The Rationality of Rational Basis Review

Through the “rational basis” test, the Supreme Court asserts the authority to assess whether laws are “rationally related to a legitimate governmental interest.” Although it gives the Court an effective substantive veto over all legislation, rationality review is poorly understood and under-theorized. Developed haphazardly over time, rationality review is not the product of either a considered formula or a particular theory of constitutional law, and though it is clothed in the language of rationality, it represents the Court’s own decidedly intuitive understanding of the proper sphere of state regulation. At one time, that understanding was based in a widely held conception of the “police power,” but the connection to the police power was severed after the Court’s decision in United States v. Carolene Products. Since then, the Court has developed ad hoc a conception of the proper role of government that has become almost entirely utilitarian in nature.

This Article examines the Court’s view of how rationality should (and by virtue of the power of judicial review must) feature in legislation by tracing the development of rationality review and comparing it to more rigorous understandings of political rationality. Comparison reveals the Court’s limited conception of rationality, which allows the Court to avoid difficult questions in pursuit of seemingly uncontroversial instrumental ends. Examination of the Court’s approach to rationality demonstrates the need for a broader conception of legislative rationality – one that includes “constitutive ends.” Recognizing constitutive legislative ends, combined with an information-forcing rule for revealing those ends, can both improve democratic discourse in the legislature and lead to a richer and more intellectually honest form of rationality review.

RJR Nabisco and the Runaway Canon

In last term’s RJR Nabisco, Inc. v. European Community,[1] the U.S. Supreme Court held that the private remedy in the Racketeer Influenced and Corrupt Organizations Act (“RICO”)[2] does not extend to foreign injuries, even if those injuries were caused by a U.S. company operating within the United States.[3] In doing so, the Court finished transforming the presumption against extraterritoriality from a tool meant to effectuate congressional intent into a tool for keeping Congress in check. The presumption against extraterritoriality has become a means for judges (particularly Justices) to override Congress in defining the proper scope of litigation in U.S. courts.

The RJR Nabisco case, like many transnational cases, was both global and local in scope. The European Community and twenty-six of its member states had been investigating major tobacco companies for their role in cigarette trafficking and money laundering into and through Europe.[4] While other tobacco companies eventually reached settlements with the European Commission, RJR Nabisco did not and continued—according to the European Community’s complaint—to engage in illegal activity,[5] specifically by scheming “to sell cigarettes to and through criminal organizations and to accept criminal proceeds in payment for cigarettes.”[6] This conduct was causing harm in Europe, but the European Community believed it was “directed and controlled” by “[h]igh-level managers and employees” from RJR Nabisco’s headquarters in the United States.[7]

The Supreme Court threw out the lawsuit after invoking the presumption against extraterritoriality. That canon of statutory interpretation instructs judges to assume “that legislation of Congress, unless a contrary intent appears, is meant to apply only within the territorial jurisdiction of the United States.”[8] In applying the presumption in RJR Nabisco, however, a majority of four Justices[9] rejected multiple indications that Congress intended RICO’s private right of action to extend abroad[10] while raising the bar on what Congress must do to make its extraterritorial expectations clear.[11]

Besides the worrisome implications for separation of powers, the majority’s opinion was also disappointing on practical grounds. By applying the presumption too aggressively, the Court missed an opportunity to provide much-needed guidance to judges on how to interpret statutes that rebut the presumption. For despite the Court’s recent wariness of extraterritorial laws,[12] Congress does sometimes intend its statutes to apply abroad.[13] Those extraterritorial statutes nonetheless have limits—but the Court has not clearly explained how judges are to identify them.[14] Without such guidance, judges may be tempted to cling too tightly to the presumption in order to avoid the doctrinal black hole on the other side.

This Essay thus concludes with advice to judges about how to interpret statutes that do indicate Congress’s extraterritorial intent: First, while judges are bound by RJR Nabisco’s holding, they should not feel obligated to repeat its problematic modes of reasoning.[15] Second, judges should not be wary of finding the presumption rebutted for fear of what comes next. On the one hand, there are other doctrines that can help judges navigate jurisdictional conflict; on the other, extraterritorial statutes on their own terms have outer limits, and the Court has provided clues elsewhere for how judges might identify them.[16]

I. The Modern Presumption Against Extraterritoriality

Though considered a “longstanding principle of American law,”[17] the presumption against extraterritoriality fell into disuse after the 1940s.[18] The Restatement (Third) of Foreign Relations Law, published in 1987, did not even bother to include it.[19] But starting in the 1990s, the Rehnquist and Roberts Courts turned back to the presumption as a means for curbing the scope of transnational litigation in U.S. courts.[20]

When the Court in 1991 breathed new life into the presumption in Equal Employment Opportunity Commission v. Arabian American Oil Co. (Aramco),[21] however, it also transformed it.[22] On the one hand, Aramco invoked the old presumption as it had last been defined half a century earlier: It is a tool, the Court explained, “whereby unexpressed congressional intent may be ascertained,”[23] with judges looking for “language in the [relevant Act] [that] gives any indication of a congressional purpose to extend its coverage” beyond U.S. territory.[24] But on the other hand, and in the same breath, Aramco required an “affirmative intention of the Congress clearly expressed” before a statute could be construed to apply extraterritorially.[25]

This language came from a separate line of cases applying the Charming Betsy[26] canon. As Professor John Knox has explained, that canon assumes Congress does not intend to violate international law and thus requires Congress to indicate clearly when it is doing so.[27] This conflation of the traditional presumption with Charming Betsy’s stricter requirement has predictably led to the presumption increasingly resembling a clear statement rule—even while the Court continues to insist that it is not.[28] In 1993, the Court required “clear evidence of congressional intent” to overcome the presumption.[29] By 2010, the Court could state more bluntly, “[w]hen a statute gives no clear indication of an extraterritorial application, it has none.”[30] And in RJR Nabisco, the Court phrased the inquiry as how far Congress “has affirmatively and unmistakably instructed” the statute to reach.[31]

Along the way, the Court has formalized the modern presumption into a two-step inquiry.[32] At step one, a judge must look for this “clear indication” of extraterritorial effect.[33] If the judge does not find such an indication—that is, if the presumption is not rebutted—then the judge continues to step two, in which she determines the “focus” of the statute.[34] For a case to fall under the statute’s ambit, its connections to the United States must match the statute’s “focus.”[35]

Like the ratcheting up of the language required to overcome the presumption, this two-step framework moves the presumption further away from the purported search for congressional intent.[36] As Professor Lea Brilmayer has explained, step one requires Congress to be emphatic when it wishes its statutes to apply abroad, while at step two, judges get to decide what domestic contacts count in which cases (determining the “focus” of a statute, after all, is a rather mushy directive).[37] And while it is helpful to give judges such a clear structure for thinking about how to handle transnational cases, that guidance has been lopsided: The Court has not provided similar guidance on what to do when the presumption is rebutted.[38]

RJR Nabisco presented such an opportunity, as the Justices unanimously agreed that RICO’s substantive provisions do extend extraterritorially.[39] RICO targets “racketeering activity,” which is comprised of certain state or federal criminal offenses that the RICO statute terms “predicate acts.”[40] These predicate acts, listed in 18 U.S.C. § 1961, include some crimes that explicitly reach conduct beyond U.S. borders.[41] Section 1962—the heart of RICO—prohibits four different ways by which a “pattern of racketeering activity” may be used to infiltrate, control, or operate an “enterprise.”[42] Because § 1962 incorporates § 1961’s definition of racketeering activity, which in turn incorporates other explicitly extraterritorial statutes, the Court had no trouble concluding that § 1962 reaches foreign racketeering activity, at least in some cases.[43]

The next step should have been to consider whether this extraterritorial statute was nonetheless subject to other limits, whether on its own terms or due to other comity-based doctrines.[44] Instead, the Court announced a new requirement that the presumption be applied separately to every statutory provision, whether substantive, remedial, or jurisdictional.[45] Based on that requirement, the four-Justice majority applied the presumption separately to RICO’s private right of action, found in § 1964(c),[46] and concluded that it did not independently overcome the presumption.[47] Thus a “private RICO plaintiff . . . must allege and prove a domestic injury to its business or property.”[48] That second application of the presumption was ill considered and provides a problematic model for the lower courts.

II. The Runaway Canon

The majority’s application of the presumption to RICO’s private right of action was ill considered along at least two dimensions. First, the majority rejected two standard legislative methods by which Congress could efficiently signal its extraterritorial intent. Second, the new requirement that Congress express its extraterritorial intent in every provision of a statute reflects an unrealistic understanding of how Congress works. In the hands of the RJR Nabisco majority, the presumption has become less a method for interpreting statutes than a pronouncement on the proper scope of access to U.S. courts, a pronouncement that Congress must labor to displace.[49]

A. Ignoring Congressional Intent

One common way that Congress indicates its geographic intent is to incorporate by reference another statute that is more explicitly extraterritorial. Indeed, that was the basis on which the Court determined that RICO’s substantive provisions reach abroad: “The most obvious textual clue” to § 1962’s extraterritorial scope, the Justices agreed, was its incorporation of § 1961, which in turn incorporated statutes that “plainly apply to at least some foreign conduct.”[50] “Short of an explicit declaration,” Justice Alito reasoned, “it is hard to imagine how Congress could have more clearly indicated that it intended RICO to have (some) extraterritorial effect.”[51] Yet when it came to RICO’s private right of action, the majority had no trouble imagining a very different congressional intent behind § 1964(c)’s incorporation of § 1962.[52] Even though § 1964(c) provides a remedy for “[a]ny person injured in his business or property by reason of a violation of section 1962,”[53] the majority refused to treat that incorporation of § 1962 as rebutting the presumption against extraterritoriality.[54]

Alternatively, the Court might have looked to Congress’s decision to model RICO’s private right of action after that of the Clayton Act, which the Supreme Court had previously held does allow recovery for injuries suffered abroad.[55] As Justice Ginsburg noted in dissent, “[t]he similarity of language in [the two statutes] is, of course, a strong indication that [they] should be interpreted pari passu.”[56] But that, too, the majority determined, is not sufficient to indicate Congress’s intent for RICO’s § 1964(c) to similarly apply to foreign harms.[57] In rejecting the statutory analogy to the Clayton Act, the majority pointed to the different definition of “person” under the Clayton Act, which explicitly extends to foreign business organizations—even though that difference had not prevented the Court from interpreting RICO’s private right of action to align with the Clayton Act’s private right of action in the past.[58]

To add insult to injury, the majority seized on the Clayton Act-like language in § 1964(c) to bolster its conclusion that § 1964(c) does not extend extraterritorially. To model RICO’s private right of action after that of the Clayton Act (which, again, the Supreme Court had previously found extended to foreign injuries), Congress limited § 1964(c) to injuries to “business or property.”[59] Rather than treat this language as an indication of Congress’s intention that the two Acts should be interpreted similarly, the majority reasoned that this language “signaled” Congress’s intent that RICO’s “civil remedy is not coextensive with [its] substantive provisions,” and thus that the remedy’s geographic scope presumably differed from that of the rest of the statute.[60] It seems that Congress cannot win.

B. Raising the Bar

In rejecting both statutory incorporation and statutory modeling as indications of congressional intent, the RJR Nabisco majority made it harder for Congress to efficiently rebut the presumption against extraterritoriality. Nor did the majority indicate any preferable alternative, short of a clear statement of extraterritoriality. At the same time, it introduced a new requirement that Congress reiterate its extraterritorial intent in every provision of a statute, whether jurisdictional, substantive, or remedial.[61] Even if the Court’s view of congressional intent (and ability) were realistic, it keeps moving the goal further down the field. The result is not a search for congressional intent, but an effort to put the brakes on what Congress can do.[62]

The Court seems to presume that it is not difficult for Congress to state its extraterritorial intent, but that ignores several realities. First there is the difficulty of the drafting process itself (and the inertia for amending misinterpreted statutes).[63] Professors Abbe Gluck and Lisa Bressman have also shown that congressional staffers are simply not aware of such judicially required clear-statement rules.[64] And then there is the possibility of introducing more unintended errors the more that Congress does say explicitly. For example, Congress tried to overturn in part the Court’s narrowly territorial interpretation of the Securities Exchange Act in Morrison v. National Australia Bank, Ltd.,[65] but there is some question whether its amendment to the Securities Exchange Act was phrased and framed correctly to achieve this purpose.[66] After all, providing the clear statement that the Court seems to want is not as simple as stating “this provision applies extraterritorially.” Drafters have to account for the limits on U.S. extraterritorial jurisdiction under international law, limits that courts may sometimes be better situated to interpret and apply through the Charming Betsy canon.

Further, the Court’s new insistence that judges seek such clear extraterritorial intent in every provision is ill-advised (and one is to hope short-lived). As Professor Bill Dodge has cogently argued, applying the presumption to jurisdictional provisions would be deeply disruptive, as well as irreconcilable with the Court’s reasoning in other recent cases—including other portions of RJR Nabisco itself.[67] It also cannot possibly reflect existing congressional intent, as Congress has not been in the habit of writing extraterritoriality into the separate jurisdictional and remedial provisions of statutory schemes that are unarguably intended to be extraterritorial in scope. As Professors Hannah Buxbaum and Pam Bookman have noted, for example, when Congress overrode the Supreme Court’s narrow interpretation of Title VII in Aramco by revising the law to clarify its extraterritorial reach, it did not separately clarify the extraterritorial reach of the law’s remedial provisions.[68] If RJR Nabisco were applied strictly, then, that clear congressional intent behind the Civil Rights Act of 1991 would be nullified.

In short, the presumption has run away from its stated purpose of effectuating congressional intent. Instead it is generating an ever-growing series of hoops through which Congress must jump if it wants its laws to extend beyond U.S. borders.[69] In applying this transformed presumption, the Supreme Court poses as a faithful agent of congressional intent, but it is in fact a disciplinarian of Congress’s global aspirations.

III. What Comes Next

In overextending the presumption against extraterritoriality in RJR Nabisco, the Court missed an opportunity to give judges better guidance on what can rebut the presumption, and if it is rebutted, what happens next. This final Part offers some suggestions about what judges might do to help bring this runaway canon back home.

A. Applying the Presumption

When applying the presumption to other statutes in the future, the best option for judges is to do what the Court says in RJR Nabisco, not what it does. The majority’s rhetoric does not constrain the lower courts, and the modes of reasoning the majority used or discounted do not dictate the modes of reasoning lower courts must use when analyzing other statutes.[70]

First, RJR Nabisco should not be read as casting doubt on the relevance of incorporated statutes or analogous statutes in determining congressional intent. Indeed, judges need only look to the Court’s analysis of § 1962 to confirm the continued viability of statutory incorporation as a means for Congress to indicate the extraterritorial scope of a statute.

Second, judges should be wary of repeating the majority’s vague functional concerns about foreign relations. At the outset of its analysis of § 1964(c), the RJR Nabisco majority seemed to suggest that the presumption should be applied more rigorously when there is a danger of “international friction” or “risk of conflict” with foreign law.[71] This passage was largely rhetorical—a calling out of the seemingly inconsistent positions of European governments in this and other cases involving the presumption against extraterritoriality[72] (a point to which we will return[73]). Whatever its purpose, that language risks a dangerous ratcheting up of an already strict presumption. To the extent the general concern is legitimate—that courts should try to promote international comity by avoiding controversy—that concern is already embodied in the presumption itself, which is meant to help prevent unintentional discord with other nations.[74] There is no need to apply the presumption more rigorously when comity is at stake, as the presumption assumes comity is always at stake when U.S. law applies outside of U.S. territory.

Besides which, friction and comity cut both ways: Worse international discord might be caused by denying foreign plaintiffs remedies for the wrongdoing of U.S. nationals, including on U.S. territory, while at the same time allowing U.S. plaintiffs to sue foreign defendants for comparable conduct.[75] As Justice Ginsburg explained in dissent, “[m]aking such litigation available to domestic but not foreign plaintiffs is hardly solicitous of international comity or respectful of foreign interests.”[76] Rather, the better place to address specific comity concerns is in the post-presumption analysis.

B. What Happens Next

In applying the presumption, then, judges should not be afraid to find it is rebutted. While the Supreme Court has not clarified what the post-presumption analysis should look like,[77] the landscape on the other side of the presumption is not quite as messy as it might at first appear. Here are four guideposts for managing that analysis.

First, a point about semantics. The Court suggested in Microsoft Corp. v. AT&T Corp. that the presumption against extraterritoriality continues to apply even when a statute is explicitly extraterritorial; in that instance, the Court said, the presumption “remains instructive in determining the extent of the statutory exception.”[78] This is a question of labels, and saying a presumption applies after it has been rebutted will only sow confusion. Rather, the Court’s analysis in Microsoft turned on F. Hoffman-La Roche Ltd. v. Empagran, S.A.,[79] a case that did not discuss the presumption against extraterritoriality as such. Both Microsoft and Empagran are really cases about how to interpret statutes that rebut the presumption, and the language used to identify that analysis should signal as much.

Which brings us to a second point: The post-presumption analysis is still a question of statutory interpretation.[80] The Court in RJR Nabisco helpfully clarified that a statute’s “focus,” invoked at step two of the Morrison framework, is irrelevant to interpreting the scope of an extraterritorial law.[81] Instead, the relevant canon post-presumption is the Charming Betsy canon,[82] or the assumption that Congress does not legislate beyond the bounds of international law. Under international law, there are generally accepted limits on a nation’s prescriptive (or law-making) power. Most traditionally, countries can assert prescriptive jurisdiction over their nationals, their territory, and ships flying their flag; countries may also legislate regarding harm to their nationals or threats to their security; and all countries can legislate regarding a set of universal crimes.[83] When a particular interpretation of an extraterritorial statute would exceed these bases of jurisdiction under international law, judges should presume that was not Congress’s intent absent a clear statement.

The third point is that this rather clear outer limit from Charming Betsy—when combined with other doctrines like personal jurisdiction that help define the scope of transnational litigation in U.S. courts—will adequately help judges resolve most cases. Cases involving the extraterritorial application of federal statutes will often fall comfortably within the core jurisdictional zones of U.S. power. When cases stretch those limits, Charming Betsy provides a hard stop. Meanwhile, other doctrines can also help address comity concerns;[84] indeed, some of these doctrines have themselves been refined in recent years to better account for international comity.[85] As emphasized by Justice Ginsburg in her RJR Nabisco dissent, for example, the recent contraction of general jurisdiction will limit the risk that foreign defendants with thin ties to the United States can be hauled before U.S. courts.[86] Between those constitutional due process limits and international limits on prescriptive jurisdiction (as filtered through the Charming Betsy canon), there should be few cases that raise otherwise unaddressed comity concerns.

But what should judges do if confronted with one of those few remaining cases? This brings us to the fourth point, which is also the most speculative. One could read the Court’s opinions in Empagran and Microsoft as suggesting that an additional, Charming Betsy-inflected inquiry might apply to these zones of jurisdictional conflict.[87] The limits under international law are not always clear-cut, and the closer one approaches to those limits, the more likely one will cause consternation among nations whose core jurisdictional prerogatives—such as their own territorial jurisdiction—are affected.[88] This gray zone at the edge of permissible exercises of jurisdiction was the source of the Court’s unease in Empagran and Microsoft.

In Empagran, for example, the Court had to determine whether a provision of the Sherman Act (which is explicitly extraterritorial) extended to foreign injuries caused primarily by the foreign conduct of foreign actors that also (but separately) caused domestic injuries.[89] In holding that it did not, the Court asserted that it “ordinarily construes ambiguous statutes to avoid unreasonable interference with the sovereign authority of other nations,” and it suggested there is a “rule of statutory construction” that “cautions courts to assume that legislators take account of the legitimate sovereign interests of other nations when they write American laws.”[90] Though this formulation is unhelpfully vague, the Court was groping for a way to put some softer outer limits on U.S. law (as its reference to the Charming Betsy canon suggests).[91] The idea is that, even without a direct conflict with international law (which could only be overcome by a clear statement of Congress), judges should still be wary of interpreting statutes as reaching right up to those outer limits because doing so can infringe on the widely recognized sovereign interests of other states.

When it comes to effectuating this idea, however, Empagran did not provide a workable framework. As the Empagran Court seemed to recognize, this should not be an open-ended balancing or a vague standard that allows functional concerns (like the risk of “international friction”) to balloon over time.[92] Indeed, the malleability of such generalized functional concerns has led the Court to curtail their relevance in other foreign relations doctrines, like the act of state doctrine[93] and the political question doctrine.[94] Under those doctrines, judges’ protestations about their incompetence in foreign affairs had led not to greater deference to the political branches, but to the growth of judicial power as judges too readily decided not to decide cases.[95] Similarly here, broadly phrased concerns about “unreasonable interference with the sovereign authority of other nations”[96] could encourage judges to back too quickly away from cases that Congress (and those other nations) would really rather they keep.

Rather, the inquiry should still be tied to methods of statutory interpretation: When the case for the exercise of U.S. prescriptive jurisdiction under international law becomes attenuated, then judges should look more searchingly for clues that Congress did, indeed, mean to legislate that far.[97] That is, functional concerns justify the inquiry but do not themselves resolve it. And if congressional intent for a statute to apply in a particular context is clear, functional concerns should not override that application.[98] In addition, this inquiry should be treated as a rare exception to an otherwise strong default. Once a judge has determined that Congress intended a statute to apply extraterritorially, she should assume it does apply extraterritorially, at least up to the limits of international law. Many cases will fall squarely in this zone, without implicating the gray space at the edges where the thinness of U.S. jurisdictional ties in fact generate friction and controversy.[99]

This is the difference between Morrison, Empagran, and Kiobel v. Royal Dutch Petroleum Co., where foreign allies intervened to voice concerns about the reach of U.S. laws, and RJR Nabisco, where they themselves invoked the U.S. law. The former cases were “foreign-cubed,” involving foreign plaintiffs, foreign defendants, and foreign harms; the reach of U.S. prescriptive jurisdiction under international law in those cases was at its lowest ebb. In RJR Nabisco, in contrast, “[a]ll defendants are U.S. corporations, headquartered in the United States, charged with a pattern of racketeering activity directed and managed from the United States, involving conduct occurring in the United States.”[100] As Justice Ginsburg summed up, “this case has the United States written all over it.”[101] The difference in foreign reaction across these cases was not hypocritical, as the majority delighted in suggesting.[102] Rather, the reason why RJR Nabisco did not raise international comity concerns was not because the plaintiffs were the foreign governments themselves, but because the defendants were U.S. citizens operating on U.S. territory.[103] In such a case involving traditional and strong bases for exercising prescriptive jurisdiction under international law, judges should not shy away from applying extraterritorial laws as Congress has written them.

The precise contours of this task of statutory interpretation in the gray zone of jurisdictional conflict, however, still requires refinement. Courts and commentators have tried and largely discarded the Restatement (Third) of Foreign Relations Law’s reasonableness balancing test, and Empagran has rarely been invoked outside the antitrust context. Besides which, the search for congressional intent regarding jurisdictional conflict might be quixotic, as Congress may well never have thought about the jurisdictional configurations at issue. The solution to that remaining uncertainty, however, is not avoidance, but engagement and reasoned elaboration.[104]


The real challenge in RJR Nabisco was not the hunt for congressional intent—which was not that hard to find—but the fact that RICO itself is overbroad. I am not unsympathetic to the majority’s concern that extraterritorial application of RICO, just like territorial application of RICO, could sweep too broadly. But unilateral judicial corrections for unwise legislation raises more concerns than it resolves. What Congress needs, if not a faithful agent, is a faithful partner in managing jurisdictional conflict in a globalized economy. It may now fall to the lower courts to step into that partnership and nudge the presumption back towards home.


[1]136 S. Ct. 2090 (2016).

[2]18 U.S.C. §§ 1961–68 (2012).

[3]RJR Nabisco, 136 S. Ct. at 2111.

[4]Brief for Respondents at 6, RJR Nabisco, 136 S. Ct. 2090 (No. 15-138), 2016 WL 447643, at *6.

[5]Id. at 7–8.

[6]Id. at 9 (internal quotation marks and citation omitted). For example, the European Community alleged that RJR Nabisco was “knowingly sell[ing] their products to organized crime, arrang[ing] for secret payments from organized crime, and launder[ing] such proceeds in the United States or offshore venues known for bank secrecy.” Id. (alterations in original) (citation omitted).

[7]Id. at 9–11.

[8]Equal Emp’t Opportunity Comm’n v. Arabian Am. Oil Co. (Aramco), 499 U.S. 244, 248 (1991) (quoting Foley Bros. v. Filardo, 336 U.S. 281, 285 (1949)).

[9]Justice Sotomayor took no part in the consideration or decision of the case. RJR Nabisco, 136 S. Ct. at 2096. With a seat vacant following the death of Justice Scalia, a Court of seven decided the case.

[10]See infra Section II.A.

[11]See RJR Nabisco, 136 S. Ct. at 2090, 2101, 2106, 2108; see also infra Section II.B.

[12]In addition to RJR Nabisco, see, for example, Morrison v. National Australia Bank Ltd., 561 U.S. 247 (2010), and Kiobel v. Royal Dutch Petroleum Co., 133 S. Ct. 1659 (2013).

[13]The Court has recognized that some statutes, like the Sherman Act, 15 U.S.C. §§ 1–7, and the Clayton Act, 15 U.S.C. §§ 12–27, apply extraterritorially. See, e.g., F. Hoffmann-La Roche Ltd. v. Empagran S.A., 542 U.S. 155, 173 (2004) (Sherman Act); Pfizer Inc. v. Gov’t of India, 434 U.S. 308, 312 (1978) (Sherman and Clayton Acts); id. at 320 (Burger, C.J., dissenting) (Sherman and Clayton Acts). On other occasions, when the Court has interpreted statutes not to apply extraterritorially, Congress has amended those statutes to make its extraterritorial intent clear. See, e.g., Zachary D. Clopton, Replacing the Presumption Against Extraterritoriality, 94 B.U. L. Rev. 1, 13–15 (2014) (discussing Congress’s response to Aramco and Morrison, among other examples).

[14]The closest direction might be Empagran, discussed below in Section III.B, but the discussion in Empagran was not explicitly tied to the presumption against extraterritoriality.

[15]See infra Section III.A.

[16]See infra Section III.B.

[17]Equal Emp’t Opportunity Comm’n v. Arabian Am. Oil Co. (Aramco), 499 U.S. 244, 248 (1991).

[18]See William S. Dodge, The Presumption Against Extraterritoriality in Two Steps, 110 Am. J. Int’l L. Unbound 45, 45 n.1 (2016).

[19]See Restatement (Fourth) of Foreign Relations Law 27 (Am. Law Inst., Tentative Draft No. 2, 2016); Restatement (Third) of Foreign Relations Law 27 (Am. Law Inst. 1987).

[20]In addition to Morrison v. National Australia Bank Ltd., 561 U.S. 247 (2010), Kiobel v. Royal Dutch Petroleum Co., 133 S. Ct. 1659 (2013), and Aramco, 499 U.S. 244, these cases include Microsoft Corp. v. AT&T Corp., 550 U.S. 437, 454–55 (2007), and Smith v. United States, 507 U.S. 197, 203–04 (1993), among others. See also Pamela K. Bookman, Litigation Isolationism, 67 Stan. L. Rev. 1081, 1098–99 (2015) (framing the renewed focus on extraterritoriality as a transnational litigation avoidance strategy). For a more sympathetic view of the Court’s efforts to rein in transnational litigation, see Paul B. Stephan, Response Essay – Empagran: Empire Building or Judicial Modesty?, in International Law in the U.S. Supreme Court: Continuity and Change 553, 553 (David L. Sloss et al. eds., 2011); Paul B. Stephan, Private Litigation as a Foreign Relations Problem, 110 Am. J. Int’l L. Unbound 40 (2016) [hereinafter Stephan, Private Litigation as a Foreign Relations Problem].

[21]499 U.S. 244 (1991).

[22]For a thorough account of the presumption’s history and this modern turn towards a stricter doctrine, see John H. Knox, A Presumption Against Extrajurisdictionality, 104 Am. J. Int’l L. 351, 361–76 (2010).

[2]Aramco, 499 U.S. at 248 (emphasis added) (internal quotation marks omitted) (quoting Foley Bros. v. Filardo, 336 U.S. 281, 285 (1949)).

[24]Id. (first alteration in original) (emphasis added) (quoting Foley Bros., 336 U.S. at 285).

[25]Id. (emphasis added) (quoting Benz v. Compania Naviera Hidalgo, S.A., 353 U. S. 138, 147 (1957)).

[26]Murray v. Schooner Charming Betsy, 6 U.S. (2 Cranch) 64 (1804).

[27]See Knox, supra note 22, at 365–66, 374–75. In fact, Benz (the case quoted by Aramco for this proposition) did not entail the extraterritorial application of a statute.

[28]See RJR Nabisco, 136 S. Ct. at 2102; Morrison v. Nat’l Austl. Bank Ltd., 561 U.S. 247, 265 (2010).

[29]Smith v. United States, 507 U.S. 197, 204 (1993).

[30]Morrison, 561 U.S. at 255.

[31]RJR Nabisco, 136 S. Ct. at 2100.

[32]For further discussion of this two-step framework, see Dodge, supra note 18.

[33]RJR Nabisco, 136 S. Ct. at 2101.


[35]Morrison, 561 U.S. at 266.

[36]See Lea Brilmayer, New Extraterritoriality: Morrison v. National Australia Bank, Legislative Supremacy, and the Presumption Against Extraterritorial Application of American Law, 40 Sw. U. L. Rev. 655, 664 (2011) (“[W]hile citing the principle of legislative supremacy, Justice Scalia’s opinion [in Morrison] has actually increased the opportunity for judicial policy making and diminished the importance of congressional preferences.”).

[37]See id. at 667–68.

[38]See RJR Nabisco, 136 S. Ct. at 2101 (noting that Morrison and Kiobel v. Royal Dutch Petroleum Co. left this question unaddressed).


[40]18 U.S.C. § 1961 (2012).

[41]Some of these cross-referenced statutes, for example, explicitly reach conduct that occurred “outside the United States.” RJR Nabisco, 136 S. Ct. at 2101–02 (internal quotation marks omitted) (quoting 18 U.S.C. § 1203(b) (2012) (hostage taking) and 18 U.S.C. § 1957(d)(2) (2012) (money laundering)).

[42]See id. at 2097 (summarizing § 1962).

[43]Id. at 2102.

[44]For my thoughts on what such inquiry might look like, see infra Part III.

[45]See RJR Nabisco, 136 S. Ct. at 2101, 2106. Though this requirement was initially stated in the portion of the opinion joined by the dissenters, it is not clear that the dissenters fully embraced it. See id. at 2113 n.2 (Ginsburg, J., dissenting). For a critique of this new requirement, see infra Section II.B.

[46]The majority did not explain how RICO’s criminal provision, § 1963, or the civil remedies available to the government, § 1964(a) and (b), would rebut the presumption, though it seemed to assume they would. See Anthony J. Colangelo, The Frankenstein’s Monster of Extraterritoriality Law, 110 Am. J. Int’l L. Unbound 51, 55 (2016).

[47]RJR Nabisco, 136 S. Ct. at 2106.


[49]Cf. Colangelo, supra note 46, at 51, 55 (raising similar concerns).

[50]RJR Nabisco, 136 S. Ct. at 2101.

[51]Id. at 2102–03.

[52]See id. at 2113 (Ginsburg, J., dissenting) (critiquing the majority’s reasoning for this inconsistency); see also Pamela K. Bookman, Doubling Down on Litigation Isolationism, 110 Am. J. Int’l L. Unbound 57, 58 (2016) (same); Colangelo, supra note 46, at 54 (same).

[53]18 U.S.C. § 1964(c) (2012) (emphasis added).

[54]RJR Nabisco, 136 S. Ct. at 2108.

[55]See Pfizer Inc. v. Gov’t of India, 434 U.S. 308, 313–20 (1978).

[56]RJR Nabisco, 136 S. Ct. at 2114 (Ginsburg, J., dissenting) (second and third alterations in original) (quoting Northcross v. Bd. of Educ. 412 U.S. 427, 428 (1973)).

[57]Id. at 2109–11 (majority opinion).

[58]Justice Ginsburg identified three other occasions on which the Court had interpreted RICO’s § 1964(c) to align with § 4 of the Clayton Act. See id. at 2114 (Ginsburg, J., dissenting). As a further irony, the majority’s focus on the definition of “person” suggests that § 4 of the Clayton Act extends extraterritorially because it incorporates another extraterritorial provision (the definition of “person”), even though the majority had just rejected a similar interpretive move for § 1964(c).

[59] 18 U.S.C. 1964(c) (2012).

[60]RJR Nabisco, 136 S. Ct. at 2108–09 (majority opinion).

[61]Id. at 2101, 2106.

[62]As Professor Pam Bookman puts it, “[i]t is . . . hard to argue that the presumption tracks congressional intent when it keeps raising the hurdle that Congress must clear in order to rebut it.” Bookman, supra note 52, at 61.

[63]See, e.g., Abbe R. Gluck & Lisa Schultz Bressman, Statutory Interpretation from the Inside—An Empirical Study of Congressional Drafting, Delegation, and the Canons: Part I, 65 Stan. L. Rev. 901, 912 (2013).

[64]See id. at 945.

[65]561 U.S. 247 (2010).

[66]See Stephan, Private Litigation as a Foreign Relations Problem, supra note 20, at 42 n.9 (citing SEC v. Chi. Convention Ctr., LLC, 961 F. Supp. 2d 905, 909–17 (N.D. Ill. 2013)).

[67]William S. Dodge, The Presumption Against Extraterritoriality Still Does Not Apply to Jurisdictional Statutes, Opinio Juris (July 1, 2016), [].

[68]See Bookman, supra note 52, at 59; Hannah L. Buxbaum, The Scope and Limitations of the Presumption Against Extraterritoriality, 110 Am. J. Int’l L. Unbound 62, 64 (2016).

[69]For a similar view, see Colangelo, supra note 46, at 55 (“[T]he canon has taken on a life of its own, and now seems simply to run roughshod over anything that stands in the way of its myopic quest to quash the private right of action in transnational cases.”).

[70]See generally Richard M. Re, Narrowing Supreme Court Precedent from Below, 104 Geo. L.J. 921, 925–27 (2016) (arguing that lower courts can legitimately narrow Supreme Court precedent through reasonable application of its directives).

[71]See RJR Nabisco, 136 S. Ct. at 2107 (“[W]here such a risk is evident, the need to enforce the presumption is at its apex.”).

[72]See id. at 2106–08.

[73]See infra Section III.B.

[74]See, e.g., Aramco, 499 U.S. at 248.

[75]See RJR Nabisco, 136 S. Ct. at 2115–16 (Ginsburg, J., dissenting); see also, e.g., Cassandra Burke Robertson, Foreign Plaintiffs and the Presumption Against Extraterritoriality, PrawfsBlawg (June 20, 2016), 2016/06/foreign-plaintiffs-and-the-presumption-against-extraterritoriality.html [‌DY] (raising a similar point in response to RJR Nabisco); Ralf Michaels, Main Essay—Empagran’s Empire: International Law and Statutory Interpretation in the U.S. Supreme Court of the Twenty-First Century, in International Law in the U.S. Supreme Court: Continuity and Change, supra note 20, at 533, 544 (raising the concern more generally).

[76]RJR Nabisco, 136 S. Ct. at 2115 (Ginsburg, J., dissenting).

[77]See id. at 2101 (majority opinion) (noting that Morrison and Kiobel did not address this question).

[78]550 U.S. 437, 455–56 (2007).

[79]542 U.S. 155 (2004).

[80]See, e.g., RJR Nabisco, 136 S. Ct. at 2101 (“The scope of an extraterritorial statute thus turns on the limits Congress has (or has not) imposed on the statute’s foreign application . . . .”).


[82]See Hartford Fire Ins. Co. v. California, 509 U.S. 764, 814–15 (1993) (Scalia, J., dissenting); cf. Clopton, supra note 13, at 22–29 (arguing that the presumption against extraterritoriality should be replaced in civil litigation with an emphasis on the Charming Betsy canon).

[83]For a standard account, see Restatement (Third) of Foreign Relations Law § 402 (Am. Law Inst. 1987).

[84]Cf. Buxbaum, supra note 68, at 65 (noting the relevance of other comity doctrines).

[85]For discussions of these doctrinal developments, see Bookman, supra note 20; Stephen B. Burbank, International Civil Litigation in U.S. Courts: Becoming a Paper Tiger?, 33 U. Pa. J. Int’l L. 663 (2012); Maggie Gardner, Retiring Forum Non Conveniens, 92 N.Y.U. L. Rev. (forthcoming 2017).

[86]RJR Nabisco, 136 S. Ct. at 2115 (Ginsburg, J., dissenting). A clearer doctrine for evaluating the import of parallel foreign litigation would help as well. See Gardner, supra note 85. Justice Ginsburg also invoked in her dissent the doctrine of forum non conveniens, RJR Nabisco, 136 S. Ct. at 2115 (Ginsburg, J., dissenting), but as I argue elsewhere, that doctrine is outdated, unhelpful, and only obfuscates the comity analysis, see Gardner, supra note 85.

[87]The following approach has much in common with that proposed by the draft Restatement (Fourth) of Foreign Relations Law. See Restatement (Fourth) of Foreign Relations Law § 204 (Am. Law Inst., Tentative Draft No. 2, 2016) (proposing that “U.S. courts may interpret federal statutory provisions to include other limitations on their applicability as a matter of prescriptive comity” in order “[t]o avoid unreasonable interference with the legitimate sovereign authority of other states”).

[88]For a discussion of these controversial margins, see Maggie Gardner, Channeling Unilateralism, 56 Harv. Int’l L.J. 297, 303–06 (2015).

[89]See Empagran, 542 U.S. at 159–60. For a critique of the Court’s characterization of the dispute in Empagran, see Michaels, supra note 75, at 539–40.

[90]Empagran, 542 U.S. at 164.

[91]See id.; see also Hartford Fire Ins. Co. v. California, 509 U.S. 764, 814–18 (1992) (Scalia, J., dissenting) (similarly groping).

[92]See William S. Dodge, Response Essay – Loose Canons: International Law and Statutory Interpretation in the Twenty-First Century, in International Law in the U.S. Supreme Court: Continuity and Change, supra note 20, at 547, 549 n.18 (noting that Empagran rejected case-by-case balancing); see also Michaels, supra note 75, at 535 (critiquing Empagran for nonetheless replacing international law concerns about actual conflicts with international relations concerns about potential conflicts). This was the problem with the Restatement (Third) of Foreign Relations Law’s reasonableness inquiry, see Restatement (Third) of Foreign Relations Law § 403 (Am. Law Inst. 1987).

[93]See W.S. Kirkpatrick & Co. v. Envtl. Tectonics Corp., Int’l, 493 U.S. 400, 409 (1990) (“The act of state doctrine does not establish an exception for cases and controversies that may embarrass foreign governments, but merely requires that, in the process of deciding, the acts of foreign sovereigns taken within their own jurisdictions shall be deemed valid.”).

[94]See Zivotofsky v. Clinton, 132 S. Ct. 1421, 1427 (2012) (holding a dispute was not a political question without applying the functional factors listed in Baker v. Carr, 369 U.S. 186, 217 (1962), which include concerns about “expressing lack of the respect due coordinate branches of government” and “the potentiality of embarrassment” from many voices addressing one question); see also id. at 1432 (Sotomayor, J., concurring in part and concurring in the judgment) (noting the omission of these more prudential concerns from the traditional political question analysis).

[95]See Jack L. Goldsmith, The New Formalism in United States Foreign Relations Law, 70 U. Colo. L. Rev. 1395, 1396 (1999); Tara Leigh Grove, The Lost History of the Political Question Doctrine, 90 N.Y.U. L. Rev. 1908 (2015); cf. Harlan Grant Cohen, Formalism and Distrust: Foreign Affairs Law in the Roberts Court, 83 Geo. Wash. L. Rev. 380, 436 (2015) (describing the Court’s apparent worry that “[f]oreign affairs functionalism . . . like the keys to the family car or no curfew, might just be too much of a temptation”).

[96]Empagran, 542 U.S. at 164.

[97]Cf. Knox, supra note 22 (suggesting a similar approach pre-Morrison, though Knox would interweave this inquiry with the presumption against extraterritoriality, a route Morrison may not have left open).

[98]See Restatement (Fourth) of Foreign Relations Law § 204 cmt. c, at 36 (Am. Law Inst., Tentative Draft No. 2, 2016)).

[99]Further, the identification of that gray space can be aided by the interventions of the U.S. government and foreign governments.

[100]RJR Nabisco, 136 S. Ct. at 2114 (Ginsburg, J., dissenting).

[101]Id. at 2115.

[102]See id. at 2107–08 (majority opinion).

[103]See Bookman, supra note 52, at 60–61 (raising a similar observation).

[104]For example, in the context of specific statutes like the Lanham Act and the Bankruptcy Code that do apply extraterritorially, lower courts have developed “a variety of tests” to limit those laws’ geographic reach.  Restatement (Fourth) of Foreign Relations Law at 38–40 (Am. Law Inst., Tentative Draft No. 2, 2016)) (gathering cases).

What’s Wrong with Sentencing Equality? Sentencing Legality: A Response to Professors Bierschbach & Bibas

In 2005, I was a public defender in Bronx County, New York. Contemplating a transition to academia, I developed an idea for an article about plea-bargaining and innocence.[1] Early on, I came across a tremendously helpful paper, written by Professor Stephanos Bibas.[2] Several months later, I began a teaching fellowship. On the first day, I was pleased to find Bibas’s name on the office door next to mine. Unfortunately, Bibas had already left for another institution. Our paths seemed destined not to cross. Still, I took a chance and emailed him. Bibas responded with warm words and constructive advice. Over the next decade, our relationship would become one of the most valuable of my professional career. His generosity is unparalleled, and my scholarship is demonstrably better for it. Sometimes we disagree, though perhaps less so recently (which only speaks to the great influence he has had on my thinking).

Coincidentally, I knew Professor Rick Bierschbach even before I left criminal practice. We were acquaintances—at opposite ends of a large circle of thirty-something lawyers. Rick’s reputation preceded him. He was (and is) a mensch.[3] And, true to his kind nature, he has always been available to lend an ear and to offer useful feedback.

What is the point of these brief testimonials? It is, of course, a law review convention to begin a response with kind words for the article’s authors. But that is not my principal aim. By this genuine and personal expression of affection for two profoundly decent individuals, I hope to show the power of narrative—the capacity for detail to reach comparatively more than form. The entrenched form is merely to celebrate the authors’ professional qualifications and achievements. But my narrative aspires to reach something deeper and richer. The evaluation of an academic’s worth (or lack thereof) entails much more than a recitation of her accomplishments. Similarly, the evaluation of an offender’s blameworthiness (or lack thereof) entails much more than legal and factual guilt. No single set of criteria—promulgated ex ante—is competent to tell the complete story in all its intricacies.

Bierschbach and Bibas understand this, of course. Indeed, it is a central premise of their remarkable article, What’s Wrong with Sentencing Equality? They explain that positive sentencing law has unduly prioritized sentencing “math” over other relevant (indeed, potentially more relevant) moral and prudential considerations.[4] Mandatory rules operate to sort offenders into predetermined boxes and types, typically defined by criminal records and crimes of conviction.[5] Like outcomes are thereafter imposed for each offender of every broad type.

The authors trace the source of the prevailing approach to the equality principle—or, rather, to our dominant conception of it. But I am not so sure. The first-order question is why our criminal justice system has settled upon such a formalistic conception of equality. The unanswered question is what makes sentencing math so attractive, as compared to some alternative qualitative approach to equality that might accommodate more detail. The answer to that question lies with another contested principle—the legality principle, which Professor Herbert Packer famously termed “the first principle of criminal law.”[6] The root of what is wrong with sentencing equality arises from our positive conception of this first principle, not from our positive conception of equality itself. Our fetish for formal legality is what drives our commitment to formal equality.[7] But equality qua equality is tangential, at best.

I. Three Conceptions of Equality

Bierschbach and Bibas recognize that there may be more than one viable conception of equality[8] They distinguish between our positive (and problematic) substantive conception, which aims to guarantee equal results, defined formally; and a procedural conception, which aims to guarantee equal opportunities to argue for defendant-favorable results.[9] They use the descriptor “outcomes-oriented” to describe the prevailing substantive approach, and they discuss its underappreciated tradeoffs.[10] Likewise, they defend alternative procedural approaches to equality (even random processes, like lotteries and dice rolls) as consistent with what John Rawls called “pure procedural justice.”[11]

Nevertheless, they fail to appreciate that there are, in fact, two very different strands of “outcomes-oriented” equality. There is a formal strand and an equitable strand. Pursuant to the equitable strand, there is no necessary tradeoff between individualization and equality. To the contrary, individualization is the means by which equitable equality is achieved. As I have explained elsewhere:

A justice system that admits equitable considerations is premised on the fact that legally identical cases should sometimes be handled differently for normative reasons. This does not mean, however, that equitable [variation] deviates unduly from a defensible notion of equality. . . . [A] contextualized approach to criminal justice necessarily demands more than just a rigid application of legal rules pursuant to formal designations. It demands an evaluation of relative blameworthiness to ensure that equitably distinct cases are recognized as such, even if those cases happen to be legally identical under insufficiently discriminating statutes.[12]

With respect to both the formal and equitable strands of equality, case outcomes provide the relevant reference points. In this way, both approaches remain substantive. The difference is only whether these outcomes are determined to be equitably or formally distinct or alike.

The reason for the misconception—for describing equitable equality as procedural equality—is the pivotal role that narrative plays in “equitable judgment.”[13] Narrative is, of course, a procedural endeavor. But the practice is only a means to the decisive end—a means “to look into things more deeply, to see whether we may have missed some unusual impediment that deformed the process of character formation.”[14] It is only once we have attended to the complete “narrative history” that we can determine whether a given penalty really fits the particular crime—or whether the prescribed sentence, instead, has failed to account for some “unusual hardship or inequality.”[15] On this reading, the stories we tell shape the sentences we impose. We contrast one story with the next to realize whether we have adequately grasped the differences between them.[16]

Thus, there are (at least[17]) three conceptions of equality: a procedural conception that promises like opportunities to argue; a substantive equitable conception that promises like normative results; and a substantive formal conception that promises like legalistic results. Results matter only with respect to the two substantive conceptions. But each substantive conception entails a radically different method by which to discover and ultimately compare blameworthiness.

The preceding is, to some degree, no more than a small taxonomical quibble. But I think it necessary to define our terms correctly in order to discern properly why the criminal justice system is so allergic to equitable equality. Bierschbach and Bibas do not make clear enough that the problem with positive sentencing law is not its focus on outcomes, but rather its fixation with law—a fixation that has produced results that are more obviously ordered than equal.

II. What’s Exceptional with Criminal Justice?

Sentencing was once different. Sentencing law “traditionally permitted the story of the defendant’s character-formation to come before the judge or jury in all its narrative complexity . . . .”[18] Over time, however, determinate sentencing regimes have reduced or eliminated the judge’s opportunities for “sympathetic assessment” and “merciful mitigation”—and, for that matter, for penalty enhancements for particularly bad actors and heinous acts.[19] Ironically, our most severe punishment—the death penalty—describes the one constitutional context in which the practice of narrative has continued to hold sway.[20]

What changed? Bierschbach and Bibas take as given the conventional wisdom that mandatory sentencing regimes developed as compromises between progressives (intent on reining in racial, ethnic, and class discrimination) and conservatives (intent on reining in lenient judges).[21] But the conventional wisdom is incomplete. It provides only an explanation for why both sides prioritized equality over other principles and values, but it does not account for why they settled on a formal conception of substantive equality. The answer to that question depends on an older trend.

Going back to the Enlightenment, political theorists have championed the legality principle as an “important prophylaxis against the arbitrary and abusive exercise of discretion in the enforcement of the penal law.”[22] The classical liberal view is that well-defined rules are the best means to achieve legality’s objectives—that, to the extent possible, the terms of criminal culpability and punishment must remain prospective and precise.[23] And that view gained greater currency in response to the atrocities committed by last century’s totalitarian powers.[24] Thus, even as the rest of the law witnessed a “revolt against formalism,” the law of crime—including sentencing law—grew more rule-bound.[25] This is the idea behind “the rule of law as a law of rules”—an idea grounded in legal formalism.[26] And it is this same impulse that also informs our formal conception of equality and, by extension, our rule-bound sentencing law.

The authors seem genuinely curious as to why we speak pejoratively about sentencing “disparities,” while we have elsewhere defended variability as the acceptable (or even virtuous) byproduct of “localism,” “pluralism,” or “laboratories of democracy.”[27] But there is no mystery. The “law of crime” is thought exceptional precisely because criminal justice is exceptionally harsh and stigmatic.[28] The coercion of conviction and sentence carries with it a corresponding “especial need for certainty,” which is considered essential to prevent liberal punishment from slipping into rank oppression.[29] Ultimately, then, it is our prevailing notion of the rule of law that is doing the bulk of the work. Indeed, Professor Peter Westen has observed that there is always some equality-independent principle—some alternative “moral standard”—that is doing the bulk of the work:

Equality is an empty vessel with no substantive moral content of its own. Without moral standards, equality remains meaningless, a formula that can have nothing to say about how we should act. . . . 

              . . . . 

Relationships of equality (and inequality) are derivative, secondary relationships; they are logically posterior, not anterior . . . . To say that two persons are the same in a certain respect is to presuppose . . . a prescribed standard for treating them . . . . Before such a rule is established, no standard of comparison exists.”[30]

Bierschbach and Bibas commit a category error. They mistake the triumph of a formal conception of legality with the triumph of a formal conception of equality. In fact, our obsession is not with equal outcomes as much as highly predictable and ordered outcomes. This is what the authors do not quite grasp. Consider this observation: “The stale sentencing debate[] of . . . rules versus standards needs to stop treating equality as if it were a single concept.”[31] But the rules-standards debate is neither stale nor peripheral. To the contrary, our false impression that equality is a single concept is a direct byproduct of our fidelity to rules. Equality comes in different shapes and sizes, but the dominant conception of legality is built to perceive just one—a breed of equality born of rules. When the authors celebrate a “more elastic approach[]” to sentencing equality, they are only pushing a “more elastic approach[]” to legality.[32] The equality question is a mere echo of the legality debate.

III. What’s Right (and Natural) About Sentencing Equity?

In the space provided, I cannot possibly defend thoroughly the ambitious claim that an “elastic approach” to legality is nonetheless consistent with the rule of law. Elsewhere, I do more to support this bold proposition.[33] I have argued even that a softer conception of legality might provide better protection against rough punishment, at least in some contexts.[34] To be sure, there are limits to any workable and defensible equitable approach, as even committed moral particularists have recognized.[35] No system is competent to attend to every relevant detail.[36] Legal standards set the outer boundaries. And, because resources are finite, sentencing proceedings can accommodate only so much scrutiny. But even within these practical parameters, an evaluative system necessarily has the capacity to perceive more than a mechanistic system. The process is imperfect, but not obviously arbitrary.[37]

The misapprehension—that equitable evaluation is incompatible with legality—is based upon the tendency of particularistic methodologies to reveal unwelcome disparities. But there is a difference in kind between creating a disparity and exposing what was always there. Formal legality paves over incongruence; its methods are mechanistic, facile, and somewhat fictive. Equitable legality engages incongruence; its methods are evaluative, complex, and relatively honest. Formal legality promotes a conception of equality that is predictable but thin. Equitable legality promotes a conception of equality that is indeterminate but thick. When it comes to the equitable approach, what we construe to be cacophony may just be consistency by another name—succinctly, individualization in the service of a thoroughgoing qualitative comparison. According to Professors Martha Nussbaum and Dan Kahan, “It’s when the law falsely denies its evaluative underpinnings that it is most likely to be incoherent and inconsistent; it is when the law refuses to take responsibility for its most contentious choices that its decision makers are spared the need to be principled . . . .”[38]

There is, after all, nothing inherently equality-enhancing about a rule that provides: sell X grams of heroin; receive Z years in prison. To the contrary, commentators have long observed that “sentencing math” promotes inconsistency by failing to account meaningfully for the offender’s genuine role in the offense.[39] Little fish are treated like big fish, and big fish trade information for undeserved cooperation pleas.[40] One response is that prosecutors retain the charging and bargaining discretion not only to use over-inclusive sentencing rules as threats, but also to correct for these rules’ overreach. Thus, they may choose not to treat little fish like big fish, even if the law ostensibly commands that they do so. Put differently, they may pursue individualized “substantive justice,” even (or especially) within mandatory regimes.[41]

But this prospect is hardly comforting. As Bierschbach and Bibas recognize, even if prosecutors “have the perspective and power to balance individual blameworthiness against systemic demands,” they are also subject to “incentives to clear cases quickly,” as well as other institutional and cognitive biases that may undercut their willingness (or even their ability) to exercise equitable discretion consistently, fairly, and effectively.[42] The problem is not only that the prosecutor is a professional, but also that she is partial. Here, the authors generously reference my scholarship to support the proposition that prosecutorial “decisions often turn on legalistic habits of charging and plea bargaining.”[43] But I am equally troubled (if not more so) by the manner in which prosecutors may indulge their extra-legalistic habits—their “‘nonlegal impetus’” to pursue their own vested interests and their own idiosyncratic notions of moral or prudential blameworthiness.[44]

If nothing else, the scope of prosecutorial power reveals a profound truth about almost any purportedly mandatory rule. It is destined, by some degree, to fail. The discretion we stamp out at one stage reappears at another. The authority we strip from one stakeholder works its way to another.[45] Like water through a weak dike, discretion finds the cracks—and there are always cracks. With this in mind, it makes little sense to construct a sentencing system that serves to delegate equitable authority to the least transparent and most biased parties—specifically, the prosecutors who control the pivotal decisions over whether to file mandatory charges (and whether to negotiate around them, thereafter).

I do not mean to suggest that the authors fail to appreciate this concern. To the contrary, they note, “[s]ome of the blameworthiness factors . . . . inform low-visibility but influential decisions by . . . prosecutors . . . to decline or divert charges, to plea bargain, and to strike cooperation deals, among other things. But the hydraulic pressures to dispose of cases quickly make these decisions invisible, unchecked, unaccountable, and highly variable.”[46] Yet the inevitability of discretion reveals something more profound still. Discretion is natural, whereas rule-bound reasoning is artificial. At best, rule-bound reasoning is infantile and small-minded.[47] The mature mind strives to understand things more deeply.[48] The criminal justice system is a human system, and no human system is prepared to dispense with context wholesale. This, then, seems to be what Bierschbach and Bibas mean when they write that “laymen care about a good deal more” than rigid sentencing rules.[49] Laymen also care about motive, social circumstance, and character (and also, for that matter, innumerable other moral and prudential questions and considerations).

Don’t get me wrong. As between a system stripped free of equity and a system that assigns equitable discretion exclusively to the executive, I prefer the latter—but only because one is impossible and even less desirable than the other. But neither is all that attractive. One of Bibas’s most important contributions is his brilliant book, The Machinery of Criminal Justice.[50] He hit upon certain fundamental realities—that the “machinery of criminal justice” is the province of neither sovereign prerogative nor the unbending rule.[51] In truth, liberal criminal justice is not machinery at all.


A mandatory sentence is like a store-bought greeting card. It strives to express a moral sentiment. But, except by rough fit, it cannot manage a genuine connection. It is just an abstract product of what typical people typically feel about typical groups. Preset categories are all that describe who should be considered similarly situated to whom. The core problem, however, is not that the mass-produced sentence (or, for that matter, the mass-produced greeting card) says the same things to everyone, but rather that it says empty things to everyone. The mandatory sentence is shallow.

Bierschbach and Bibas still need to identify the source of equality’s shallowness, as it applies to positive law. The authors’ point of attack is a particular approach to the principle. But by giving such primacy to the dominant conception of equality, they unintentionally buy into it. The real problem is legality, as conventionally formulated and expressed.


[1]Josh Bowers, Punishing the Innocent, 156 U. Pa. L. Rev. 1117, 1117 (2008).

[2]Stephanos Bibas, Bringing Moral Values into a Flawed Plea-Bargaining System, 88 Cornell L. Rev. 1425 (2003).

[3]“Mensch” is the Yiddish word for a good person. See Mensch,, [].

[4]Richard A. Bierschbach & Stephanos Bibas, What’s Wrong with Sentencing Equality? 102 Va. L. Rev. 1447, 1455, 1465 (2016) (“Where moral disagreement was endemic, math supplanted morality.”).

[5]Josh Bowers, Legal Guilt, Normative Innocence, and the Equitable Decision Not to Prosecute, 110 Colum. L. Rev. 1655, 1688–92 (2010) [hereinafter Bowers, Legal Guilt]; see also Wayne A. Logan, Proportionality and Punishment: Imposing Life Without Parole on Juveniles, 33 Wake Forest L. Rev. 681, 703 (1998) (“Crime is now typically conceived solely in terms of the relative seriousness of a given offense, to the exclusion of offender culpability.”).

[6]Herbert L. Packer, The Limits of the Criminal Sanction 79–80 (1968).

[7]On our overly formal conception of the legality principle, see Josh Bowers, Probable Cause, Constitutional Reasonableness, and the Unrecognized Point of a “Pointless Indignity”, 66 Stan. L. Rev. 987, 996–98 (2014) [hereinafter Bowers, Pointless Indignity]; Josh Bowers, Legality & Rough Justice 6–7 (May 30, 2014) (unpublished manuscript) (on file with author); Josh Bowers, Understanding the Police 1 (Oct. 5, 2016) [hereinafter Bowers, Understanding the Police] (unpublished manuscript) (on file with author).

[8]Bierschbach & Bibas, supra note 4, at 1492 (observing that “many alternative conceptions of sentencing equality” exist); cf. Peter Westen, The Empty Idea of Equality, 95 Harv. L. Rev. 537, 537 (1982) (tracing the endurance of the principle to its ability to shift shape).

[9]Bierschbach & Bibas, supra note 4, at 1447.

[10]Id. at 1456–57 (“[O]ur main goal is to show how sentencing equality, as it has come to be conventionally understood in outcomes-oriented terms, interacts with the institutional structure and goals of punishment, and how exposing that interaction complicates the tradeoffs that inhere in sentencing design.”).

[11]John Rawls, A Theory of Justice 75 (rev. ed. 1999) (defining “pure procedural justice” as a “fair procedure” that produces a result that is “likewise correct or fair, whatever it is, provided that the procedure has been properly followed”); see Bowers, Legal Guilt, supra note 5, at 1677 (“[T]here is no persuasive reason why equal treatment must be measured according to substantive outcomes only.”); Vincent Chiao, Ex Ante Fairness in Criminal Law and Procedure, 15 New Crim. L. Rev. 277 (2012).

[12]Bowers, Legal Guilt, supra note 5, at 1673–74; Logan, supra note 5, at 703 n.108 (“All defendants are not alike, just as all crimes, even if given the same label, are not identical.” (internal quotation marks and citation omitted)).

[13]Martha C. Nussbaum, Equity and Mercy, 22 Phil. & Pub. Aff. 83, 85–86, 92 (1993) (defining “equitable judgment” as “judgment that attends to the particulars,” and as “a gentle art of particular perception, a temper of mind that refuses to demand retribution without understanding the whole story”). My conception of equitable judgment is consistent with what Aristotle called epieikeia or “fair-mindedness.” Lawrence B. Solum, Virtue Jurisprudence: A Virtue-Centered Theory of Judging, 34 Metaphilosophy 178, 205 (2003).

[14]Dan M. Kahan & Martha C. Nussbaum, Two Conceptions of Emotion in Criminal Law, 96 Colum. L. Rev. 269, 368–70 (1996) (comparing mechanistic and evaluative approaches).

[15]Id. In any event, even our conventional legalistic conception of equality depends upon procedural methods. Here, the means consist of the conventional (and relatively technical) deconstructive craft of legal analysis, as opposed to the constructive craft of narrative. Bowers, Legal Guilt, supra note 5, at 1690–91 (describing what it means to think and reason like a lawyer).

[16]Packer, supra note 6, at 88 (“It is not enough to say: this man goes to jail because he did something bad. There is obligation to relate the particular bad thing that this man did to other bad things that have been created as criminal in the past.”).

[17]Cf. infra note 30 and accompanying text (discussing Peter Westen’s view that equality is secondary to—and defined by—other enumerable moral principles).

[18]Kahan & Nussbaum, supra note 14, at 367; see also Bierschbach & Bibas, supra note 4, at 1473 (“[T]he criminal justice system once did and could again make a point of promoting remorse, apology, forgiveness, and reconciliation . . . . about treating victims and offenders with dignity and respect . . . . But these considerations . . . . require context-specific judgments of real human beings . . . .” (footnote omitted)).

[19]Kahan & Nussbaum, supra note 14, at 367.

[20]Enmund v. Florida, 458 U.S. 782, 801 (1982) (“[P]unishment must be tailored to . . . personal responsibility and moral guilt.”); Lockett v. Ohio, 438 U.S. 586, 604 (1978) (holding that the capital sentencing jury is entitled to consider “as a mitigating factor, any aspect of a defendant’s character or record and any of the circumstances of the offense that the defendant proffers as a basis for a sentence less than death”); see Josh Bowers, Mandatory Life and the Death of Equitable Discretion, in Life Without Parole: America’s New Death Penalty? 25, 25 (Charles J. Ogletree Jr. & Austin Sarat eds., 2012) [hereinafter Bowers, Mandatory Life].

[21]Bierschbach & Bibas, supra note 4, at 1459; see Michael Tonry, Sentencing Matters 6, 9, 147 (1996). I do not reject the conventional story entirely. Indeed, I have even articulated it previously. Josh Bowers, Contraindicated Drug Courts, 55 UCLA L. Rev. 783, 825 (2008) (“Stakeholders of varied political stripes came together to counteract what some saw as racist inequities in sentencing and what others saw as overly lenient discretionary sentencing.”); Bowers, Mandatory Life, supra note 20, at 30 (“Left-liberals saw determinate sentencing to be an antidote to racial and economic inequalities in discretionary sentencing. Law-and-order conservatives saw determinate sentencing to be an antidote to lenient liberal judges.” (footnote omitted)).

[22]Richard J. Bonnie et al., Criminal Law 81 (3d ed. 2010); John Calvin Jeffries, Jr., Legality, Vagueness, and the Construction of Penal Statutes, 71 Va. L. Rev. 189, 201, 212 (1985) (“The rule of law signifies the constraint of arbitrariness in the exercise of government power.”).

[23]Christine Sypnowich, Utopia and the Rule of Law in Recrafting the Rule of Law: The Limits of Legal Order 178, 179–80 (David Dyzenhaus ed., 1999) (“[T]he rule of law . . . refer[s] to the idea that law should meet certain procedural requirements so that the individual is enabled to obey it. . . . [It must] be relatively certain, clearly expressed, open, . . . adequately publicised . . . . [and] prospective . . . . The practical effect . . . is to set limits to the discretion of legislators, administrators, judges and the police.”).

[24]Bonnie et al., supra note 22, at 83 (“Would a Puritan theocracy or an Islamic state or a Marxist dictatorship have a comparable commitment to protecting . . . . the principle of legality as a fundamental ideal of the penal law dictated by liberal democracy and its underlying assumptions about the relation of the state to individual citizens?”); cf. Packer, supra note 6, at 86–87 (describing development of the legality principle and concluding that “after centuries of retrospective law-making by judges, . . . . the process of judicial law-making in the criminal field has . . . come to a halt” (emphasis omitted)).

[25]Louis Michael Seidman, Points of Intersection: Discontinuities at the Junction of Criminal Law and the Regulatory State, 7 J. Contemp. Legal Issues 97, 98, 101, 103 (1996) (“[A]lthough realism’s lessons for criminal law seem obvious, formalism continues to dominate criminal jurisprudence.”).

[26]Jeffries, supra note 22, at 212 (describing the “quite conventional” prevailing conception of the rule of law and the principle of legality); Antonin Scalia, Essay, The Rule of Law as a Law of Rules, 56 U. Chi. L. Rev. 1175 (1989). According to Professor John Jeffries: “[T]he agencies of official coercion should, to the extent feasible, be guided by rules” as a means to promote “regularity and evenhandedness in the administration of justice and accountability in the use of government power.” Jeffries, supra note 22, at 201, 212 (explaining that “appeals to the ‘rule of law,’” as they apply to the penal law tend to entail “the resort to legal formalism as a constraint against unbridled discretion”); Bowers, Pointless Indignity, supra note 7, at 989–98 (examining and critiquing the prevailing perspective); Bowers, Understanding the Police, supra note 7, at 1 (same).

[27]Bierschbach & Bibas, supra note 4, at 1450–51, 1489 (“One might even argue that the arguments and observations of Gerken, Leib, and Schragger should have special purchase at sentencing, with its lack of easy policy answers, difficult moral tradeoffs, and inextricable connection to community norms.”); see also id. at 1487 (“The assumption in all of this is that punishment should not turn on local views.”); id at 1490–91, 1495 (arguing that “normative variation” may be “a virtue, not a vice”).

[28]Egon Bittner, The Police on Skid-Row: A Study of Peace Keeping, 32 Am. Sociological Rev. 699, 700 (1967) (“[C]rime belongs wholly to the law, and its treatment is exhaustively based on considerations of legality . . . .”); Seidman, supra note 25, at 97.

[29]Stephen R. Perry, Judicial Obligation, Precedent, and the Common Law, 7 Oxford J. Legal Stud. 215, 256 (1987); see also H.L.A. Hart, Legal Responsibility and Excuses, in Punishment and Responsibility 28, 44–47 (1968) (comparing certainty in criminal punishment to certainty in private law); Kenneth I. Winston, On Treating Like Cases Alike, 62 Calif. L. Rev. 1, 37 (1974) (discussing criminal law’s long tradition of “strict adherence to rules”); cf. Sherry F. Colb, Freedom from Incarceration: Why is This Right Different from All Other Rights?, 69 N.Y.U. L. Rev. 781, 821 (1994) (explaining that criminal justice is different in kind from other forms of legal regulation and that “liberty from confinement cannot be relegated to the status of unprotected aspects of daily life”).

[30]Westen, supra note 8, at 545, 547–48 (footnotes omitted).

[31]Bierschbach & Bibas, supra note 4, at 1520 (emphasis added).

[32]Id at 1514.

[33]Bowers, Pointless Indignity, supra note 7, at 1030–43; Bowers, Understanding the Police, supra note 7, at 1.

[34]Specifically, I claim that equitable oversight is critical to regulating appropriately the enforcement and adjudication of low-level crimes. Bowers, Pointless Indignity, supra note 7, at 1036–37; Bowers, Understanding the Police, supra note 7, at 20.

[35]Bowers, Legal Guilt, supra note 5, at 1670 (citing sources).

[36]Id. at 1670–72; Nussbaum, supra note 13, at 93 (“[T]he ‘matter of the practical’ can be grasped only crudely by rules given in advance, and adequately only by a flexible judgment suited to the complexities of the case.”); Solum, supra note 13, at 206 (“[T]he infinite variety and complexity of particular fact situations outruns our capacity to formulate general rules.”).

[37]Kenneth Culp Davis, Discretionary Justice: A Preliminary Inquiry 21 (1969) (“[T]he conception of equity that discretion is needed as an escape from rigid rules [is] a far cry from the proposition that where law ends tyranny begins.”).

[38]Kahan & Nussbaum, supra note 14, at 274, 373–74 (noting that evaluations of normative blameworthiness “are better because they are brutally and uncompromisingly honest,” whereas “[m]echanistic doctrines . . . tend to disguise contentious moral issues”); William J. Stuntz, Unequal Justice, 121 Harv. L. Rev. 1969, 2039 (2008) (“[W]hen prosecutors have enormous discretionary power, giving other decisionmakers discretion promotes consistency, not arbitrariness. . . . [I]nstitutional competition curbs excess and abuse.”). In this vein, Professor Bill Stuntz argued that even localism is compatible with equality. Id. at 1995, 2031–33 (noting that “equality and local democracy [may] go hand in hand”).

[39]United States v. Justice, 877 F.2d 664, 666 (8th Cir. 1989) (describing federal sentencing as a “mechanical process”); Douglas A. Berman, The Virtues of Offense/Offender Distinctions, in Criminal Law Conversations 611, 615 (Paul H. Robins, Stephen P. Garvey & Kimberly Kessler Ferzan eds., 2009) (criticizing federal sentencing guidelines as “a sentencing process that [has] been drained of its humanity”); Bowers, Mandatory Life, supra note 20, at 29 (“[T]he vagaries of life outstrip ‘sentencing math’ that tends to count most that which can be counted most easily—like drug weight and monetary loss, as opposed to moral blameworthiness.” (emphasis omitted) (citing Berman, supra, at 615)); Jack B. Weinstein, Comment, A Trial Judge’s Second Impression of the Federal Sentencing Guidelines, 66 S. Cal. L. Rev. 357, 364 (1992).

[40]Daniel C. Richman, Cooperating Defendants: The Costs and Benefits of Purchasing Information from Scoundrels, 8 Fed. Sentencing Rep. 292, 292 (1996). The authors understand this, of course. They emphasize “more granular” approaches that turn on “less-quantifiable values” than determinate facts and figures, like drug weight and monetary gain. Bierschbach & Bibas, supra note 4, at 1455.

[41]Malcolm M. Feeley, The Process Is the Punishment: Handling Cases in a Lower Criminal Court xix (1979); Bowers, Legal Guilt, supra note 5, at 1708. It is well understood that determinate sentencing empowers prosecutors. Rachel E. Barkow, Institutional Design and Policing of Prosecutors: Lessons from Administrative Law, 61 Stan. L. Rev. 869, 877 (2009) (“With the prevalence of mandatory minimum laws, a prosecutor’s decision to bring or not bring charges can dictate whether a defendant receives a mandatory five-, ten-, or twenty-year term, or whether he or she is sentenced far below that floor.”); David Bjerk, Making the Crime Fit the Penalty: The Role of Prosecutorial Discretion Under Mandatory Minimum Sentencing, 48 J.L. & Econ. 591, 593–95 (2005); William J. Stuntz, Plea Bargaining and Criminal Law’s Disappearing Shadow, 117 Harv. L. Rev. 2548, 2564 (2004).

[42]Bierschbach & Bibas, supra note 4, at 1482; see also Bowers, Legal Guilt, supra note 5, at 1687 (“[P]rosecutors possess the human capacity for practical reason. But, in their professional roles, they are first and foremost legally trained institutional actors. And their position and profession may profoundly limit the degree to which they are willing and able to exercise equitable discretion—particularly in the petty cases where such discretion is most warranted.”); Bowers, Mandatory Life, supra note 20, at 36 (observing that a “‘mechanistic, impersonal, lawyerized criminal justice’ may interfere with what some have identified as an intuitive ‘deep human need’ to humanize and particularize retributive questions” (quoting Stephanos Bibas, Forgiveness in Criminal Procedure, 4 Ohio St. J. Crim. L. 347, 348 (2007))).

[43]Bierschbach & Bibas, supra note 4, at 1482 (emphasis added) (citing Bowers, Legal Guilt, supra note 5, at 1701–02).

[44]Bowers, Understanding the Police, supra note 7, at 22 (quoting Frederick Schauer, Analogy in the Supreme Court, Lozman v. City of Riviera Beach, Florida, 2013 Sup. Ct. Rev. 405, 429 (defining a “nonlegal impetus” as, inter alia, an “idiosyncratic reaction to . . . the very particular facts of the case”)).

[45]Bierschbach & Bibas, supra note 4, at 1470 (observing that individualization that is “omitted” from sentencing “show[s] up elsewhere in the system”); Bowers, Legal Guilt, supra note 5, at 1687 n.146 (“[D]iscretion is a hydraulic force. An effort to eradicate it may play out like an attempt to squeeze air out of a partially inflated balloon: What disappears from one spot pops up in another.”); Debra Livingston, Police Discretion and the Quality of Life in Public Places: Courts, Communities, and the New Policing, 97 Colum. L. Rev. 551, 593 (1997) (“Limiting the discretion that police exercise on the street simply by demanding specificity in the laws that they enforce is so hopeless . . . . ‘Elimination of discretion at one choice point merely causes the discretion that had been exercised there to migrate elsewhere in the system.’” (quoting Jerry L. Mashaw, Prodelegation: Why Administrators Should Make Political Decisions, 1 J.L. Econ. & Org. 81, 97 (1985))).

[46]Bierschbach & Bibas, supra note 4, at 1470 (footnote omitted); see also Stephanos Bibas, Essay, Transparency and Participation in Criminal Procedure, 81 N.Y.U. L. Rev. 911, 914, 931 (2006) (“On average . . . [professional] insiders are more concerned with and informed about practical constraints . . . . [Lay] [o]utsiders, knowing and caring less about practical obstacles and insiders’ interests, focus on . . . offenders’ just desserts. . . .” They “care about a much wider array of justice concerns than do lawyers, including . . . blameworthiness, and apologies.”).

[47]Jerome Frank, Law and the Modern Mind 178 (Transaction Publishers ed. 2009) (“The constant effort to achieve a stable equilibrium . . . is regressive, infantile, and immature.”); Bowers, Legal Guilt, supra note 5, at 1690–91.

[48]Nussbaum, supra note 13, at 94 (“[T]he equitable person is characterized by a sympathetic understanding of ‘human things.’”); Kahan & Nussbaum, supra note 14, at 287 (observing that an Aristotelian conception of appropriate conduct in a particular context requires “asking what a person of practical wisdom would do and feel in the situation,” not by asking mechanistically what the law commands).

[[49]Bierschbach & Bibas, supra note 4, at 1473; David Garland, Punishment and Modern Society: A Study in Social Theory 1 (1990) (noting that punishment falls short of societal expectations because “we have tried to convert a deeply social issue into a technical task for specialist institutions”).

[50]Stephanos Bibas, The Machinery of Criminal Justice (2012).

[51]Bierschbach & Bibas, supra note 4, at 1483–84 (“No one institutional player should hold all the cards. . . . An outcomes-focused conception of equality bent on centralizing sentencing and reducing discretion is in tension with this checks-and-balances approach.” (footnote omitted)).