaii would not have been necessary if this were not the case. See Trump v. Hawaii, 138 S. Ct. 2392, 2414–15 (2018).

  7. [7]. As noted above, such allegations of impermissible presidential motive have a long history spanning Presidential administrations of both parties. See supra note 1 (collecting examples).

  8. [8]. Ray and Shaw have done excellent work on how to judicially assess claims of presidential animus, but do not focus on motives that do not implicate individual rights. See supra note 4. Renan’s important work on “The President’s Two Bodies,” discusses presidential motive at points, but is fundamentally concerned with identifying—and incisively articulating—the tension between the President’s personal and institutional interests, rather than in identifying whether and how Article II limits the President’s motive, how to define permissible and impermissible motives, whether such motive limit is normatively defensible, or how to go about enforcing it. See Daphna Renan, The President’s Two Bodies, 120 Colum. L. Rev. 1119, 1997–201 (2020). Kent, Leib, and Shugerman’s originalist work on “Faithful Execution,” which I discuss and seek to build on in Part I, touches on presidential motive, but brackets most of the questions addressed here. See, e.g., Andrew Kent, Ethan J. Leib & Jed Handelsman Shugerman, Faithful Execution and Article II, 132 Harv. L. Rev. 2111, 2120 (2019) (“The correct method of interpreting and applying the Constitution in the present day is endlessly contested, because it is unclear how to evaluate a President’s subjective motives and what to do about mixed motive cases.”). Perhaps the closest work on our topic is Micah Schwartzman’s terrific article focusing on whether looking to executive branch officials’ motive is morally justifiable. See Micah Schwartzman, Official Intentions and Political Legitimacy: The Case of the Travel Ban, in Nomos LXI: Political Legitimacy 201, 206 (Jack Knight & Melissa Schwartzberg eds., 2019) (“My question is this: [C]an we give a moral justification for rejecting the legitimacy of an official action on the basis of the intention or motivation behind it?”). Schwartzman’s article does not focus on legal, rather than moral, permissibility or how motive inquiry might apply to the peculiar features of presidential power. That said, I am deeply indebted to Schwartzman’s work, which this Article seeks to build directly upon.

  9. [9]. For recent public law work on the topic, see, e.g., Richard H. Fallon, Jr., Constitutionally Forbidden Legislative Intent, 130 Harv. L. Rev. 523, 525–29 (2016); Benjamin Eidelson, Reasoned Explanation and Political Accountability in the Roberts Court, 130 Yale L.J. 1748, 1751–58 (2021). For recent work in philosophy on the topic, see, e.g., T.M. Scanlon, Moral Dimensions: Permissibility, Meaning, Blame 8–88 (2008); Deborah Hellman, When Is Discrimination Wrong? 136–68 (2008); S. Matthew Liao, Intentions and Moral Permissibility: The Case of Acting Permissibly with Bad Intentions, 31 L. & Phil. 703, 703–06 (2012); Judith Jarvis Thomson, Physician-Assisted Suicide: Two Moral Arguments, 109 Ethics 497, 517 (1999); Schwartzman, supra note 8, at 212.

  10. [10]. Indeed, this is an action that President Biden could take immediately, and, if undertaken, would likely serve to bind future Presidents as well. See infra Section IV.A.

  11. [11]. The classic article on the topic is Lawrence Gene Sager, Fair Measure: The Legal Status of Underenforced Constitutional Norms, 91 Harv. L. Rev. 1212, 1212–13 (1978). The lack of a clear coercive enforcement mechanism that ensures the President fully complies with constitutional legal requirements is a central feature of our system of constitutional law. See Jack Goldsmith & Daryl Levinson, Law for States: International Law, Constitutional Law, Public Law, 122 Harv. L. Rev. 1791, 1794 (2009) (“Constitutional law . . . shares with international law the absence of an enforcement authority capable of coercing powerful political actors to comply with unpopular decisions.”).

  12. [12]. Schwartzman, supra note 8, at 205.

  13. [13]. Id.; see also Andrew Verstein, The Jurisprudence of Mixed Motives, 127 Yale L.J. 1106, 1123 (2018) (“[M]otives are a reason for action which an actor takes to be guiding.” (emphasis omitted)).

  14. [14]. See Schwartzman, supra note 8, at 205 (using the terms interchangeably).

  15. [15]. I similarly am focused here on delegations of power to the President, by name, by the Constitution or statute, bracketing questions about the President’s ability to direct how agencies exercise power delegated directly to them. For classic work on this topic, see, e.g., Elena Kagan, Presidential Administration, 114 Harv. L. Rev. 2245, 2326–31 (2001); Kevin M. Stack, The President’s Statutory Powers to Administer the Laws, 106 Colum. L. Rev. 263, 270–76 (2006).

  16. [16]. Cf. Fallon, supra note 9, at 580 (“A question then can arise about how we know individual legislators’ private intentions or motivations. This is a genuine problem, but often nowhere nearly so severe as some suggest. [T]he criminal law recurrently requires courts to determine defendants’ intentions. In imputing intentions to people whom we know, we often rely on a mix of contextual factors, biographical information, and explicit statements. We can do the same with legislators.” (footnotes omitted)); see also Schwartzman, supra note 8, at 205.

  17. [17]. Schwartzman, supra note 8, at 205 (“[T]here is nothing epistemically distinctive about trying to discern the intentions of a particular public official for constitutional purposes.”).

  18. [18]. See generally Goldsmith & Levinson, supra note 11 (discussing the lack of central enforcement mechanisms for constitutional law).

  19. [19]. See U.S. Const. art. II, §§ 1, 3; Inaugurations and the White House, White House Hist. Ass’n,
    -sheet [].

  20. [20]. Some might question whether it is meaningful to speak of legal obligations that are not subject to robust external judicial or even congressional enforcement. I will not delve into this debate here but accept for purposes of this article that one can have meaningful legal obligations, even if they are not robustly externally enforced. Cf. Goldsmith & Levinson, supra note 11, at 1823; Hans A. Linde, Due Process of Lawmaking, 55 Neb. L. Rev. 197, 244 (1976) (“It is not mere theory to distinguish between constitutional law and judicial review.”); Trump v. Hawaii, 138 S. Ct. 2392, 2424 (2018) (Kennedy, J., concurring) (“There are numerous instances in which the ... actions of Government officials are not subject to judicial scrutiny or intervention. That does not mean those officials are free to disregard the Constitution ... .”).

  21. [21]. See, e.g., Tamsin Shaw, William Barr: The Carl Schmitt of Our Time, N.Y. Rev. (Jan. 15, 2020), [].

  22. [22]. Memorandum from Bill Barr to Rod Rosenstein, Deputy Att’y Gen., U.S. Dep’t of Just., & Steve Engel, Assistant Att’y Gen., U.S. Dep’t of Just. 1 (June 8, 2018) [hereinafter Barr Memo].

  23. [23]. For example, Barr states that “[t]he President’s discretion in these areas has long been considered ‘absolute,’ and his decisions exercising this discretion are presumed to be regular and are generally deemed non-reviewable.” Id. at 9. As I read it, Barr’s characterization of the President’s discretion as “absolute” does not seem to be a claim that he can act for any reasons at all—rather it is a claim that such actions are not generally reviewable by a Court. See id. at 14. Similarly, Barr’s claim that “[t]he Constitution itself places no limit on the President’s authority to act on matters which concern him or his own conduct,” id. at 10, seems to be a claim that there is no categorical limit on such conduct so long as it is properly motivated, not that there is no limitation on the reasons for engaging in such conduct. See id. at 12 (“Implicit in the Constitution’s grant of authority over such cases . . . is the recognition that Presidents have the capacity to decide such matters based on the public’s long-term interest.” (emphasis added)).

  24. [24]. See, e.g., Renan, supra note 8, at 1158 (citing Barr memorandum for position that “[t]here is no legally forbidden motive when the incumbent exercises a core power of Article II”). I take Renan’s claim to be about whether the President’s motive can be judicially reviewed. Barr’s position is that it cannot be and, in that sense, there is no “legally forbidden motive” that a court will enforce. See id. In this sense, I am in full agreement with Renan’s characterization of Barr’s memorandum.

  25. [25]. Barr Memo, supra note 22, at 11.

  26. [26]. Id.

  27. [27]. Id. at 9–12.

  28. [28]. Id. at 12 (emphasis added).

  29. [29]. See id. at 9.

  30. [30]. See id. at 12.

  31. [31]. See, e.g., id. at 10–11 (“The Framers’ idea was that, by placing all discretionary law enforcement authority in the hands of a single ‘Chief Magistrate’ elected by all the People, and by making him politically accountable for all exercises of that discretion by himself or his agents, they were providing the best way of ensuring the ‘faithful exercise’ of these powers.” (emphasis added)).

  32. [32]. Id. at 10; see also id. at 11 (“Every four years the people as a whole make a solemn national decision as to the person whom they trust to make these prudential judgments. In the interim, the people’s representatives stand watch and have the tools to oversee, discipline, and, if they deem appropriate, remove the President from office.”); id. at 12 (“The fact that President is answerable for any abuses of discretion and is ultimately subject to the judgment of Congress through the impeachment process means that the President is not the judge in his own cause.”). Barr also objects to the chilling effect he thinks criminal liability would have on the President’s ability to fulfil his responsibilities. See id. at 10. But, again, Barr’s claim is not that Article II imposes no limit on the President’s motive, but rather that criminalizing presidential conduct is an inappropriate way to enforce those limits. See id. at 14–15, 19.

  33. [33]. See, e.g., Hemel & Posner, supra note 5, at 1302 (“[N]either Taney in Jewels of the Princess Orange nor Scalia in Morrison argued that the president’s prosecutorial discretion grants him the power to pursue or drop a case for any reason whatsoever. Other advocates of the unitary executive theory do not make that claim either ... .” (footnote omitted)); id. at 1302 n.138 (“[E]ven Justice Scalia appeared to accept the proposition that intervening in an investigation for partisan purposes would be improper.”). Even President Trump’s first impeachment defense—where his motive was central—primarily argued about the dangers of Congress enforcing a motive limit, rather than that no such limit exists. See 166 Cong. Rec. S319–20 (daily ed. Jan. 21, 2020)(Trial Memorandum of President Donald J. Trump) [hereinafter Trump Brief] (raising objections relating to dangers of Congress impeaching based on motive, but not claiming motive irrelevant to legality of conduct). I also take Joshua Blackman and Seth Barrett Tillman’s arguments against looking to motive in impeachment as grounded in arguments based on the dangers of Congress politicizing motive inquiries rather than claiming that motive is irrelevant to legality under Article II. See Josh Blackman & Seth Barrett Tillman, Defining a Theory of ‘Bribery’ for Impeachment, Lawfare (Dec. 6, 2019, 12:43 PM), [].

  34. [34]. Kent et al., supra note 8, at 2120; id. at 2192 (“[O]ur findings here at least suggest that the President . . . must pursue the public interest in good faith republican fashion rather than pursuing his self-interest ... .”); id. at 2141 (concluding faithful execution came to signify duty to act “impartially in the best interest of the public”).

  35. [35]. See id. at 2191 (finding that faithful execution “likely means that when the Executive acts or refrains from acting, he must be motivated by the right kinds of reasons”); see also Ethan J. Leib & Stephen R. Galoob, Fiduciary Political Theory: A Critique, 125 Yale L.J. 1820, 1836 (2016) (“[Fiduciary norms require that] it is impossible to act loyally by accident. If an agent’s behavior and/or deliberation happen to match the pattern specified by a fiduciary norm, but the interests or ends of the principal do not influence the agent’s practical deliberation in the right way, then the agent has not complied with the fiduciary norm.”).

  36. [36]. See Kent et al., supra note 8, at 2119.

  37. [37]. See, e.g., Samuel L. Bray & Paul B. Miller, Against Fiduciary Constitutionalism, 106 Va. L. Rev. 1479, 1482 (2020). For their part, Kent et al. are careful to explain that they “do not claim that the drafters at Philadelphia took ready-made fiduciary law off the shelf and wrote it into Article II. But we do assert that the best historical understanding of the meaning of the Faithful Execution Clauses is that they impose duties that we today ... would call fiduciary.” Kent et al., supra note 8, at 2119.

  38. [38]. See Richard Primus, The Elephant Problem, 17 Geo. J.L. & Pub. Pol’y 373, 398 (2019).

  39. [39]. See id. (“Both the Constitution and a typical power of attorney are legal documents by which some authoritative actor empowers some other actor to exercise power, and in both scenarios the power is supposed to be exercised for the benefit of the power-conferring actor rather than that of the power-exercising actor.” (emphasis added)); id. at 404–05 (“[A]nalogizing the Constitution to a power of attorney can be salutary if it reminds people who exercise governmental power that they are required to act for the sake of something other than their own benefit.” (emphasis added)).

  40. [40]. Seth Davis provides a forceful critique of incorporating the judicial remedial structure of fiduciary law to public law but does not engage with whether the President might be required to be motivated by the public interest as a matter of Article II. See Seth Davis, The False Promise of Fiduciary Government, 89 Notre Dame L. Rev. 1145, 1146 (2014). Bray and Miller also focus primarily on problems with applying fiduciary judicial review principles to public officials under the Constitution. See, e.g., Bray & Miller, supra note 37, at 1512–17, 1530. Bray and Miller, admittedly, sound skeptical about incorporating a “good faith” requirement as a criterion of legal permissibility, but their focus is on showing that Article II’s “faithful execution” language cannot be filled in by reference to fiduciary law, rather than providing any affirmative argument for what “faithful execution” does entail. Id. at 1507–09.

  41. [41]. See, e.g., Kent et al., supra note 8, at 2190 (“Ultimately, our effort here is not to develop clear rules of constitutional law. But the finding of a fiduciary duty of loyalty in the Faithful Execution Clauses is an important development and must be considered along with other modalities of constitutional interpretation in finding answers to pressing modern problems.”).

  42. [42]. See, e.g., In re Debs, 158 U.S. 564, 584 (1895) (“Every government, [is] intrusted by the very terms of its being with powers and duties to be exercised and discharged for the general welfare.”).

  43. [43]. See, e.g., William N. Eskridge, Jr. & Victoria F. Nourse, Textual Gerrymandering: The Eclipse of Republican Government in an Era of Statutory Populism, 96 N.Y.U. L. Rev. 1718, 1760 (2021) (“The Constitution [and the Founding Era] established our government as a republic of representatives, not as a populist democracy. As Madison expressed it in Federalist No. 10, the goal of a representative, deliberative democracy is ‘to refine and enlarge the public views, by passing them through the medium of a chosen body of citizens, whose wisdom may best discern the true interest of their country, and whose patriotism and love of justice, will be least likely to sacrifice it to temporary or partial considerations.’”).

  44. [44]. Id. at 1789 (“[W]e have a republican government, one which filters democracy through representation.”); Evan D. Bernick, Faithful Execution: Where Administrative Law Meets the Constitution, 108 Geo L.J. 1, 22 (2019) (“Representative governments rest upon the premise that government power ought to be exercised in order to achieve ends that are valuable to members of the public.”); see also Alan Rozenshtein, The Virtuous Executive 12 (Feb. 25, 2022) (unpublished manuscript) (on file with author) (“Th[e] constitutional requirement [of faithful execution] is closely related to a basic requirement of political morality in a representative state: that leaders act to further the common good.”).

  45. [45]. Joseph Story, Commentaries on the Constitution of the United States § 1524 (1833). Story concludes his treatises’ section on the “rights, powers, and duties of the executive department,” with this admonition:

    All, that seems desirable in order to gratify the hopes, secure the reverence, and sustain the dignity of the nation, is, that it should always be occupied by a man of elevated talents, of ripe virtues, of incorruptible integrity, and of tried patriotism; one, who shall forget his own interests, and remember, that he represents not a party, but the whole nation; one, whose fame may be rested with posterity, not upon the false eulogies of favorites, but upon the solid merit of having preserved the glory, and enhanced the prosperity of the country.

    Id. § 1566.

  46. [46]. For a useful overview of the debates in political theory, see Suzanne Dovi, Political Representation, Stan. Encyc. of Phil. (Aug. 29, 2018),
    ical-representation [].

  47. [47]. See, e.g., Margaret H. Lemos, Three Models of Adjudicative Representation, 165 U. Pa. L. Rev. 1743, 1743–44 (2017) (“The standard account of political representation emphasizes three features that serve to legitimize and democratize representation: Representatives must be authorized to act on the people’s behalf; there must be some means by which the people can hold their representatives accountable for their actions; and the representatives must in fact endeavor to advance the people’s interests.”); Dovi, supra note 46 (“Political representation occurs when political actors speak, advocate, symbolize, and act on the behalf of others in the political arena.”); Hanna F. Pitkin, The Concept of Representation 209 (1967) (“[R]epresenting
    ... means acting in the interest of the represented, in a manner responsive to them.”); id. at 208 (“If ‘to represent’ as an activity is to have a substantive meaning, it must be ‘to act in the interest of’ or ‘to act according to the wishes of,’ or some such phrase.”); see also Andrew Rehfeld, Representation Rethought: On Trustees, Delegates, and Gyroscopes in the Study of Political Representation and Democracy, 103 Am. Pol. Sci. Rev. 214, 215 (2009) [hereinafter Rehfeld, Representation Rethought] (noting that core feature of representation under either conventional model is “to aim at” good of constituency); Andrew Rehfield, Towards a General Theory of Political Representation, 68 J. Pol. 1, 3 (2006) [hereinafter Rehfeld, Towards a General Theory of Political Representation] (“Under the standard account, a political representative ... has substantive obligations to act on behalf of another’s interests as ipso facto what it means to be a political representative.”).

  48. [48]. See, e.g., Pitkin, supra note 47, at 213 (emphasis added); Lemos, supra note 47, at 966.

  49. [49]. Pitkin, supra note 47, at 164. This ideal of representation was not lost on the Founders. Cf. Cass R. Sunstein, Interest Groups in American Public Law, 38 Stan. L. Rev. 29, 41 (1985) (“Madison regarded [representation] as an opportunity for achieving governance by officials devoted to a public good distinct from the struggle of private interests.”); The Federalist No. 57 (James Madison) (“The aim of every political constitution is, or ought to be, first to obtain for rulers men who possess most wisdom to discern, and most virtue to pursue, the common good of the society; and in the next place, to take the most effectual precautions for keeping them virtuous whilst they continue to hold their public trust.”); Daryl J. Levinson, Parchment and Politics: The Positive Puzzle of Constitutional Commitment, 124 Harv. L. Rev. 657, 668–69 (2011) (“Madison hoped that ... indirect election of senators and the President would select for representatives who would ‘possess most wisdom to discern, and most virtue to pursue, the common good of the society.’”); Fallon, supra note 9, at 559 & n.168 (“Today, almost no one questions that our Constitution includes deliberative norms that require legislators to eschew reliance on forbidden considerations when acting in their official capacities.” “A [] minimal version might hold that the legislators should not act on the basis of ‘naked preferences’ unsupported by publicly spirited reasons.”).

  50. [50]. Representation is typically thought to be present in the legislature, rather than the executive, but “there can be little doubt that the President fulfills ... representative functions.” Kathy B. Smith, The Representative Role of the President, 11 Pres. Stud. Q. 203, 206 (1981); see also Gary L. Gregg II, The Presidential Republic: Executive Representation and Deliberative Democracy 65 (1997) (“Despite the differences with the legislative bodies, ... the president’s origin, powers, and the nature of his responsibilities make him a central representative in the American political system... . Th[e] indirect mode of election, like that for the Senate, does not make the presidency less of a republican institution or its holder less of a political representative.”); Karen S. Hoffman, Misunderstanding Executive Representation: Causes and Consequences, 28 Cong. & the Presidency 185, 202 (2001) (“[E]xecutive representation exists on some unidentified plane as a phenomenon that is accepted and understood, but never defined. It is somehow different than congressional representation, but exactly what the differences are is not known, because executive representation is for the most part an unexamined assumption that most people accept at face value.”); Story, supra note 45, at § 1566 (“All, that seems desirable in order to gratify the hopes, secure the reverence, and sustain the dignity of the nation, is, that [the executive department] should always be occupied by a man of elevated talents, of ripe virtues, of incorruptible integrity, and of tried patriotism; one, who shall forget his own interests, and remember, that he represents not a party, but the whole nation; one, whose fame may be rested with posterity, not upon the false eulogies of favourites, but upon the solid merit of having preserved the glory, and enhanced the prosperity of the country.” (emphasis added)). For a discussion of the history of contestation regarding the President’s role as a representative, see generally Jeremy D. Bailey, The Idea of Presidential Representation: An Intellectual and Political history (2019).

  51. [51]. See U.S. Const. art. II.

  52. [52]. It is not fully settled whether reference to the “Laws” in the take care clause includes the Constitution. See Goldsmith & Levinson, supra note 11, at 1836 (noting debate). Regardless, constitutional powers are clearly part of executing the “Office” of the Presidency, and so the President would be required by her oath to faithfully execute constitutional powers. See, e.g., Andrew Kent, Can Congress Do Anything About Trump’s Abuse of the Pardon Power?, Lawfare (July 24, 2020, 11:36 AM),
    -pardon-power [] (stating that even if the pardon power falls outside of take care clause’s use of “Laws,” “the full office of the presidency must be faithfully executed, as must the Constitution” regardless of whether it falls under Take Care obligation).

  53. [53]. Kent, et al., supra note 8, at 2132; Shalev Roisman, Presidential Factfinding, 72 Vand. L. Rev. 825, 856 (2019) (noting modern dictionaries are consistent with this meaning).

  54. [54]. See, e.g., Gregg, supra note 50, at 18 (“In none of the[] three types of presidential action does the officer act as a private citizen. All that he does is done in the name of the people or in the name of their fundamental law. In this way his doings transcend his person and he acts as a representative of things larger than himself. To have it otherwise would mean that we have empowered at the center of our government a force alien to the genius of our constitutional democracy. In his various capacities to act officially, the American president provides representation to a diverse people under one government and one constitutional tradition.”).

  55. [55]. See, e.g., Pitkin, supra note 47, at 118–19 (“When we act for someone else we may not act on impulse; we may not risk what others have at stake ‘just because we happened to feel like it.’ We are expected to act as if we would eventually have to account for our actions. Thus we ought to have reasons for what we do, and be prepared to justify our actions to those we act for, even if this accounting or justification never actually takes place.”); Rehfeld, Towards a General Theory of Political Representation, supra note 47, at 3 n.8 (“Another’s interests may be identical to the representative’s own interests, but the obligation to represent the former accrues only by virtue of their interests being someone else’s. A more complex case arises when the representative is part of the group whose interests she is representing. This is not terribly difficult to understand as a case in which the representative qua representative is not representing her own interests but rather the interests of the group of which she, as a citizen, happens to be a member.”).

  56. [56]. One could argue that faithful execution does not require the President to be motivated by pursuing the public interest, but rather that it only requires belief that conduct is in the public interest. In this way, the President would aim at the public interest, so long as she believes her conduct is in the public interest, even if it is not motivated by such public interest. I cannot dismiss this suggestion out of hand but have not seen any full-throated argument to this effect in the presidential power realm. Cf, e.g., Schwartzman, supra note 8, at 206 (discussing this distinction in other context); see also Douglas Husak, The Philosophy of Criminal Law: Selected Essays 65 (2010) (“It is hard to understand why mere belief should have a bearing on liability unless it influences conduct.”). It strikes me that “faithful” execution likely requires both belief that conduct is in the national interest and that the President be motivated by this view.

  57. [57]. For a terrific account noting the core tension between the President as a person and as an institution as a constitutive feature of our understanding of presidential power, see Renan, supra note 8.

  58. [58]. David E. Pozen, Constitutional Bad Faith, 129 Harv. L. Rev. 885, 914 (2016) (quoting The Federalist No. 51 (James Madison)).

  59. [59]. See id. (“Rather than rely on the republican virtue of officeholders, the Madisonian model seeks to ‘supply[], by opposite and rival interests, the defect of better motives.’” (alteration in original) (quoting The Federalist No. 51 (James Madison))).

  60. [60]. See generally, e.g., The Federalist No. 10 (James Madison) (“It is in vain to say that enlightened statesmen will be able to adjust these clashing interests, and render them all subservient to the public good. Enlightened statesmen will not always be at the helm. Nor, in many cases, can such an adjustment be made at all without taking into view indirect and remote considerations, which will rarely prevail over the immediate interest which one party may find in disregarding the rights of another or the good of the whole.”).

  61. [61]. This can be seen in Madison’s famous justification for a large republic. See id.

    The two great points of difference between a democracy and a republic are: first, the delegation of the government, in the latter, to a small number of citizens elected by the rest; secondly, the greater number of citizens, and greater sphere of country, over which the latter may be extended.

    The effect of the first difference is, on the one hand, to refine and enlarge the public views, by passing them through the medium of a chosen body of citizens, whose wisdom may best discern the true interest of their country, and whose patriotism and love of justice will be least likely to sacrifice it to temporary or partial considerations. . . .

    In the next place, . . . it will be more difficult for unworthy candidates to practice with success the vicious arts by which elections are too often carried; and the suffrages of the people being more free, will be more likely to cent[er] in men who possess the most attractive merit and the most diffusive and established characters.

    Id. (emphasis added).

  62. [62]. Neil S. Siegel, After the Trump Era: A Constitutional Role Morality for Presidents and Members of Congress, 107 Geo. L.J. 109, 133 (2018) (footnote omitted); see also Gordon S. Wood, Empire of Liberty: A History of the Early Republic, 1789–1815, at 9 (2009) (“In the 1780s James Madison had his doubts about this moral capacity of the people stretched to the limit, but even he admitted that ordinary people had to have sufficient ‘virtue and intelligence to select men of virtue and wisdom’ or ‘no theoretical checks, no form of government, can render us secure.’” (quoting Gordon S. Wood, The Creation of the American Republic, 1776
    –1787, at 544 (1998))); Ralph Ketcham, James Madison: A Biography 300 (1990) (“Madison insisted repeatedly that the future of republican government was hopeless without some confidence in human virtue ... .”).

  63. [63]. The Federalist No. 51 (James Madison).

  64. [64]. Daryl J. Levinson & Richard H. Pildes, Separation of Parties, Not Powers, 119 Harv. L. Rev. 2311, 2314 (2006) (emphasis added).

  65. [65]. Id. at 2318.

  66. [66]. See id. (“Democratic politics is unlikely to generate such officials.”); see also id. at 2323 (“[F]rom the outset of government under the Constitution, practical politics undermined the Madisonian vision of rivalrous branches pitted against one another in a competition for power... . In Madison’s terms, ‘the interests of the man’ have become quite disconnected from the interests of ‘the place.’”).

  67. [67]. See U.S. Const. art. II. The argument that Article II requires the President to be motivated by the public interest might not be sufficiently proven to some. This is not an area where we have definitive judicial precedent, and some might need more analysis of original public meaning or historical gloss to be fully convinced. For those who remain unconvinced, some of the normative discussion below might be relevant, as it might help “fill in” or “construct” the Constitution’s meaning. See Lawrence B. Solum, Originalism and Constitutional Construction, 82 Fordham L. Rev. 453, 455–58 (2013) (“[I]n [some] cases, the constitutional text does not provide determinate answers to constitutional questions... . We can call this domain of constitutional underdeterminacy ‘the construction zone.’”); id. at 472 (“[C]onstruction is essentially driven by normative concerns.”). See also generally Ronald Dworkin, Law’s Empire (1986); David A. Strauss, The Living Constitution (2010). However, the duty strikes me as sufficiently straightforward to be confident in its existence, and I will treat the duty as having been established for the remainder of the Article; see also Rozenshtein, supra note 44, at 14–16 (providing structural support for a related theory of constitutional presidential virtues).

  68. [68]. See supra note 9 (collecting sources). To be clear, the goal of this section is not to break new ground in the moral philosophy literature on the relevance of motive to moral permissibility. Instead, the goal is to take lessons from this literature and apply them to the peculiar context of presidential power. It turns out—or so I will argue—that lessons in the moral philosophy and public law literature on motive reveal that many of the features of the President’s power render looking to motive justifiable in a way that might not be true of other actors. It also identifies certain categories of presidential conduct where motive inquiry is more and less justified. In short, the question this section seeks to answer is not whether motive ought ever be relevant to permissibility in some fundamental moral sense, but rather whether, practically speaking, it is normatively defensible to make the President’s motive relevant to the legality of her conduct.

  69. [69]. For variations of this example, see Alison McIntyre, Doctrine of Double Effect, Stan. Encyc. of Phil. (Dec. 24, 2018), [
    /EB4T-8QPW]; see also Scanlon, supra note 9, at 28–32; Schwartzman, supra note 8, at 213–14.

  70. [70]. See supra note 9 (collecting sources).

  71. [71]. Scanlon, supra note 9, at 22–23; see also Schwartzman, supra note 8, at 214 (“Those who argue for the irrelevance of intentions draw a sharp distinction between the moral justification of an act and whether an agent is motivated by that justification. The permissibility of an act depends on whether there are sufficient reasons for allowing it. For a given act, such reasons may exist, even if an agent does not act on the basis of them and, indeed, even if an agent is unaware of them.”).

  72. [72]. See generally Fallon, supra note 9.

  73. [73]. Scanlon, supra note 9, at 38–40, 58–60, 90, 110–12.

  74. [74]. See, e.g., Fallon, supra note 9, at 529­–31, 557–69.

  75. [75]. See, e.g., id. at 531 (“[F]ew would judge it tolerable for courts to strike down a law prohibiting murder if historical examination revealed that most members of the legislature voted for it solely for the constitutionally forbidden purpose of enforcing one of God’s commandments.”).

  76. [76]. See, e.g., John Ferejohn, Power in Public Law: Some Reactions, 130 Harv. L. Rev. F. 9, 14 (2016) (“The distinctive feature of democratic agency in a liberal democracy—in which people are free to hold diverse values and preferences and make decisions by voting—is that the democratic principal (the ‘people’) is heterogeneous, and its members cannot generally agree on the objective the agent should pursue.” (footnote omitted)); id. at 22–23 (doubting “that the public good is a unitary objective that can be or ought to be pursued single-mindedly in a heterogeneous society. We are a diverse people who disagree deeply about many things. Different communities and interests have their own views of the good, and reconciling these with each other and with some conception of national interests is necessarily difficult and conflictual”); Davis, supra note 40, at 1150.

  77. [77]. See, e.g., Barack Obama, President of the United States, Address to the Nation on the Situation in Libya in U.S. Gov’t Publ’g Office (Mar. 28, 2011), at 2,
    tent/pkg/DCPD-201100206/pdf/DCPD-201100206.pdf [] (“[I]f we waited one more day, Benghazi ... could suffer a massacre that would have reverberated across the region and stained the conscience of the world. It was not in our national interest to let that happen.”); Authority to Use Military Force in Libya, 35 Op. O.L.C. 20, at 23 (2011) (“The President . . . identified several national interests supporting United States involvement in the planned operations ... . Left unchecked, we have every reason to believe that Qaddafi would commit atrocities against his people. Many thousands could die. A humanitarian crisis would ensue. The entire region could be destabilized, endangering many of our allies and partners. The calls of the Libyan people for help would go unanswered. The democratic values that we stand for would be overrun. Moreover, the words of the international community would be rendered hollow.”).

  78. [78]. See, e.g., Blackman & Tillman, supra note 33 (“[C]onsider the public motivation. Trump’s request would directly promote foreign and domestic policy interests: Ukraine would investigate possible corruption regarding an American citizen.”). Ethics organizations called for investigations of President Trump’s oldest son, Donald Trump Jr., on similar grounds after he allegedly unlawfully received a permit to hunt an endangered sheep after privately meeting with the Mongolian President. See, e.g., Jael Holzman, Agency Says It Will Review Trump Jr.’s Sheep Hunt in Mongolia, Roll Call (Dec. 18, 2019, 11:31 AM), [].

  79. [79]. Cf. Roisman, supra note 53, at 846–51 (distinguishing between findings of “fact” and “policy judgment”).

  80. [80]. I take Ely to be somewhat sympathetic to this view. See John Hart Ely, Legislative and Administrative Motivation in Constitutional Law, 79 Yale L.J. 1205, 1273 (1970); cf. Rozenshtein, supra note 44, at 13 (“Where a decision has high stakes and uncertain long-term outcomes, depends on unique factual circumstances ... and requires the exercise of individual judgment on the part of the president, no appeal to an ex ante rule ... may be possible. In these cases the best way to make a moral evaluation of the action may be with reference to the qualities of the person who undertook it.”).

  81. [81]. See, e.g., Frederick Schauer & Virginia J. Wise, Nonlegal Information and the Delegalization of Law, 29 J. Legal Stud. 495, 498 (2000) (“[A]lthough lawyers, particularly, are likely to be preoccupied with dusk when people ask them about the distinction between night and day, we do not believe that the existence of borderline cases undercuts the existence of a workable distinction ... .”).

  82. [82]. See Roisman, supra note 53, at 848–49 nn.102–04 (collecting numerous examples).

  83. [83]. Nonconsequentialists might also find this relevant. Even nonconsequentialist critics of motive inquiry have suggested that looking to motive can be justified when it is predictive of impermissible behavior. See Scanlon, supra note 9, at 12, 67. So, for our purposes, if we believe that a President who is personally interested is likely to evaluate the national interest in a skewed way—or simply to ignore it—we might think that the actor’s conduct is likely to be worse when it is personally motivated. The personal motive might thus be predictive of conduct that is not in the national interest and thus indirectly relevant to the permissibility of the conduct. See id.

  84. [84]. See 50 U.S.C. § 4613(a) (2018); Presidential Discretion to Delay Making Determinations Under the Chem. and Biological Weapons Control and Warfare Elimination Act of 1991, 19 Op. O.L.C. 306, 309 (1995) (discussing how factually contingent this power is).

  85. [85]. See Elena Kagan, Private Speech, Public Purpose: The Role of Governmental Motive in First Amendment Doctrine, 63 U. Chi. L. Rev. 413, 414 (1996).

  86. [86]. See id. at 509 (“Why do we wish to discover improper motive? Perhaps because we wish to discover adverse effects, but cannot do so directly; because we know that actions tainted with certain motives tend to have such consequences; because although a focus on motive will prove imprecise, we can think of no better way to gauge the effects of an action on the state of public discourse.”).

  87. [87]. See id.

  88. [88]. Schwartzman, supra note 8, at 216–17; see also infra note 83 (collecting sources on this point).

  89. [89]. See Scanlon, supra note 9, at 57; Liao, supra note 9, at 712–15 (discussing how this is a “must act” case).

  90. [90]. See Fallon, supra note 9, at 557.

  91. [91]. See Liao, supra note 9, at 722 (“One confounding factor is that these cases appear to be cases in which the agent ‘must act’.”); Schwartzman, supra note 8, at 216–17 (“These are cases of moral urgency in which an agent is usually forced to confront a matter of life or death. In such cases, there is often only one permissible response, which is to save lives. The cases also typically require agents to make a single decision and to do so immediately and under pressure, without time for reflection and deliberation. There are no chances to exercise discretion, to think about the reasons for an action, or to revisit it in the future.”). But see Hellman, supra note 9, at 166 (suggesting that intent is irrelevant, even in discretionary areas).

  92. [92]. Liao, supra note 9, at 716 (“[I]f the enemy is not required to push the button, then it seems perfectly cogent to require the enemy not to push the button if the enemy is going to push the button with the wrong intention/for the wrong reason. The intention principle straightforwardly explains how the enemy’s bad intention can change a ‘may act’ into a ‘may/must not act’ in this case.”).

  93. [93]. Id. at 715–16.

  94. [94]. Id. at 716.

  95. [95]. Id.

  96. [96]. See Schwartzman, supra note 8, at 216.

  97. [97]. Id. at 218.

  98. [98]. See, e.g., Authority to Use Military Force in Libya, supra note 77, at 20; April 2018 Airstrikes Against Syrian Chem.-Weapons Facilities, 42 Op. O.L.C. 1, 10 (2018).

  99. [99]. See Curtis Bradley & Jack Goldsmith, OLC’s Meaningless ‘National Interests’ Test for the Legality of Presidential Uses of Force, Lawfare (June 5, 2018, 3:13 PM), https://www.lawfareblog
    .com/olcs-meaningless-national-interests-test-legality-presidential-uses-force [

  100. [100]. April 2018 Airstrikes Against Syrian Chem.-Weapons Facilities, supra note 98, at 10 (“These interests ... grant the President a great deal of discretion.”).

  101. [101]. See, e.g., Roisman, supra note 53, at 847–48 (collecting examples).

  102. [102]. See, e.g., Adam B. Cox & Cristina M. Rodríguez, The President and Immigration Law 5 (2020) (noting how this is a core element of immigration enforcement today and common to many other areas of presidential power).

  103. [103]. See U.S. Const. art. II, §§ 1, 3.

  104. [104]. See 50 U.S.C. § 4613(a).

  105. [105]. Presidential Discretion to Delay Making Determinations Under the Chem. and Biological Weapons Control and Warfare Elimination Act of 1991, supra note 84, at 309.

  106. [106]. In practice, the President may well have enforcement discretion as to which duties to fulfill and which not to. This would justify continued inquiry into motive in such areas.

  107. [107]. This transactional, hypothetical-heavy method is consistent with much of the nonconsequentialist literature on the topic. See Schwartzman, supra note 8, at 212.

  108. [108]. See also generally Rozenshtein, supra note 44 (making analogous point by reference to virtue theory). To put this in Scanlon’s terms, although motive might be irrelevant to an individual action, it can be relevant to a “larger intention,” which changes the consequences of the individual action. Scanlon, supra note 9, at 41–44. We might think that, even if discrete conduct by a President might be in the national interest, if it is part of a “larger plan” to help the President’s self-interest, it should still be impermissible. Cf. id. at 42.

  109. [109]. See, e.g., Fallon, supra note 9, at 585 (“In some instances, we have interests in the ‘meanings’ of actions as expressions of a person’s values or attitudes. For example, the meaning of a gift will depend on whether the giver sought to express affection or to curry favor. Just as we have reason to care about the meanings of actions independently of whether they are morally permissible, we may have reasons—including reasons of constitutional stature—to care about the meanings of statutes as expressions of the values of some, even if not all, members of the legislature, who are . . . elected representatives of the political community that they serve.” (footnote omitted)); Kagan, supra note 85, at 510–11 (suggesting motive inquiry might be justifiable because “two actions having similar material outcomes may express different values and have different meanings. . . . An action acquires meaning in part through motive, and the meaning of an action in part defines it” (footnote omitted)). See also generally Elizabeth S. Anderson & Richard H. Pildes, Expressive Theories of Law: A General Restatement, 148 U. Pa. L. Rev. 1503 (2000) (providing broader account of expressive theories in law).

  110. [110]. Cf. Thomas Fuller & Derrick Bryson Taylor, Trump Reverses Decision to Reject California’s Request for Wildfire Relief, N.Y. Times (Jan. 25, 2021),
    /us/trump-california-wildfire-relief.html [] (quoting former senior administration official as stating “[h]e told us to stop giving money to people whose houses had burned down from a wildfire because he was so rageful that people in the state of California didn’t support him and that politically it wasn’t a base for him”). But see id. (noting how such denials are not historically unusual: “from 1974 to 2016 presidents denied requests for disaster relief an average of 2.9 times per year during nonelection years, and 2.1 times in a year with a presidential election”).

  111. [111]. Cf. Peter M. Shane, Madison’s Nightmare: How Executive Power Threatens American Democracy 181 (2009) (“Under the founders’ vision ... [t]he minority of voters who lose in an election contest [ought] not lose their expectation that their interests—their lives and well-being—will continue to be respected even by those officials they did not support in the election.”); Siegel, supra note 62, at 115–16 (arguing the president should represent whole country, including those who did not vote for her).

  112. [112]. Cf. Eidelson, supra note 9, at 1819–20 (“[T]he supposition that only an administration’s bottom-line policy choices matter politically—that the public is indifferent to the reasons for those choices—is implausible. [T]he meaning of any action necessarily depends on how the actor took different facts to bear on his or her choice... . Certainly politicians appear to care about how their reasons are understood: That is why they are constantly defending their important decisions—articulating the values they understood a choice to serve, the considerations they weighed, and the like—rather than just reciting the policies they adopted ... .” (emphasis omitted)).

  113. [113]. A final way in which looking to motive can be justified is if it is required of the President as a role obligation. See Scanlon, supra note 9, at 70; Hellman, supra note 9, at 164. This strikes me as compelling but does not really respond to the main question we are addressing here, which is whether such an obligation ought to be part of the role of being President.

  114. [114]. See, e.g., Victor Tadros, The Ends of Harm: The Moral Foundations of Criminal Law 155–56 (2011) (“Perhaps the most important objection [to looking to motive as relevant to permissibility] is [that a] person who is deliberating about what to do should not normally focus on what motivates her. She should focus outwardly, on the effects that her action would have on herself and on others, not inwardly on her attitudes.”); Fallon, supra note 9, at 565–66 (making same point).

  115. [115]. Scanlon, supra note 9, at 20.

  116. [116]. It is worth noting here that many of the reasons I have given for looking to motive would be accepted by Scanlon as “indirectly” relevant—such as motive as proxy, enforcement prioritization, and expressive effects—even if they are not “directly” relevant to permissibility in some fundamental sense. See id. at 12–13, 62–73; Schwartzman, supra note 8, at 211.

  117. [117]. Cf. Kagan, supra note 85, at 509 (making point regarding motive in First Amendment doctrine).

  118. [118]. For another example, if the President informed a foreign ally that foreign aid was contingent on arranging a position on the board of a major company for the President’s son, this would be personally motivated. The reason the President offers foreign aid is to advance a personal interest—the career or finances of her family member—and, therefore, Article II prohibits such an exercise of power. For more examples, see, e.g., Hemel & Posner, supra note 5, at 1313–14.

  119. [119]. Of course, there are other impermissible reasons—such as animus, see generally Ray, supra note 4 and accompanying text, and, on some accounts, a desire to undermine statutory law. See, e.g., Kent et al., supra note 8, at 2113; Zachary S. Price, Enforcement Discretion and Executive Duty, 67 Vand. L. Rev. 671, 746 (2014).

  120. [120]. See Hemel & Posner, supra note 5, at 1314–15.

  121. [121]. Cf. id. (“Let us assume that the president writes a memo clearly stating that his only reason for adopting the ‘soft on pot’ policy is to win votes—he thinks it is otherwise a bad policy. Imagine that he also observes that the policy would throw the opposing party into turmoil, destroying its electoral prospects for years to come. Isn’t his motive ‘narrowly partisan’? We think that the president’s motive is legitimate.”).

  122. [122]. The same goes if the goal was to serve partisan ends, rather than his own political fortunes. Although the party is distinct from the human occupying the office, acting to help one’s party is still not an action motivated by serving the public interest. Just as the President cannot use public power to help a family member, he cannot use it to help a political party (or corporation). The President can only act if his purpose is to help the public.

  123. [123]. See supra notes 47­–49 and accompanying text.

  124. [124]. See, e.g., Rehfeld, Representation Rethought, supra note 47, at 214 (“[T]he central normative problem of democracy is often restated in terms of the relationship between citizens and their representatives: how closely must a representative’s votes on legislation correspond to the preferences and will of his or her constituents?”).

  125. [125]. To be clear, these are not the only conceptions, but they are the classic ones. See generally Dovi, supra note 46 (summarizing prominent theories and challenges to them); Jane Mansbridge, Rethinking Representation, 97 Am. Pol. Sci. Rev. 515 (2003) (providing additional conceptions).

  126. [126]. See, e.g., Lemos, supra note 47, at 1744 (“Some commentators argue that representatives ought to serve as delegates for the people, channeling their constituents’ wishes into action. Others envision representatives as enlightened and largely independent trustees who follow their own autonomous judgment about how best to promote their constituents’ interests.”); Pitkin, supra note 47, at 145 (describing debate as asking “[s]hould (must) a representative do what his constituents want, and be bound by mandates or instructions from them; or should (must) he be free to act as seems best to him in pursuit of their welfare?”); Amy Gutmann & Dennis Thompson, The Theory of Legislative Ethics, in Representation and Responsibility: Exploring Legislative Ethics 167, 170 (Bruce Jennings & Daniel Callahan eds., 1985) (“We distinguish trustees from delegates according to the reasons for their decisions—whether they decide according to expressed preferences of the persons they represent or according to the representative’s judgment about the interests of the persons they represent.”).

  127. [127]. See Rehfeld, Representation Rethought, supra note 47, at 223 (distinguishing between representatives’ ideal “source” of judgment and “responsiveness” and concluding some delegate conceptions require the representative to look to constituents as source and be motivated by fear of electoral sanction in terms of responsiveness).

  128. [128]. See id. at 215 (noting some delegate conceptions require the representative to be motivated by fear of electoral sanction to be sufficiently “responsive”).

  129. [129]. See id. (discussing conception that representatives ought to be motivated by fear of electoral sanction). As Rehfeld points out, in principle, a trustee could also view themselves as responsive to electoral sanction, but the classic conceptions of trustees envision them as acting motivated by civic virtue, rather than electoral sanction. See id. (“‘[T]rustees’ are generally described as (1) looking out for the good of the whole (the nation’s interests), (2) based on their own judgment about that good (rather than the judgment of their constituents), and
    (3) less responsive to sanctioning (acting instead according to civic virtue), whereas ‘delegates’ are generally described as (1) looking out for the good of a part (the interests of their electoral constituents), (2) defined by a third party (their constituents’ rather than their own judgment), and (3) more responsive to sanctions (in particular, the hope of reelection).”).

  130. [130]. See, e.g., Donald J. McCrone & James H. Kuklinski, The Delegate Theory of Representation, 23 Am. J. Pol. Sci. 278, 280 (1979) (noting central condition for delegate theory is “[t]he representative must believe himself to be obliged to behave in accordance with constituency preferences, i.e., he must consider himself a delegate” (emphasis omitted)).

  131. [131]. Donald A. Gross, Representative Styles and Legislative Behavior, 31 W. Pol. Q. 359, 362 (1978) (noting that “[f]ew congressmen have the delegate style”); Roger H. Davidson, The Role of the Congressman 52 (1967) (finding plurality of congressman did not adopt “delegate” conception).

  132. [132]. See, e.g., Gregg, supra note 50, at 147 (“Presidents regularly claim to make decisions according to what their judgment tells them is in the nation’s best interest, whether those decision are popular or unpopular.”). But see id. at 148 (noting how Presidents, “at other times . . . celebrate[] their perfect harmony with the public”); see also Smith, supra note 50, at 212 (“The characteristics of the [Trustee and Delegate] models were readily apparent in past presidents’ writings. Nevertheless, the categories did not enable us to exclusively categorize presidents as either Trustee or Delegate.”).

  133. [133]. See, e.g., Terry M. Moe & Scott A. Wilson, Presidents and the Politics of Structure, 57 L. & Contemp. Probs. 1, 11 (1994) (“If there is a single driving force that motivates all presidents, it is not popularity with the constituency nor even governance per se. It is leadership.”); id. at 11 (“Reelection ... does not loom as large in [the President’s] calculations (and in the second term, of course, it is not a factor at all). [Presidents] are more fundamentally concerned with governance.”); William G. Howell & Terry M. Moe, Relic: How Our Constitution Undermines Effective Government and Why We Need a More Powerful Presidency 104 (2016) (“At their motivational core, then, presidents are strikingly different from legislators and approach public policy in a completely different way. [T]he premium they place on legacy makes them champions of the nation’s long-term interests in ways that Congress cannot and never will be.”); Matthew D. Adler, Judicial Restraint in the Administrative State: Beyond the Countermajoritarian Difficulty, 145 U. Pa. L. Rev. 759, 878–79 n.334 (1997) (noting presidentialist scholars suggest the President is not motivated solely by reelection, but rather stress “the President’s ‘autonomy,’ and drive for ‘leadership’” (citations omitted)). For a summary of studies that have attempted to measure how responsive the President’s actions are to “public opinion,” see James N. Druckman & Lawrence R. Jacobs, Presidential Responsiveness to Public Opinion, in The Oxford Handbook of the American Presidency 160 (George C. Edwards III & William G. Howell eds., 2009), which concludes that Presidents largely do track public opinion but that how much depends on “the electoral cycle, popularity, issue salience, and venue.” Id. at 170, 177. That said, as the authors note, the President, himself, can shape public opinion, id. at 174–76, and, ultimately, “[p]residential representation is much more complicated, multidimensional, and dynamic than investigations of whether the public’s policy preferences align with the president’s policies capture.” Id. at 177.

  134. [134]. See, e.g., Story, supra note 45, at § 1527 (“It is a duty of the president to ... disregard the importunities of friends; the hints or menaces of enemies; the bias of party, and the hope of popularity. The latter is sometimes the refuge of feeble-minded men; but its gleam is transient, if it is obtained by a dereliction of honest duty and sound discretion. Popular favour is best secured by carefully ascertaining, and strictly pursuing the true interests of the people. The president himself is elected on the supposition, that he is the most capable citizen to understand, and promote those interests; and in every appointment he ought to consider himself as executing a public trust of the same nature.”); Howell & Moe, supra note 133, at 14 (“The founders ... never intended ... for the president to be a tribune of the people. On the contrary, they shielded the office from popular pressures that might arise from ordinary Americans.”); Stephen Skowronek, John A. Dearborn & Desmond King, Phantoms of a Beleaguered Republic: The Deep State and the Unitary Executive 36–37 (2021) (noting how Framers sought to detach President’s selection from “electoral politics, and to favor individuals who mirrored the interests of the whole nation”); Gregg, supra note 50, at 196 (“Presidents were to be compliant instruments neither of the will of the public nor that of the legislature ... . As a representative relating to other representatives in Congress, the president’s primary responsibility was to enhance the deliberative process of government and thereby to encourage reasonable public actions.”); Druckman & Jacobs, supra note 133, at 161 (“[T]he Framers of the US Constitution explicitly positioned the president to be independent of public opinion. The president was to be politically free to pursue what policies and administrative decisions he believed best furthered the country’s overall interests.”).

  135. [135]. See, e.g., Margaret H. Lemos, Democratic Enforcement? Accountability and Independence for the Litigation State, 102 Cornell L. Rev. 929, 966–67 (2017) (“Contemporary theorists continue to debate the meaning of representation, but most [reject] the most extreme ‘mandate’ or ‘delegate’ view, under which the representative is duty-bound to follow . . . constituents’ views. [T]hey recognize that constituent preferences are an important policy input. Genuine representation means that representatives will not lightly depart from their constituents’ wishes. Decisions that deviate from public preferences are permissible, but they must be capable of justification in terms of the public interest.” (footnotes omitted)). To be clear, contemporary theorists typically reject pure versions of the trustee model as well. See, e.g., id. at 967 (collecting sources); Pitkin, supra note 47, at 165–66 (“[T]he basic question of the mandate-independence controversy is wrongly put. It poses a logically insoluble puzzle, asking us to choose between two elements that are both involved in the concept of representation.”).

  136. [136]. Pitkin, supra note 47, at 165 (emphasis added); see also id. at 166 (“[T]here is no universal, safe principle to guide one in [the] dilemma [of whether to follow his constituents’ wishes or to do what he thinks is in their interest]. Neither ‘follow their wishes’ nor ‘ignore their wishes’ will do; the decision must depend on why they disagree, and in a practical case that means his judgment on why they disagree. But the standard by which he will be judged as a representative is whether he has promoted the objective interest of those he represents.”).

  137. [137]. See id. at 164–65 (“[T]he representative’s duty, his role as representative, is generally not to get reëlected, but to do what is best for those he represents. In a democracy, the voters pass the final judgment ... on their representative by reëlecting him or refusing to do so. But it does not follow that whatever will get him reëlected is what he is obligated to do, or is equivalent to ‘true’ representation.”).

  138. [138]. As Pitkin states, although representatives will typically act consistent with their constituents’ wishes, that

    is not equivalent to saying that [a representative] represents only when he acts in accord with [her constituent’s] actual, conscious wishes. Quite the contrary: leadership, emergency action, action on issues of which the people know nothing are among the important realities of representative government. They are not deviations from true representation, but its very essence. It is often for that very purpose that people choose representatives.

    Id. at 163.

  139. [139]. See id. at 166.

  140. [140]. These problems arise most starkly under conceptions that would require the President to be motivated by electoral sanction as part of what it means to be a representative. See Rehfeld, Representation Rethought, supra note 47, at 223 (suggesting scholars who might adopt this view). To be clear, I cannot fully capture the enormous literature on political representation in this short section. Some theories of political representation will impose greater or lesser problems for the straightforward answer. See generally Dovi, supra note 46 (summarizing various contemporary normative conceptions of representation and discussing how representatives will be judged differently depending on the conception).

  141. [141]. I take this to be why Hemel and Posner and Michael Dorf seem to reject this view, at least in the context of presidential obstruction of justice and impeachment, respectively. See Hemel & Posner, supra note 5, at 1314–15; Michael C. Dorf, Dershowitz’s “L’état c’est Trump” Is Not As Crazy As It Sounds, But It Doesn’t Benefit Trump, Dorf on L. (Jan. 31, 2020), http://www.dor [] (“Given the ubiquity of political motives, every president would be committing multiple impeachable acts on a routine basis” if they could be impeached “for taking an otherwise lawful step . . . for a political motive.”).

  142. [142]. Cf. Martti Koskenniemi, From Apology to Utopia: The Structure of International Legal Argument 17 (2005) (“A law which would base itself on principles which are unrelated to State behaviour, will or interest would seem utopian, incapable of demonstrating its own content in any reliable way.”).

  143. [143]. Recall that, at present, we are assuming the President is acting for a single motive, rather than for mixed motives. So the claim must be that the President frequently takes action motivated exclusively by re-election, not by the public interest.

  144. [144]. Cf. Smith, supra note 50, at 210 (giving examples of presidents acting based on their own conscience, rather than political motives); id. at 211 (“[I]ndividual elements of the Trustee model, particularly the reliance on conscience, are factors in many presidents’ interpretations of their role.”).

  145. [145]. Hemel & Posner, supra note 5, at 1314.

  146. [146]. Indeed, there has long been a powerful norm—enforced albeit imperfectly in various bodies of case law—against using governmental power for partisan purposes. See Katherine Shaw, Partisanship Creep 3, 6­–10 (Apr. 19, 2022) (unpublished manuscript) (on file with author).

  147. [147]. See, e.g., Savannah Behrmann, RNC: Trump Criticized for Using White House as a Backdrop for the Convention, USA Today (Aug. 26, 2020, 12:05 AM),
    /news/politics/elections/2020/08/25/rnc-white-house-convention-speeches-ethics-hatch-act-trump/5628864002 []; Baker, supra note 1.

  148. [148]. See, e.g., Gregg, supra note 50, at 25 (“Symbolically, presidents are expected to be above party while still being of party. This tension becomes particularly strained during national election cycles when the nation is charged with partisanship amidst the quest of candidates and their supporters for public office. The tension, however, remains always beneath the surface of any presidency.” (emphasis omitted)).

  149. [149]. See, e.g., Howell & Moe, supra note 133, at 107 (stating legacy concerns are “motivator that most forcefully drives presidential behavior”).

  150. [150]. How to deal with “mixed motive” cases is addressed in Section III.B.

  151. [151]. Cf. Pitkin, supra note 47, at 222 (“Even if the representative does not examine his conscience as to the national interest on every issue, he may still be following a course of action designed to promote that interest. He may be playing his complicated role in the institutionalized political system in such a way that it strikes us as—that it is—representing.”).

  152. [152]. Focusing our efforts on protecting our democratic process is familiar in constitutional law as the basis for John Hart Ely’s famous version of “political process theory,” which sought to justify special judicial scrutiny where there were fundamental breakdowns in the democratic process. See John Hart Ely, Democracy and Distrust: A Theory of Judicial Review 73–104 (1980); Aaron Tang, Reverse Political Process Theory, 70 Vand. L. Rev. 1427, 1439 (2017) (describing theory as suggesting that “judges should intervene to enforce the rules of the game” when the political process is not working under political process theory). Although political process theory no longer seems to hold much sway in the Court (or academy), see id., the intuition that there is something especially important about protecting our democratic process certainly remains.

  153. [153]. Hemel & Posner, supra note 5, at 1312 (emphasis omitted).

  154. [154]. Id. at 1314.

  155. [155]. Id.

  156. [156]. See id. (emphasis added).

  157. [157]. See id.

  158. [158]. See id.

  159. [159]. Cf. Ezra Klein, The Unpersuaded: Who Listens to a President?, New Yorker (Mar. 12, 2012), [
    9N-T54Y] (“Put [] simply, the President’s party can’t win unless the other party loses.”).

  160. [160]. Hemel & Posner, supra note 5, at 1314.

  161. [161]. Id. at 1314–15.

  162. [162]. Id. at 1314.

  163. [163]. Id. at 1315.

  164. [164]. Id.

  165. [165]. Id. (internal quotation marks omitted).

  166. [166]. Id.

  167. [167]. Id.

  168. [168]. Id.

  169. [169]. Id.

  170. [170]. Id. at 1316. They conclude this is so, despite seeming “partisan rather than public-spirited,” because President Obama’s concern about “partisan polarization is close enough to a legitimate conception of the public interest that applying the obstruction statutes in such a case would threaten his ability to do what he believes is best for the nation.” Id.

  171. [171]. Id.

  172. [172]. Id.

  173. [173]. Id.

  174. [174]. Recent work seeking to identify potential rules and reforms necessary to entrench our democracy might help to develop such an alternative. See, e.g., Tom Ginbsurg & Aziz Z. Huq, How to Save a Constitutional Democracy 205-37 (2018).

  175. [175]. See generally Verstein, supra note 13 (categorizing the treatment of mixed motives in numerous areas of law).

  176. [176]. See id. at 1134–43 (describing tests and collecting cases).

  177. [177]. Id. at 1134.

  178. [178]. Id. at 1136.

  179. [179]. Id. at 1137. For work suggesting this standard should apply to the President in the context of obstruction of justice and animus in immigration, see Hemel & Posner, supra note 5, at 1319–20; Ray, supra note 4, at 66–69.

  180. [180]. Verstein, supra note 13, at 1139–41.

  181. [181]. Id.

  182. [182]. Id. at 1141.

  183. [183]. Id. at 1132–33 (describing such cases of “tiny” motive).

  184. [184]. We might understand the Court’s discussion of motive in the challenge to the Trump Administration’s putting a citizenship question back on the census to support this. The Court seemed comfortable with “politics” playing a role, likely so long as the public-interested motive played at least some meaningful role. See, e.g., Eidelson, supra note 9, at 1789–90 (discussing Dep’t of Commerce v. New York, 139 S. Ct. 2551, 2573 (2019)).

  185. [185]. For an argument against but-for motive standards in general, see generally Andrew Verstein, The Failure of Mixed-Motives Jurisprudence, 86 U. Chi. L. Rev. 725 (2019). I think the context of presidential power and politically motivated behavior is sufficiently distinct from Verstein’s focus such that a but-for motive is at least potentially defensible here. This is true for several reasons: (1) the impermissible motive here—acting based on politics—is not nearly as bad as the impermissible motive Verstein focuses on relating to animus; cf. id. at 784–88; cf. Ray, supra note 4, at 65–66; (2) unlike the situations Verstein focuses on, impermissible motives are likely to be more common for the President than the impermissible motives he focuses on; cf. Verstein, supra, at 778 (“Most people don’t have high levels of [impermissible motive] ... .”); and (3) it is not clear to me that the benefit of “airing” the bad motive in the context of the President is high, cf. id. at 793–94, rather it might actually be more beneficial to focus on why conduct is in the public interest, rather than private interest.

  186. [186]. See Hemel & Posner, supra note 5, at 1312; Ray, supra note 4, at 65.

  187. [187]. See Verstein, supra note 13, at 1137–39.

  188. [188]. See, e.g., Verstein, supra note 185, at 778 (“Most people don’t have high levels of [impermissible] Motive [in context of employment discrimination], making this region of protection useful to many people and making the region of risk less frightening to most people.”); see also Ray, supra note 4, at 65.

  189. [189]. See Ray, supra note 4, at 65.

  190. [190]. See id.; Hemel & Posner, supra note 5, at 1319–20.

  191. [191]. Hemel and Posner suggest a but/for mixed motive test for narrow pecuniary, personal, or partisan motivated behavior. See Hemel & Posner, supra note 5, at 1319–20. Ray suggests such a standard for animus in immigration. See Ray, supra note 4, at 65.

  192. [192]. See supra note 11 and accompanying text.

  193. [193]. See e.g., Curtis A. Bradley & Trevor W. Morrison, Presidential Power, Historical Practice, and Legal Constraint, 113 Colum. L. Rev. 1097, 1098 (2013) (“A variety of justiciability limitations
    ... are regularly invoked by courts as a basis for declining to resolve issues of presidential power, especially when individual rights are not directly implicated.”); Roisman, supra note 53, at 892
    –900 (discussing congressional and judicial review of presidential fact-finding).

  194. [194]. See also Roisman, supra note 53, at 873–92 (arguing for greater attention to internal executive branch constraints).

  195. [195]. See Shalev Roisman, Presidential Law, 105 Minn. L. Rev. 1269, 1326–27 (2021) (discussing such incentives).

  196. [196]. See id. at 1331; see also Andrew Rudalevige, The Contemporary Presidency: Executive Orders and Presidential Unilateralism, 42 Presidential Studs. Q. 138, 148–51 (2012).

  197. [197]. The requirement could be limited to sufficiently “important” directives, if requiring it for all directives proved too cumbersome. Cf. Roisman, supra note 53, at 888 (discussing creating “significance” threshold for presidential directives).

  198. [198]. See Exec. Order No. 11,030, § 2(a), 3 C.F.R. § 610 (1959–1963); see Roisman, supra note 53, at 876 (discussing process required by this order). Although such explanations are common when directives are issued, they are not always provided or required. The order does require inclusion of a letter “explaining the nature, purpose, background, and effect of the proposed Executive order or proclamation.” See Exec. Order No. 11,030, § 2(a), 3 C.F.R. § 610 (1959–1963). Such a letter may well often include a public-regarding reason, but it does not, by itself, require it, and the letter need not be published. A better approach would be to require a public-facing public-regarding reason for such action.

  199. [199]. See, e.g., Ashley S. Deeks, Secret Reason-Giving, 129 Yale L.J. 612, 626–36 (2020) (discussing benefits of public reason-giving and addressing critiques that it is insincere or ossifying); Stephen Holmes, In Case of Emergency: Misunderstanding Tradeoffs in the War on Terror, 97 Calif. L. Rev. 301, 354 (2009) (“[P]ublic officials perform best, even during emergencies, when forced to give reasons for their actions.”); cf. Sunstein, supra note 49, at 78 (“Requiring justifications does not, to be sure, guarantee ‘reasoned analysis’ on the part of the legislature. Boilerplate, representing not the actual process of decision but instead a necessary bow to the courts, is hardly an unambiguous good and would undoubtedly be increased by the proposed requirements. But requiring justifications does serve an important prophylactic function... . [I]dentification of the legitimate public purposes purportedly served by statutory classifications should improve representative politics by ensuring that the deliberative process is focused on those purposes and the extent to which the classifications serve them ... .”).

  200. [200]. Cf. Sunstein, supra note 49, at 78 (“[P]rocedural requirements occasionally have substantive consequences. Identification of the actual purposes served by a statute may, for example, diminish the likelihood of its enactment.”).

  201. [201]. Cf. Harold H. Bruff, Judicial Review and the President’s Statutory Powers, 68 VA. L. Rev. 1, 46 (1982) (suggesting costs of explanation requirement are “tolerable, although they are not insignificant” in light of benefits).

  202. [202]. See, e.g., Deeks, supra note 199, at 616 (arguing that “secret reason-giving confers a variety of benefits, even if those benefits manifest themselves in a somewhat different way than in the public context”).

  203. [203]. See generally Daphna Renan, Presidential Norms and Article II, 131 Harv. L. Rev. 2187 (2018) (providing general account of presidential norms).

  204. [204]. How to define what qualifies as a “high crime or misdemeanor” is a topic that has sustained several books. See supra note 3 (collecting sources).

  205. [205]. See id. (collecting sources).

  206. [206]. Gerhardt, supra note 3, at 219.

  207. [207]. See, e.g., Sunstein, supra note 3, at 119–24 (providing such examples as easy cases).

  208. [208]. See, e.g., Trump Brief, supra note 33, at S319 (“[B]y making impeachment turn on nearly impossible inquiries into the subjective intent behind entirely lawful conduct, House Democrats’ standard would open virtually every presidential decision to partisan attack based on questioning a President’s motives.”); Blackman & Tillman, supra note 33 (“Opening this door will not root out ‘corruption.’ Rather, the impeachment process will simply drive normal politics and compromise underground.”).

  209. [209]. See U.S. Const. art. I, § 3.

  210. [210]. See Roisman, supra note 53, at 892 (collecting examples).

  211. [211]. Cf. Roisman, supra note 185, at 1331­­–35 (addressing potential constitutional objections to such procedural requirements).

  212. [212]. See, e.g., Nixon v. Fitzgerald, 457 U.S. 731, 756–57 (1982) (noting “an inquiry into the President’s motive . . . could be highly intrusive” and would “subject the President to trial on virtually every allegation” thus “depriv[ing] absolute immunity of its intended effect”).

  213. [213]. See, e.g., Eidelson, supra note 9, at 1805 n.279 (“I take the risk of inappropriately outcome-driven decisionmaking to be endemic to judicial review ... .”); David A. Strauss, The Supreme Court 2014 Term—Foreword: Does the Constitution Mean What It Says?, 129 Harv. L. Rev. 1, 61 (2015) (“Constitutional law has its share of complexity and mystery, but in the end it is law—subject to being manipulated and abused ... as all law is, but also capable of being applied in good faith.”). It is worth noting that people’s empirical assumptions about the likelihood of judicial error might justifiably change over time, because, inter alia, the composition of the courts change over time.

  214. [214]. This section brackets the complex question of what precise method or evidentiary standard courts ought to use to identify the President’s motive. Cf. Eidelson, supra note 9, at 1794 n.222, 1812 (bracketing issue). This is obviously an important question but one that I do not have space to address. Whatever the ideal standard, judicial error is a sufficiently large concern such that the threshold for finding the President’s motive impermissible ought to be quite high.

  215. [215]. Both Verstein and Fallon suggest variations of this approach in some contexts. See Verstein, supra note 13, at 1134­–36; Fallon, supra note 9, at 579 (“[T]he legislature’s breach of its deliberative obligations should bring deference to an end and ... should provoke elevated judicial scrutiny.”).

  216. [216]. See 50 U.S.C. 4613(a) (providing sanctions power upon this finding).

  217. [217]. Cf. Sierra Club v. Trump, 929 F.3d 670, 680 (9th Cir. 2019) (finding border wall not a “military construction project[]”).

  218. [218]. See Trump v. Hawaii, 138 S. Ct. 2392, 2413, 2420 (2018).

  219. [219]. See Roisman, supra note 53, at 847 (distinguishing between findings of fact and policy judgment); cf. Zivotofsky ex rel. Zivotofsky v. Clinton, 566 U.S. 189, 197 (2012) (“Our precedents have . . . found the political question doctrine implicated when there is ‘a lack of judicially discoverable and manageable standards for resolving’ the question before the court.” (quoting Nixon v. United States, 506 U.S. 224, 228 (1993))).

  220. [220]. But see Verstein, supra 185, at 769 (“If a president’s executive order is vital for national security, the Department of Justice can defend it by establishing the objective importance of the law on a nondeferential standard. Considerations of expertise, secrecy, and deference make this inappropriate in cases of unimpeached motives, but quite fitting when the defendant-executive has revealed problematic judgment.”). Similarly, Fallon’s proposal to rely exclusively on substantive objective tests strikes me as unworkable in this domain, at least until we have a developed substantive test to evaluate presidential exercises of power, which we currently lack. Fallon notes that “different constitutional provisions give rise to different substantive norms,” Fallon, supra note 9, at 560, and concludes that courts should never strike down a law solely based on a legislature’s impermissible motive, rather than based on the substantive test. He focuses on laws motivated by reasons that violate certain individual rights protections and thus he would apply judicial strict scrutiny in these contexts. See id. at 530. But while strict scrutiny is well-suited to evaluating individual rights claims, it is not set up to ask about broad policy questions like whether or not the entry of certain noncitizens is detrimental to the interests of the United States. And, at least to date, we do not have a clear alternative substantive test that applies to presidential determinations of this sort—although I am working on creating one in a work-in-progress, tentatively titled “The President’s Subjective and Objective Legal Obligations.”

  221. [221]. See generally Note, The Presumption of Regularity in Judicial Review of the Executive Branch, 131 Harv. L. Rev. 2431, 2431 (2018) (“[C]ourts will sometimes presume that official duties have been properly discharged until the challenger presents clear evidence to the contrary.” (internal quotation marks omitted)).

  222. [222]. See Eidelson, supra note 9, at 1801–02.

  223. [223]. This suggestion is in line with recent work on how to structure judicial review of presidential directives more broadly. See, e.g., Lisa Manheim & Kathryn A. Watts, Reviewing Presidential Orders, 86 U. Chi. L. Rev. 1743, 1814–15 (2019) (calling for requirement that “the president would be required to set forth a nonarbitrary justification in the text of her orders themselves”).

  224. [224]. See Eidelson, supra note 9, at 1758 (noting reasoning in these cases “begins from a simple premise: Political accountability sometimes depends on the public’s understanding not only what the government has done, but why”).

  225. [225]. See DHS v. Regents of the Univ. of Cal., 140 S. Ct. 1891, 1089 (2020).

  226. [226]. See Dep’t of Com. v. New York, 139 S. Ct. 2551, 2574–76 (2019).

  227. [227]. See Eidelson, supra note 9, at 1822–23.

  228. [228]. Id. at 1826.

  229. [229]. To be clear, this is a normative proposal. Unlike for agencies, such reason-giving is not currently required of the President by positive law. Agencies are required to give such reasons pursuant to administrative law which does not apply to the President. See Franklin v. Massachusetts, 505 U.S. 788, 800–01 (1992) (noting the Administrative Procedure Act does not apply to the President); see also Trump v. Hawaii, 138 S. Ct. 2392, 2409 (2018) (labeling as “questionable” claim that statute required President to make relevant finding “with sufficient detail to enable judicial review”).

  230. [230]. Although extra procedure inevitably imposes some costs on the President’s internal ability to take action, I do not think requiring the President to give a public-interested reason when exercising power would be overly costly given the apparatus and practices he already uses to approve written directives. See supra note 188 and accompanying text. But, as noted above, allowing courts to review the President’s motive for pretext would impose new costs in the form of litigation, discovery costs, as well as the potential for judicial error.

  231. [231]. Cf. Dep’t of Com., 139 S. Ct. at 2576 (Thomas, J., concurring in part and dissenting in part) (“It is not difficult for political opponents of executive actions to generate controversy with accusations of pretext, deceit, and illicit motives. Significant policy decisions are regularly criticized as products of partisan influence, interest-group pressure, corruption, and animus. Crediting these accusations on evidence as thin as the evidence here could lead judicial review of administrative proceedings to devolve into an endless morass of discovery ... .”).

  232. [232]. Cf. id. at 2575 (majority opinion) (“We are presented ... with an explanation for agency action that is incongruent with what the record reveals about the agency’s priorities and decisionmaking process. It is rare to review a record as extensive as the one before us when evaluating informal agency action—and it should be. But having done so for the sufficient reasons we have explained, we cannot ignore the disconnect between the decision made and the explanation given. Our review is deferential, but we are ‘not required to exhibit a naiveté from which ordinary citizens are free.’” (quoting United States v. Stanchich, 550 F.2d 1294, 1300 (2d Cir. 1977))).

  233. [233]. Cf. Sunstein, supra note 49, at 78 (making this argument in context of legislation).

  234. [234]. See, e.g., Kathrine Shaw, Conventions in the Trenches, 108 Calif. L. Rev. 1955, 1974 (2020) (“Both the travel ban litigation and the Census citizenship litigation featured district court inquiries into executive-branch decisions and conventions. In both cases, such inquiries generated valuable forms of public knowledge, created important pressure on executive-branch actors, and led to meaningful policy change.”); Z. Payvand Ahdout, Enforcement Lawmaking and Judicial Review, 135 Harv. L. Rev. 937, 960 (2022). I am especially indebted to Kate Shaw for making this point clear to me.

  235. [235]. The census case, although dealing with the Secretary of Commerce, not the President, is one such example. See Dep’t of Com., 139 S. Ct. at 2574–76.


Associate Professor of Law, University of Arizona James E. Rogers College of Law. The author served in the Office of Legal Counsel in the U.S. Department of Justice from 2015 to 2017.

The views expressed are the author’s own and are based entirely on publicly available materials. Thank you to Bernard Bell, Kiel Brennan-Marquez, Jacob Bronsther, Daniel Hemel, Aziz Huq, Toni Massaro, Nina Mendelson, Michael Sant’Ambrogio, Steve Schaus, Kate Shaw, Will Thomas, Andrew K. Woods, the students in my Spring 2022 Presidential Power Seminar, as well as participants at various workshops, including the Duke-Virginia Foreign Relations Law Roundtable, the Administrative Law New Scholarship Roundtable hosted at Yale Law School, the Administrative Law “New Voices” Panel at the AALS Annual meeting, the Twelfth Annual Constitutional Law Colloquium at George Washington University, the Research Accountability Group, and the University of Arizona Fall Faculty Workshop. Thank you also to Vinny Venkat for excellent research assistance. I am especially indebted to Andrew Coan and Diana Newmark for repeatedly pushing me on various matters and dramatically improving the work. All errors are my own.