Note from the Editor: 

The Federalist Society takes no positions on particular legal and public policy matters. Any expressions of opinion are those of the author. Whenever we publish an article that advocates for a particular position, as here, we offer other perspectives on the issue, including ones opposed to the position taken in the article. This article is a response to an article by Dan Morenoff, which you can find here; there is also a brief reply at the end of this rebuttal. We also invite responses from our readers. To join the debate, please email us at


Dan Morenoff’s Escaping the Goldilocks Problem: How States Can Avoid Redistricting Litigation identifies and explains a significant problem: Modern redistricting invariably results in costly and uncertain litigation.[1] This problem is created by two seemingly contradictory doctrines. Fourteenth Amendment Equal Protection Clause jurisprudence generally forbids mapmakers from predominately considering race when drawing legislative districts, while the Voting Rights Act requires detailed racial considerations. To be sure, there are porridges that are “just right” and avoid violating both doctrines; presumably districts drawn with predominate racial considerations but only to comply with the Voting Rights Act satisfy strict scrutiny.[2] But to get to that conclusion, the porridge must be tested. And because the incongruous commands of the 14th Amendment and the Voting Rights Act require legislation to sit on the head of a pin, a dissatisfied voter-plaintiff’s civil rights complaint is always ready-made.

Large volume redistricting litigation is a problem, and Mr. Morenoff is correct that the commonly proposed reforms[3] will not meaningfully reduce the likelihood of litigation that entangles even the best-intentioned maps. But his proposed solution of using multimember statewide districts would not alleviate this problem. Moreover, his proportional representation solution would undermine the values of district-based representation—values that are due for a defense. A better solution to reduce litigation and protect district-based representation values is far more elegant though possibly just as controversial: get the courts out of the political thicket of districting litigation except in cases where there is discriminatory intent.

I. Gerrymandering Litigation Is Uniquely Problematic Because It Undermines the Institutional Capital of Courts and the Integrity of the Legislative Process

Litigation is how we sort out and protect constitutional and statutory rights.[4] All litigation is subject to criticism on the ground that it is too costly, and much of it is problematic because court decisions produce costly uncertainty. So why should we be specially concerned about people petitioning courts for a vindication of rights in the context of redistricting litigation? Because there are unique facets to redistricting litigation that go beyond both the cost-objection that inures to all litigation and the uncertainty objection that attaches to all totality-of-(often confounding)-circumstances jurisprudence (like Section 2 doctrine). These unique facets undermine both the judicial and legislative branches for multiple reasons; I highlight one reason for each branch here.

A. Judicial Branch Integrity

Invariably, redistricting litigation enmeshes courts in political disputes. As the Supreme Court observed in Gaffney v. Cummings, “Politics and political considerations are inseparable from districting and apportionment . . . . The reality is that districting inevitably has and is intended to have substantial political consequences.”[5] And it is not just courts that are caught up in redistricting litigation; it is the Supreme Court. This is because the grant or denial of an injunction relating to legislative districts is directly appealable to the Supreme Court.[6] As a result, the Supreme Court is asked to decide numerous politically charged cases every redistricting cycle.[7]

These are not simply cases with policy implications furthering or frustrating a particular party’s platform. These are cases affecting legislative organization and the substantive membership of legislative bodies. While this concern is most acute in partisan gerrymandering cases,[8] it is also present in apportionment[9] and VRA Section 2 cases.[10] Indeed, Section 2 cases are premised on the understanding that one kind of district constituency with an opportunity to elect one kind of preferred candidate is valid while another is not.[11] Not all candidates are the same, even within parties. Different district constituencies will produce substantively or descriptively different types of Democrats and substantively or descriptively different types of Republicans.[12]

Districting decisions have direct political implications that shape not just whether a Democrat or Republican is more likely to be elected, but what kind of Democrat or Republican will be elected, and even what those parties will look like.[13] Whether it affirms or invalidates maps, the Supreme Court’s decisions will be controversial, and thus all districting litigation requires the expenditure of political capital that can undermine the Court’s institutional legitimacy.[14] In short, the current volume of inexorably political litigation undermines the public’s perception of the judiciary as a neutral and non-political institution.

B. Legislative Branch Integrity

Redistricting litigation undermines the legislative branch because it imposes unique burdens on legislators and introduces external factors into their deliberative process by subjecting them to litigation from which they are usually immune. Typically, legislators are shielded from judicial inquiry into their legislative activities, either as an application of the Speech and Debate Clause (for members of Congress),[15] or as an application of the federal common law of legislative immunity and privilege.[16] These mechanisms protect legislators (and their aides) in their exercise of any core legislative activity, not just what they say on the floor.[17] Legislative privilege extends to those activities that are “necessary to prevent indirect impairment [of legislative] deliberations.”[18] And while the set of constitutive elements comprising core legislative activities may be open to some debate,[19] drafting legislation like redistricting laws is indisputably the core of the core of legislative activities.

The doctrines of legislative immunity and privilege are indispensable to proper democratic functioning. Compelling legislators to participate in a “private civil action . . . creates a distraction and forces [legislators] to divert their time, energy, and attention from their legislative tasks to defend litigation.”[20] This can “delay and disrupt the legislative function.”[21] Separation of powers is another concern. The “central purpose” of the protections for legislators against liability and judicial inquiry into the legislative process is to “avoid intrusion by the Executive or Judiciary into the affairs of a coequal branch,” “protect legislative independence,” and thus “‘preserve the constitutional structure of separate, coequal, and independent branches of government.’”[22]

Nevertheless, many district courts have “qualified” (a euphemism for eliminated) the legislative privilege in numerous redistricting cases.[23] The result is that legislators and their staffs have been compelled to produce testimony, documents, or both. Courts abrogating the privilege have typically reasoned that, because legislative privilege is not absolute, legislative testimony of intent is the best evidence of legislative intent, and civil rights actions are very important, it is appropriate to deviate from the norm of legislative privilege.[24]

That reasoning is dubious. First, the premise is overstated. To be sure, the Supreme Court has held that legislative privilege is not absolute in criminal proceedings.[25] But the Court has never held that a legislator may be compelled to testify in a civil action. The Court has speculated that in “extraordinary instances” legislators might be called “to testify concerning the purpose of official action, although even then such testimony frequently will be barred by the privilege.”[26] Second, the abrogating courts’ substantive logic is flawed. An individual legislator’s intent is not the same as the intent of the legislature as a body.[27] Moreover, the Supreme Court has already, in Tenney, asked and answered the question of whether Congress intended to abrogate legislative privilege in civil rights cases, saying it did not.[28]

The ease with which these courts have abrogated the privilege is mystifying.[29] Were legislative bodies, in the exercise of their legislative subpoena power, to compel judges and justices to testify about their case deliberations and individual motivations for judicial decisions, the interference with the judicial function would be obvious. Redistricting cases involve an interference with legislative branch deliberations and operations that is unlike any other kind of civil litigation. This anomaly alone should cause us special concern about the volume of redistricting litigation.

So Mr. Morenoff is right. Voluminous redistricting litigation is a unique problem that threatens the judicial and legislative branches. But his strategy for avoiding this damaging litigation will not work.

II. A Statewide Multimember District Would Not Reduce Litigation

Absent from Mr. Morenoff’s otherwise accurate description of current racial gerrymandering and Voting Rights Act jurisprudence is a history of racial gerrymandering cases. This history shows that Mr. Morenoff’s proposed solution of electing all legislators in single statewide at-large districts would not free legislatures from litigation.

In fact, suspicion of multimember districts is what drove the development of racial gerrymandering jurisprudence in the first place. In the 1960s, the Supreme Court recognized multimember districts may “operate to minimize or cancel out the voting strength of racial or political elements of the voting population.”[30] Far from alleviating litigation risks, multimember districts invited litigation because “the invidious effect” of canceling out the voting strength of racial or political elements of the voting population “can be more easily shown” in large, multimember districts that lack residential requirements for candidates.[31]

In the 1960s, the idea that multimember districts could support a race-based equal protection claim was largely theoretical. The Court recognized that multimember districts could be used to dilute the minority vote, but multimember districts were not per se unconstitutional and the Court regularly upheld the validity of multimember districts against racial gerrymandering challenges.[32] But in 1973’s White v. Regester, the Supreme Court struck down two Texas multimember legislative districts on the principle that members of political minorities “had less opportunity than did other residents of the district to participate in the political processes and to elect legislators of their choice.”[33] In these multimember districts, each primary candidate was selected by a majority of the multimember district voters. Such an arrangement turned what would be minority-majority constituencies in single-member districts into powerless minority-minority constituencies in the multi-member district.[34] In the phraseology the Whitcomb, decided two years earlier, Texas had created multimember districts that “submerge[d] minorities.”[35] Other Supreme Court cases followed White in striking down multimember districts.[36]

In 1980, the Supreme Court noted equal protection challenges to “at-large electoral schemes” had “been advanced in numerous cases before this Court[,] . . . most often with regard to multimember constituencies within a state legislative apportionment system.”[37] In that case, City of Mobile v. Bolden, the Supreme Court reversed the lower courts’ decisions that Mobile’s decades-old at-large election system for local legislators violated equal protection, and the plurality famously held that race dilution claims—like other equal protection claims—required a showing of discriminatory effect and discriminatory purpose.[38]

Section 2 of the Voting Rights Act was enacted “largely [as] a response” to City of Mobile v. Bolden.[39] Section 2 adopts as its relevant legal standard the Court’s “results” test applied in White v. Regester,[40] but eliminates any need to demonstrate a discriminatory purpose.[41] As pre-Bolden constitutional racial vote dilution challenges were typically aimed at multimember districts, so too was the first Section 2 challenge considered by the Supreme Court.[42] What had been a constitutional equal protection claim simply became a statutory claim with one less element to prove.[43] And if a constitutional vote dilution claim would have succeeded under the constitutional jurisprudence that Section 2 incorporated (which was already suspicious of multimember districts), surely it would succeed under Section 2.

Against this history of skepticism about the disproportionately negative effects multimember districts can have on minority representation, Mr. Morenoff doubles down. He proposes that states should adopt a single statewide district—a mega-multimember district—where voters choose political parties, not specific candidates. In this scheme, representatives would be selected by the parties in numbers corresponding with the statewide legislative vote. In a nutshell, he proposes proportional representation.

Mr. Morenoff asserts these statewide party-based elections are impervious to Section 2 Voting Rights Act challenges because “it would be impossible for a local majority to block any local minority’s ability to elect its preferred candidate.” Thus, he concludes no plaintiff could survive the preliminary stage of the Gingles analysis, and that Section 2 litigation would therefore be cut off at the outset.[44]

But Mr. Morenoff’s shorthand description of Gingles’ preliminary requirements, which puts load-bearing weight on term “local,” is not accurate. Gingles asks whether a minority group “is sufficiently large and geographically compact to constitute a majority in a single-member district,” whether that majority-minority hypothetical single-member district constituency is “politically cohesive,” and whether the “majority votes sufficiently as a bloc . . . usually to defeat the protected group’s preferred candidate.[45] Gingles does not require that the blocking be done by a “local majority,” but instead the majority as constituted in the district created by the law being challenged. Gingles is simply a judicial test for assessing whether the minority vote is being submerged.[46] Those factors seem to apply in any statewide scenario. As Judge Frank Easterbrook commented when he was sitting on a district court panel in a 2002 Wisconsin redistricting impasse case, “at-large election[s] of the entire Assembly . . . would likely violate the Voting Rights Act.”[47]

And in many states, a mega-district would have all of the demographic attributes necessary for a majority or plurality to submerge the representative interests of protected classes. Let’s keep with Wisconsin to illustrate. Wisconsin is a swing state, having in the past decade elected both Democrats and Republicans in each of the state’s most significant statewide elections: President, U.S. Senator, Governor, and state Attorney General.[48] Let’s stipulate it is comprised of an equal number of Democrat and Republican voters. According to the 2010 U.S. Census, approximately 7% of residents reported as “Black or African American” “alone or in combination with one or more other races.” Hispanics comprise 6% of the population.[49] Some members of these groups exhibit residential and voting patterns that satisfy Gingles’ preliminary test.[50]

If we presume that there is no demographic difference between those who vote and those who are counted in the census, and if we presume every black or Hispanic voter is a Democrat—two counterfactuals that surely overstate the percentage of Democratic votes that come from these groups[51]—then neither group makes up greater than 14% of the Democratic electorate. To be sure, the members of these groups might be present in the “party lists” offered up in European- or Israeli-style legislative elections,[52] even in demographically proportional numbers. But it is not self-evident that this would be so. Discrimination (purposeful or not—the VRA requires only disparate impact) might very well exist within the party list selection process.[53] And anytime the statewide-elected legislature or congressional delegation is demographically different than the population as a whole, a VRA plaintiff should be able to craft a pleading that survives a dismissal motion.

It is not enough to respond, as Mr. Morenoff does,[54] that the discrimination would not be the result of state action. The state action is the adoption of the statewide redistricting plan,[55] and underlying facts outside of the state action always contribute to VRA analysis. Courts consider facts ranging from the political cohesion of a minority group (a preliminary Gingles inquiry) to any number of factors that make up the “totality of circumstances” analysis that comprises the second part of the Gingles test. One of these listed in the Senate Committee on Judiciary Report accompanying the legislation is “the exclusion of members of the minority group from [the] candidate slating process.”[56]

That leaves Mr. Morenoff to rest his argument on this assertion: “where every community receives proportional representation, no community could claim to have been denied the same opportunity to elect its candidates afforded any other.”[57] But candidates aren’t elected in a proportional representation system, parties are. Whether a community is afforded the same opportunity to elect its candidates as another pushes the VRA question into a judicial inquiry into the operations of political parties: How is the party list selected and ordered? Answering this and related questions (e.g., why is a candidate favored by minority groups so low on the list?) would involve substantial judicial inquiry into the operations of political associations and may prove extremely disruptive to political participation. This may be problematic from a First Amendment perspective; at the very least, it creates tensions with First Amendment principles.

Mr. Morenoff might reply that in a proportional rep-resentation system, we would expect to see third parties flourish. Fair enough, and that may contextually make a VRA claim more difficult to prove. But it will by no means end litigation. We do not know what a VRA analysis would look like in that scenario, but one can easily imagine arguments that there is discrimination if this system produces a need to create “special interest” third parties in order for minority groups to see candidates of their choice in the legislature. Only “majority” parties in this scenario would have the benefit of having the majorities or core pluralities that enable party dominance of legislative organization and leadership that is key to moving bills and setting legislative agendas.

Thus, while Mr. Morenoff is likely correct that a statewide mega-district would avoid Shaw problems (because there are no statutory classifications as everyone is in a single district), it would invite Voting Rights Act litigation in every case in which it is adopted. At least single-member districts today carry the potential of a just-right porridge. Proportional representation morphs the analogy into Scylla and Charybdis, and gives Odysseus no choice but to sail into Scylla.

But if I am wrong that courts would still entertain VRA claims after third parties emerge, then Mr. Morenoff’s proposal’s VRA effectiveness depends on the balkanization of political parties and the emergence of parties designed to chiefly accommodate descriptive racial identities. Some may not see this as problematic, though the Supreme Court has noted that when legislators perceive themselves as just representing particular racial groups, it may “threaten[] to undermine our system of representative democracy. . . .”[58] Even if Mr. Morenoff’s proposal were to solve the litigation problem (which I do not believe it would), the negative consequence of proportional representation to “our system of representative democracy” should be better understood.

III. Proportional Representation Undermines Important Values of Representation

One salient criticism of the Supreme Court’s gerrymandering and apportionment jurisprudence—and more generally all political process jurisprudence—is that while the Court has addressed these cases through the doctrinal lens of equal protection and individual rights (whether constitutional or statutory), its opinions are largely devoid of an overarching political theory of representation.[59] This may be, in part, because the Framers did not adopt a single theory of representation, and therefore countenanced many.[60] Indeed, the Constitution established a bicameral legislature[61] that was substantively designed to curb the legislative power and structurally denies predominance to any single theory of representation. Not only must measures pass both houses before they become law,[62] but the houses were designed to reflect different interests in part based on their modes of election. As James Madison wrote in Federalist 51:

In republican government, the legislative authority necessarily predominates. The remedy for this inconveniency is to divide the legislature into different branches; and to render them, by different modes of election and different principles of action, as little connected with each other as the nature of their common functions and their common dependence on the society will admit.[63]

Without a clear historical marker for what is the proper translation of the people’s interests into a republican form of government, the Court’s treatment of this question has been (appropriately, in my view) to simply put up markers for what the Constitution does not compel. For example, proportional representation is not required by the Constitution because, among other reasons, it is fundamentally inconsistent with the concept of winner-take-all elections and multimember bodies comprised of separately elected individuals.[64] Nor are competitive districts constitutionally compelled.[65]

Every mode or manner of choosing legislators will endorse different underlying representational values. A legislature comprised of the winners of winner-take-all single-member elections in equipopulous and geographically contiguous districts (today’s dominant model for state legislatures and exclusive model for Congress) will reflect different representational values than a legislature that is the product of proportional representation derived from statewide general ticket elections. These possibilities are by no means the only ones,[66] but they are the ones to compare when evaluating the effect of Mr. Morenoff’s proposal on values other than litigation-avoidance. And the proposal undermines several current conceptions of representation, three of which are highlighted below.

A. Proportional Representation Denies Individuals a Personal Representative

Among the most troubling aspects of proportional rep-resentation is that it denies citizens a personal representative in the legislative body. It is obvious, if often overlooked, that legislators elected in geographically contiguous districts represent all of their constituents, not just the ones who voted for them. While a losing candidate’s supporters might be “without representation” by their candidate of choice,[67] it “cannot [be] presume[d] . . . the candidate elected will entirely ignore the interests of those voters.”[68] Instead, those voters “have as much opportunity to influence that candidate as other voters in the district.”[69]

This personal representation is about more than substantive influence on policy. A legislator’s job is not just substantive policymaking; “Serving constituents . . . is the everyday business of a legislator.”[70] Indeed, as one district court observed, “[t]he modern role of legislators centers less on the formal aspects of representing—e.g., legislating and policymaking—and more on maintaining the relationship between legislators and their constituents.”[71]

Proportional representation systems in which candidates are selected from a party list after a general ticket election deprive constituents of a single point of contact to influence policy or navigate government bureaucracies. A legislator elected under such a system is not dependent on the votes of any particular category of citizens, and there is thus limited incentive to forge responsive constituent relations.[72] This, in turn, would seem to undermine legislative responsiveness to constituents, a chief tenet of republicanism.

B. Proportional Representation Elevates Party Over People

For similar reasons, a proportional representation system perverts Shaw’s representative ideal that legislators represent a whole constituency and not just a part.[73] In a proportional representation system, a legislator represents the party (and after that its members and supporters), not the polity. A legislator remains or moves up on the party list because of his or her ability to please not constituents, but party leaders. This is one of the principal criticisms of the Knesset, which Mr. Morenoff holds up as a template for his proposal:

Israel is an illuminating (and discouraging) example [of party list voting]: The political parties there have been subject to withering, albeit ineffective, criticism for picking their slates more in response to the imperatives of internal party politics than by consideration of something so abstract as the public good or the capacity for public leadership. It is indeed hard to see how turning over such important decisions [as candidate selection] to a party bureaucracy necessarily maintains the values of a republican government.[74]

I would not assume parties would be wholly unresponsive to the people in the candidate selection process, of course. For example, the DNC’s changes to the power of “superdelegates” was responsive to Senator Bernie Sanders’ supporters’ claim that the party’s presidential nomination was fixed for Secretary Hillary Clinton in 2016.[75] But it took Sanders’ improbably strong primary campaign and Clinton’s improbable general election defeat for the party to make even modest changes to the candidate selection process.

Perhaps more significantly, many political scientists, reformers, and members of the public believe that increased partisan polarization is a problem with modern politics.[76] But Mr. Morenoff’s proposal, which places with party bosses the power of candidate selection and retention, would predictably exacerbate polarized voting in legislative bodies. Gone would be competitive districts, where elected officials must sometimes part ways with party platforms in order to “vote their district.” Proponents of proportional representation might see this as a feature, not a bug, as parties provide clear values for voters to choose. But political parties reflect only one type of representational value: policymaking influenced by political ideology. Citizens have dynamic representational interests that are not always ideological and that might not be captured in party platforms. Enabling those dynamic interests to flourish may be essential to curbing partisan excesses.

C. Proportional Representation Excludes All Representational Interests but One, Increasing Risks of Minority Oppression

What did James Madison mean in Federalist 51 when he observed that “different modes of election and different principles in action” would operate to mitigate potentially oppressive legislative authority? He explains the many ways in which the proposed Constitution’s bicameral legislature would accomplish this end in Federalist 62.[77] Some are dependent on the Senate’s state-equality structure and are not directly applicable to state legislatures. But the goal those mechanisms attempt to reinforce are still worth remembering and incorporating into state representative systems:

It is of great importance in a republic not only to guard the society against the oppression of its rulers, but to guard one part of the society against the injustice of the other part. Different interests necessarily exist in different classes of citizens. If a majority be united by a common interest, the rights of the minority will be insecure. There are but two methods of providing against this evil: the one by creating a will in the community independent of the majority—that is, of the society itself; the other, by comprehending in the society so many separate descriptions of citizens as will render an unjust combination of a majority of the whole very improbable, if not impracticable. . . . The second method will be exemplified in the federal republic of the United States. Whilst all authority in it will be derived from and dependent on the society, the society itself will be broken into so many parts, interests, and classes of citizens, that the rights of individuals, or of the minority, will be in little danger from interested combinations of the majority.[78]

States, of course, do not have a federal character like the United States. Nevertheless, there are distinct political communities within states: counties, cities, towns, and so forth. Reynolds v. Sims, of course, found purely area-based state legislative districting to be unconstitutional,[79] upsetting many state constitutional designs where “representatives were allocated among districts of fixed territory, typically counties and towns.”[80] Yet territorially based representation—contiguity—is still used to define district boundaries. Geographic contiguity, particularly when combined with compactness and some fidelity to municipal boundaries, recognizes that place matters. Places contain communities of interest separate and distinct from partisan ideology. Communities are distinct from one another on multiple levels: political organization (towns, cities, counties), economic character (agricultural, manufacturing, commerce), density (urban, suburban, rural), demographics (age, race), and others. Each community cross-section might be seen as a “different class of citizens” with “different interests.” Just as Madison presumed that senators would balance the interests of their states with national interests, state legislators elected in geographically contiguous districts must balance their district’s unique local interests with state interests.[81]

And those local interests often depart from the party line. In Wisconsin, for example, urban black Democrats have supported a Milwaukee-only school choice program against statewide Democrats,[82] university-town Republicans have voted against labor reforms supported by Republican state leadership,[83] and Democratic representatives have voted against their party to support tax breaks for a local development project.[84]

But in a proportional representation system, there are no countervailing place-informed interests to introduce heterogeneity into parties, and there is no way to reflect representational interests that have both local and state dimensions. Party interests, after all, cross geographic and political boundaries.[85] Without a system that recognizes the significance of place, the examples above likely never occur, and local interests (in the case of the Milwaukee school choice program and the local development project) would be subordinated to state interests. Without the internal party fracturing caused by dyadic concerns, it is far more likely for “an unjust combination of a majority of the whole” to arise.

Short of that, it seems plain that territorially elected legislatures and proportionally elected legislatures will have different focuses, with the former more concerned with local issues and the latter concerned with ideological and statewide issues. “[T]erritorial representation might well provide a kind of institutional formula for promoting governmental minimalism,” while “[p]erhaps it is no coincidence that party-based, proportional systems of representations tend to be found in nations that favor policies associated with the modern welfare state.”[86]

IV. Conclusion

Dan Morenoff’s proportional representation solution to endless litigation over district lines is likely to be both ineffective in its aims and destructive to the traditional construction of representation. A better solution to attack the former and protect the latter is far more elegant though possibly just as controversial: get the courts out of the political thicket of districting litigation except in cases where there is discriminatory intent.

After all, as Chief Justice Roberts memorably said, “The way to stop discrimination on the basis of race is to stop discriminating on the basis of race.”[87] But Section 2 of the Voting Rights Act, when applied to districting, requires mapmakers to do just that. While there is no question the government has a compelling interest in ensuring the right to vote is not denied or abridged on account of race, Shaw and its progeny protect that interest by making it unconstitutional for districting decisions to be predominately motivated by racial considerations. Moreover, it is difficult to see how any law whose compliance requires an imprecise “totality of the circumstances” test and involves meritoriously contentious, highly technical, and uncertain litigation where experts speculate on the political proclivities of racial groups in hypothetical future elections is narrowly tailored towards any ends. While the Supreme Court has assumed that complying with the Voting Rights Act is a compelling state interest, it first ought to address head on the question of whether Section 2 of the Voting Rights Act, when applied to redistricting, passes constitutional muster.



Author’s Reply

I’m grateful for Kevin St. John’s thoughtful response. While I fear Mr. St. John has missed the mark in concluding that a jurisdiction could not avoid redistricting litigation by avoiding redistricting, the first and most important point to emphasize is how broadly we agree on the core issues. We wholly agree:

  1. On the substance of existing doctrine.
  2. That the Court has never addressed whether seeking to comply with the Voting Rights Act may qualify as the kind of “compelling state interest” strict scrutiny requires for a use of race to be constitutional (and that it likely could not).
  3. That existing doctrine poses a Hobson’s Choice between legislatures’ picks of poison. Mr. St. John sees the menu as composed of a Scylla of litigation under Section 2 of the Voting Rights Act (what I, using a Goldilocks analogy, described as a map’s creation being “too cold” in its use of race) and a Charybdis of Shaw-style 14th Amendment claims under the Equal Protection Clause (that I described as a map’s creation being “too hot” in its use of race). I’m actually less sanguine than Mr. St. John that current doctrine “carries the potential of a just-right porridge”—no conceivable “temperate” use of race would spare a jurisdiction litigation in order to find out, ex-post, whether it complied with federal law.
  4. That common voting-rights reforms are red-herrings, which would neither increase the fairness of elections, nor decrease the likelihood of redistricting litigation if implemented.

Still, we have two important disagreements. The first is a “who” question. Mr. St. John concludes that the “better solution to” the dilemma redistrictors face would be to “address head on” the tension between the case law applying the Equal Protection Clause and the Voting Rights Act, even proposing that the best resolution would be to “get the courts out of the ‘political thicket’ of districting litigation except in cases where there is discriminatory intent.” No doubt there are those who sit at the necessary, commanding, Olympian heights (in Congress and the federal courts) who have that option. I don’t doubt that the optimal systematic solution to a conflict of law is to resolve it. Bracketing for another day what resolution would be best, I simply wasn’t addressing such Olympians. I wrote to the state and local legislators whom the gods and federal authorities have placed on the boat with Odysseus and required to act every ten years. They lack the option to “address head on” the conflict by removing one of the threats. Since there is little prospect that those who do have the option will exercise it before the next decennial cycle unfolds, I see value in proposing to such actors a way to limit their time in the dock.

The second goes to whether I’ve identified for legislators a real way out of the crosshairs. In saying “no,” Mr. St. John errs in at least two ways. He conflates dissimilar systems to conclude that existing law dooms the proposal. Then, he dramatically overstates the power of parties to discipline their members in proportional regimes, so generating a false entry in his parade of horribles.

In concluding that existing case law bars proportional representation systems, Mr. St. John relies on cases rejecting at-large elections (which award victory to the prevailing candidate for each seat on a first-past-the-post basis).[88] Although each involves jurisdiction-wide votes, at-large and proportional systems differ in a fundamental way: how they award seats following an election. The courts rejecting at-large systems have done so under Gingles 3, finding a risk of submergence of large, persistent minorities within the electorate—a group with 45% of the population, hypothetically producing 45% of all ballots cast through a bloc-vote, would win 0% of the resulting representation. On the other hand, a proportional system imposes no risk of submergence—the 45% minority casting 45% of hypothetical ballots through a bloc-vote would elect 45% of the resulting officials. Respectfully, the difference vitiates the applicability of the cited cases and leaves no risk of a finding that Gingles 3 has been violated.

Much of Mr. St. John’s analysis of the likely results of a proportional regime (especially the potential losses of centrist elected officials and of official accountability to voters as a result of political parties’ supposedly enhanced powers to force uniformity on members, but also his concerns for enhanced risk of litigation against jurisdictions based on how they allow parties to compile their candidate lists) is both familiar and misguided.[89] While the idea that a proportional system would undermine centrism and accountability finds support in decades-old political-science literature, more recent history has not been kind to those conclusions.

On the greater difficulties for centrists to win election in proportional regimes, the last two decades have seen American political parties, operating in first-past-the-post environments, exhibit greater and greater polarization,[90] giving rise to greater swings in policy at transitions of power;[91] the same period has seen Israeli political parties, operating in a context of proportional representation, converge toward a national consensus on most issues,[92] minimizing potential policy instability. The systems are not having the impact the literature suggests, or perhaps that impact is insufficiently strong to dictate results; either way, events have greatly weakened the deference due the theory.

On accountability, it is not clear either that American incumbents exhibiting politburo-like reelection numbers are accountable to their constituents,[93] or that parties in proportional-representation systems are not,[94] leaving that argument, too, without legs. And the claim that party-power will hold elected officials in line, whatever voters prefer, would surprise: (a) voters in Britain, where last year saw the two historically largest parties suffer mass-defections from their Parliamentary ranks of MPs unwilling to follow leadership’s chosen courses; and (b) those in Israel, where all elections since the State’s founding have seen candidates unhappy with their party leadership go their own way and win seats with new parties (or join parties with different leadership). As the last implies, governmental exposure to suit based in a community’s difficulties founding a new party and running separately are entirely a function of how easy the easy-ballot-access rules adopted for proportional representation are. Only if those rules impose meaningful hurdles that divergently impact minority constituencies would they support a claim that they afforded such groups “less opportunity than . . . other residents . . . to participate in the political processes and to elect legislators of their choice.”[95] That’s not an objection in principal; it’s a drafting guideline to bear in mind while making the move to a proportional system.

As a whole, this exchange strongly suggests that state or local governments could avoid substantive redistricting litigation by avoiding redistricting. It also highlights both that there would be real costs counterbalancing that benefit and that the benefit would remain uncertain until proved up by the Rule 12 motion practice which I contend litigation could not survive. But we’ll only find out who is correct if some intrepid jurisdiction pursues the option before Congress or the courts remake the landscape. I hope one will.



[1]  Dan Morenoff, Escaping the Goldilocks Problem: How States Can Avoid Redistricting Litigation, 21 Federalist Soc’y Rev. 96 (2020), available at


[2]  The Court has not answered directly whether Voting Rights Act compliance satisfies the Equal Protection Clause’s strict scrutiny standard, but it has “assume[d], without deciding, that [a] State’s interest in complying with the Voting Rights Act [is a] compelling” state interest. Bethune-Hill v. Virginia State Bd. of Elections, 137 S. Ct. 788, 801 (2017).


[3]  See Morenoff, supra note 1, at 98-99.


[4]  28 U.S.C. § 1983.


[5]  412 U.S. 735, 753 (1973). See also Vieth v. Jubelirer, 541 U.S. 267, 285 (2004) (plurality op. of Scalia, J.) (stating that districting is “root-and-branch a matter of politics”); Davis v. Bandemer, 478 U.S. 109, 145 (1986) (concurring op. of O’Connor, J.) (“[T]he legislative business of apportionment is fundamentally a political affair[.] . . . To turn these matters over to the federal judiciary is to inject the courts into the most heated partisan issues.”).


[6]  28 U.S.C. § 1253.


[7]  Redistricting happens every ten years following the decennial census. By my count, the Supreme Court has issued 17 opinions involving whether post-2010 Census state legislative or congressional district lines were valid. See Rucho v. Common Cause, 139 S. Ct. 2484 (2019); Lamone v. Benisek (reported with Rucho); Virginia House of Delegates v. Bethune-Hill, 139 S. Ct. 1945 (2019); North Carolina v. Covington, 138 S. Ct. 2548 (2018); Abbott v. Perez, 138 S. Ct. 2305 (2018); Benisek v. Lamone, 138 S. Ct. 1942 (2018); Gill v. Whitford, 138 S. Ct. 1916 (2018); North Carolina v. Covington, 137 S. Ct. 1624 (2017); Cooper v. Harris, 137 S. Ct. 1455 (2017); Bethune-Hill, 137 S. Ct. 788; Whitman v. Personhuballah, 136 S. Ct. 1732 (2016); Harris v. Arizona Independent Redistricting Comm’n, 136 S. Ct. 1301 (2016); Evenwel v. Abbott, 136 S. Ct. 1120 (2016); Shapiro v. McManus, 136 S. Ct. 450 (2016); Arizona State Legislature v. Arizona Independent Redistricting Commission, 135 S. Ct. 2652 (2015); Alabama Legislative Black Caucus v. Alabama, 135 S. Ct. 1257 (2015); Tennant v. Jefferson County Comm’n, 567 U.S. 758 (2012); Perry v. Perez, 565 U.S. 388 (2012). More were decided summarily. See Joshua Leavitt, All About Redistricting, (collecting 2010 cycle redistricting cases, listing Supreme Court dispositions).


[8]  See, e.g., Rucho, 139 S. Ct. at 2498-50, 2458 (holding partisan gerrymandering claims are not justiciable, observing that partisan gerrymandering claims “inevitably ask the court to make their own political judgment about how much representation political parties deserve,” and concluding courts have “no commission to allocate political power and influence in the absence of a constitutional directive. . . .”).


[9]  See, e.g., Gaffney, 412 U.S. at 749-50 (“That the Court was not deterred by the hazards of the political thicket when it undertook to adjudicate the reapportionment cases does not mean that it should become bogged down in a vast, intractable apportionment slough, particularly when there is little, if anything, to be accomplished by doing so.”).


[10]  See Morenoff, supra note 1, at 97.


[11]  52 U.S.C. § 10301(b) (providing voting rights are deemed abridged if it is shown that members of a racial class of citizens “have less opportunity than other members of the electorate to participate in the political process and to elect representatives of their choice”).


[12]  In the lingo of representation, “substantive” relates to policy outlooks and outcomes and “descriptive” relates to characteristics such as race, sex, sexual orientation, or other status. See, e.g., Kenneth Lowande, Melinda Ritchie, & Erinn Lauterbach, Descriptive and Substantive Representation in Congress: Evidence from 80,000 Congressional Inquiries, 63 Am. J. Pol. Sci. 644 (2019).


[13]  For a thoughtful exploration as to why different district lines will yield different constituencies and thus impact the substantive platforms of the candidates who represent those constituencies, see Jacob Eisler, Partisan Gerrymandering and the Illusion of Unfairness, 67 Cath. U. L. Rev. 229, 244-59 (2018). While Eisler’s article concentrates on partisan gerrymandering, there is no reason to believe that the substantive implications of line drawing are confined to the underlying intent of the drafters as opposed to the actual makeup of district constituencies—makeups that are directly or indirectly influenced by litigation.


[14]  This appears to be a central concern to Chief Justice John Roberts in resolving partisan gerrymandering cases. See Gill v. Whitford, No. 16-1611 (S. Ct.), Oral Ar. Tr. at 36-38 (identifying the “main problem” with partisan gerrymandering cases as public perception that the Court is making decisions to favor one party over another). As Professors Gibson and Caldeira have observed, “[t]he driving mechanism for change in institutional support has to do with whether the Supreme Court is seen as an ordinary political institution or whether it is judged to be distinctive. To the extent that people believe the Court is a relatively non-political institution, support for it is more easily generated. Anything that drags the Court into ordinary politics damages the esteem of the institution.” James L. Gibson & Gregory A. Caldeira, Citizens, Courts, and Confirmations: Positivity Theory and the Judgments of the American People 119-20 (2009).


[15]  U.S. Const. art. I, sec. 6, cl. 1 (providing Senators and Representatives “shall not be questioned in any other Place” “for Speech or Debate in either House”).


[16]  Lake County Estates, Inc. v. Tahoe Regional Planning Agency, 440 U.S. 391, 403 (1979); Tenney v. Brandhove, 341 U.S. 367, 376-77 (1951).


[17]  See, e.g., Eastland v. U.S. Servicemen’s Fund, 421 U.S. 491, 505-06 (Speech and Debate protection applies to congressional aide’s issuance of subpoenas as part of congressional committee inquiry); Tenney, 341 U.S. at 376-78 (state legislator’s speech at legislative investigative committee hearing entitled to legislative immunity).


[18]  Gravel v. United States, 408 U.S. 606, 625 (1982).


[19]  See, e.g., Favors v. Cuomo, 11-CV-5632, 2013 WL 11319831, *8-*9 (E.D.N.Y. Feb 8, 2013) (collecting court decisions addressing activities found to be and not to be part of legislative functions).


[20]  Eastland, 421 U.S. at 503.


[21]  Id.


[22]  United States v. Gillock, 445 U.S. 360, 369 (1980); (quoting United States v. Helstoski, 442 U.S. 477, 491 (1979)).


[23]  See, e.g., Whitford v. Gill, 331 F.R.D. 375, 378 (W.D. Wis. 2019) (vacated by Whitford v. Vos, No. 19-2066, 2019 WL 4571109, (7th Cir. July 11, 2019); Benisek v. Lamone, 241 F. Supp. 3d 566, 572–74 (D. Md. 2017); Lee v. Virginia State Bd. of Elections, No. 15-cv-357 (HEH-RCY), 2015 WL 9461505, at *5 (E.D. Va. Dec. 23, 2015); Bethune-Hill v. Virginia State Bd. of Elections, 114 F. Supp. 3d 323, 333 (E.D. Va. 2015); Favors v. Cuomo, 285 F.R.D. 187, 213–14 (E.D.N.Y. 2012); Committee for a Fair & Balanced Map v. Ill. State Bd. of Elections, No. 11 C 5065, 2011 WL 4837508 (N.D. Ill. 2011); Rodriguez v. Pataki, 280 F. Supp. 2d 89, 101 (S.D.N.Y. 2003).


[24]  See, e.g., Whitford, 331 F.R.D. at 378-82; Benisek, 241 F. Supp. 3d at 572-77.


[25]  United States v. Gillock, 445 U.S. 360, 372-73 (1980).


[26]  Village of Arlington Heights v. Metropolitan Housing Development Corp., 429 U.S. 252, 268 (1977) (emphasis added) (citing Tenney and United States v. Nixon, 418 U.S. 683, 705 (1974)).


[27]  See, e.g., United States v. O’Brien, 391 U.S. 367, 383-84 (1968) (“What motivates one legislator to make a speech about a statute is not necessarily what motivates scores of others to enact it.”).


[28]  Tenney, 341 U.S. at 369; see also Gillock, 445 U.S. at 372-73 (explaining Tenney).


[29]  While the trend appears to be that courts will pierce the privilege, see supra note 23, a couple of recent appellate decisions explicitly or implicitly have pushed back against this trend. See Whitford, No. 19-2066 (Munsingwear vacation of order to compel Speaker of Wisconsin Assembly to testify in redistricting case); Lee v. City of Los Angeles, 908 F.3d 1175, 1187 (9th Cir. 2018) (holding municipal legislative officials may not be deposed in municipal redistricting case).


[30]  Fortson v. Dorsey, 379 U.S. 433, 439 (1965).


[31]  Burns v. Richarson, 384 U.S. 73, 88 (1966).


[32]  See, e.g., Whitcomb v. Chavis, 403 U.S. 124 (1971); Fortson, 379 U.S. 433; Burns, 384 U.S. 73.


[33]  412 U.S. 755, 765-71 (1973).


[34]  Id.


[35]  Whitcomb, 403 U.S. at 158-59.


[36]  See, e.g., Rogers v. Lodge, 458 U.S. 613, 622-27 (1982) (affirming district court findings that multimember district resulted in minority exclusion from the political process); East Parish School Bd. v. Marshall, 424 U.S. 636, 639 (1976) (striking down a court-drawn plan that included multimember districts while avoiding a constitutional claim that such districts violated equal protection rights).


[37]  City of Mobile v. Bolden, 446 U.S. 55, 65, 80 (1980) (plurality op.).


[38]  Id. at 66-70 (plurality op.).


[39]  Thornburg v. Gingles, 478 U.S. 30, 35 (1986).


[40]  Id. Compare White, 412 U.S. at 766 (requiring plaintiffs to prove that a minority group’s members “had less opportunity than did other residents in the district to participate in the political processes and to elect legislators of their choice”) with 52 U.S.C. § 13101(b) (providing that a denial or abridgment of the right to vote claim is established where a “totality of the circumstances” shows that a protected class of citizens “have less opportunity than other members of the electorate to participate in the political process and to elect representatives of their choice”).


[41]  Gingles, 478 U.S. at 35.


[42]  Id. (plaintiffs challenged 6 multimember North Carolina general assembly districts).


[43]  While this might have rendered constitutional racial gerrymandering claims unnecessary, in Shaw v. Reno, the Supreme Court more or less dispensed with the “effects” components articulated in White. Rather than having to show that a minority group was frozen out of the political process, the “effect” of a classification is the separation of voters into different districts because of their race, which “reinforces stereotypes and threatens to undermine our system of representative democracy by signaling to elected officials that they represent a particular racial group rather than their constituency as a whole.” Shaw v. Reno, 509 U.S. 630, 649-51 (1993).


[44]  Morenoff, supra note 1, at 100-01.


[45]  Gingles, 478 U.S. at 50-51.


[46]  Whitcomb, 403 U.S. at 158-59.


[47]  Baumgart v. Weidelberg, No. 01-C-0121, 02-C-0366, 2002 WL 34127471, *6 (E.D. Wis. May 30, 2002).


[48]  For statewide election results, see Wisconsin Elections Commission,


[49]  United States Census, Wisconsin (2010), available at


[50]  See generally Baldus v. Members of Wisconsin Government Accountability Bd., 849 F. Supp. 2d 840, 848, 854-58 (E.D. Wis. 2012) (describing African American VRA districts that were not challenged at trial, and holding VRA required legislature to create one majority-minority Latino district).


[51]  Four percent of CNN’s 2016 Wisconsin Presidential exit poll respondents were Latino—far less than the percentage of Hispanic persons in 2010 Census figures—while seven percent were African American. CNN, “exit polls: wisconsin president,” available at The same exit poll reported that, in this close contest, 92% of African Americans voted for Clinton (with an “other/no answer” rate of 2%) and 63% of Latinos voted for Clinton (with an “other/no answer” rate of 3%)—far less than the 100% Democratic Party allegiance assumed for simplicity in our hypothetical. Id.


[52]  See Morenoff, supra note 1, at 100.


[53]  Consider Wisconsin again. Eight of the 36 Democratic members of the Wisconsin State Assembly are African American or Hispanic. See Wisconsin State Legislature, 2019 Wisconsin State Representatives, available at This 2:7 ratio is likely equal to or greater than the percentage of Wisconsin Democrats who are African American or Hispanic—a result likely influenced by VRA-compliant districts. Yet none of these minority representatives are included among the Democratic party’s 6-member legislative officer ranks. See Wisconsin State Legislature, Wisconsin State Assembly, available at Leadership positions are the result of the Assembly Democratic caucus votes, and the caucus presumably includes the same party leaders who would be responsible for developing party lists of representatives to be seated after a general ticket election.


[54]  Morenoff, supra note 1, at 100-01.


[55]  That redistricting legislation qualifies as a “voting . . . practice or procedure” within the meaning of Section 2 of the VRA is certainly contestable, but the Supreme Court has assumed it is as long as it has decided such claims. See Gingles, 478 U.S. at 47.


[56]  S. Rep. No. 97-417, 97th Cong., 2d Sess. (1982), pp. 28-29. Another factor within the state’s control is whether there are “unusually large election districts[.]” Id.

While I suspect that many readers harbor my general skepticism toward the utility of legislative committee reports in the proper interpretation of statutes, this committee report is a part of Supreme Court jurisprudence interpreting Section 2, Gingles, 478 U.S. at 36, 45, and the report draws its factors from prior Supreme Court decisions that Congress designed to incorporate into Section 2. See, e.g., White, 412 U.S. at 767 (citing minority exclusion from candidate selection process to be a factor evincing discriminatory impact of multimember district). Moreover, without these factors, the statutory language would appear to leave ample room for judicial discretion (it employs “totality of circumstances” terminology), and this discretion is sure to be filled in with a sort of jurisprudential common law. This is what courts did in the 1960s and 1970s constitutional gerrymandering decisions, with no other textual hook than the Equal Protection Clause. Absent an about-face on the pre-VRA doctrine that developed to assess racial vote dilution, the Court would surely mine these principles to assess a Section 2 claim.


[57]  Morenoff, supra note 1, at 101.


[58]  Shaw, 509 U.S. at 650.


[59]  See generally James A. Gardner, Partitioning and Rights: The Supreme Court’s Accidental Jurisprudence of Democratic Process, 42 Fla. St. U. L. Rev. 61 (2014).


[60]  The manner of holding elections to choose Representatives was left to state legislatures, subject to Congress’ laws prescribing otherwise. U.S. Const. art. I, § 4, cl. 1. In the first 50 years post-ratification, many states selected congressional delegations in general ticket elections in which the party receiving the plurality of votes would comprise the state’s entire congressional delegation. Rucho, 139 S. Ct. at 2499 (describing practice of many states post-ratification). It was only in 1842 that Congress required single-member geographically contiguous districts. Later statutes required those districts to be compact and equipopulous (though these “traditional” criteria outside of a requirement for single-member congressional districts are no longer codified by federal law). Id.


[61]  See U.S. Const. art. I, §§ 2, 3 (creating House of Representatives and Senate, and requiring bills to pass each house and be signed by the President (or overridden on reconsideration by two-thirds majorities of each house) before they become law).


[62]  See U.S. Const. art. I, § 7.


[63]  The Federalist No. 51, at 322 (James Madison) (Clinton Rossiter ed., 1961).


[64]  Bandemer, 478 U.S. at 130 (plurality op.); Whitcomb, 403 U.S. at 160.


[65]  Id. at 131 (describing Court’s holding in Gaffney as upholding collusively drawn map that tended “to deny safe district [political] minorities any realistic chance to elect their own representatives”).


[66]  For example, representation in the United States Senate is based on static geographic lines surrounding distinct sovereign entities (to the extent not delegated to the United States). Prior to Reynolds v. Sims’ holding in 1964 that state legislative seats must be apportioned on the basis of population, 377 U.S. 533, 568, a majority of states did not require equipopulous districts and recognized some component of area-based apportionment. 377 U.S. at 610-11 (Harlan, J., dissenting). Moving in another direction, one might imagine an electoral system where the districts or candidates must meet certain descriptive qualities, such as race, gender, or occupation. Approximately 50 countries “officially allocate access to political power by gender, ethnicity, or both.” Mala Htun, Is Gender Like Ethnicity? The Political Representation Of Identity Groups, Perspectives of Politics, Vol. 2, No. 3 (American Political Science Association, Sept. 2004).


[67]  Whitcomb, 403 U.S. at 153.


[68]  Bandemer, 478 U.S. at 132 (plurality op.).


[69]  Id. at 131. See also Whitford v. Gill, 218 F. Supp. 3d 837, 954 (W.D. Wis. 2016) (explaining how political minorities influence elected representatives) (Griesbach, J., dissenting), overruled on jurisdictional grounds by Gill, 138 S. Ct. 1916.


[70]  McCormick v. United States, 500 U.S. 257, 272 (1991).


[71]  Gordon v. Griffith, 88 F. Supp.2d 38, 47 (E.D.N.Y. 2000) (attributing increasing significance of legislator-constituent relationship to voter-demand for assistance in navigating modern state bureaucracies) (citing Malcolm E. Jewell, Representation in State Legislatures 10-18 (1982)).


[72]  This could be addressed somewhat by assigning constituent-services responsibilities to representatives or requiring party lists to include representatives from distinct geographic areas. Doing so, however, might reintroduce the VRA problems Mr. Morenoff seeks to avoid and could not fully substitute for the powerful pro-constituent-service incentive structure created by single-member, geographically contiguous districts.


[73]  Shaw, 509 U.S. at 650.


[74]  Sanford Levinson, Gerrymandering and the Brooding Omnipresence of Proportional Representation: Why Won’t It Go Away?, 33 U.C.L.A. L. Rev. 257, 273 (1985).


[75]  See Adam Levy, DNC changes superdelegate rules in presidential nomination process, CNN (Aug. 25, 2018) available at


[76]  See generally Nolan McCarty, Polarization: What Everyone Needs To Know (2019).


[77]  The Federalist No. 62, at 377-78 (James Madison) (Clinton Rossiter ed., 1961).


[78]  The Federalist No. 51, at 323-24 (James Madison) (Clinton Rossiter ed., 1961).


[79]  See supra note 66.


[80]  James A. Gardner, What is “Fair” Partisan Representation, and How Can It Be Constitutionalized? The Case For A Return to Fixed Election Districts, 90 Marq. L. Rev. 555, 560 (2007).


[81]  For a detailed discussion of territories and their interests, see generally James Gardner, Representation Without Party: Lessons From State Constitutional Attempts To Control Gerrymandering, 37 Rutgers L. Rev. 881 (2006).


[82]  See, e.g., Gary C. George and Walter C. Farrel, School Choice and African American Students: A Legislative View, 59 Journal of Negro Education, 521, 521-55 (1990) (legislator-author explains that school choice initiative was “supported sizable segment of Milwaukee’s low-income African-American community,” and legislator worked to enact choice plan that would satisfy local interests while responding to some of the more significant criticisms offered by fellow Democrats).


[83]  Wisconsin State Representative Travis Tranel, whose Mississippi River-bordering district includes UW-Platteville, voted against Act 10, Governor Scott Walker’s signature public sector labor reform bill. In the subsequent election cycle (2012), Tranel outperformed Republican presidential candidate Mitt Romney by 11 points. The Wisconsin Assembly’s roll call vote on Act 10 is available at For discussion of the controversy surrounding the bill, see State v. Fitzgerald, 798 N.W.2d 436, 442-443 (Wis. 2011) (Prosser, J., concurring).


[84]  In 2017, Representative Peter Barca, who had been elected minority leader, was one of four assembly Democrats to vote in favor of a tax break package that aimed to bring FoxConn—and 13,000 promised jobs—to Racine County. See Jason Stein and Patrick Marley, Wisconsin Assembly sends $3 billion Foxconn incentive package to Scott Walker, Milwaukee J. Sentinel (September 14, 2017), available at Barca’s district straddled Racine and neighboring Kenosha County. Two of the other three Democrats voting for the measure were from Racine or Kenosha. For roll call votes on the measure, see (August 17, 2017 Assembly vote sending measure to Senate) and (September 14, 2017 roll call vote concurring in measure as amended by Senate).


[85]  See Gardner, supra note 80, at 573 (“[T]o represent voters by territory is to organize the electorate according to bonds of local community and interest; to represent voters by party, in contrast, is to represent them according to bonds and interests that are found statewide, and that by definition transcend the boundaries of any single district.”).


[86]  Id. at 580-81.


[87]  Parents Involved in Community Schools v. Seattle School Dist. No. 1, 551 U.S. 701, 748 (2007).


[88]  See supra notes 30-43 and accompanying text.


[89]  I readily admit this is not true of all of his analysis. Mr. St. John is correct that a move to proportional representation would prioritize one value (“fairness”) over another (the centrality of locality and geographic community). Similarly, Mr. St. John’s contention that a proportional system could give rise to a balkanization into ethnically-based parties is entirely accurate, although I cynically note that this reality would arise from ethnic groups’ divergent preferences, not from a potential shift to proportional representation. Indeed, the frequency with which jurisdictions defend suits under Shaw and the VRA by arguing that they have engaged solely in legal partisan gerrymandering strongly suggests that we largely already live in the world Mr. St. John fears might emerge from the shift.


[90]  See Cynthia R. Farina, Congressional Polarization: Terminal Constitutional Dysfunction?, 115 Colum. L. Rev. 1689, 1701 (2015) (“[P]olarization has been steadily and consistently increasing since the 1980s.”); Nolan McCarty, Reducing Polarization: Some Facts for Reformers, 2015 U. Chi. Legal F. 243, 249 (2015) (“The current trend towards greater and greater polarization began in the late 1970s and was detectable by academics as early as 1982.”).


[91]  See Nolan McCarty, Polarization, Congressional Dysfunction, and Constitutional Change, 50 Ind. L. Rev. 223, 231 (2016) (“Polarization should simply lead to wider policy swings upon a change in power, not paralysis.”).


[92]  For this counter-intuitive conclusion, see Natan Sachs, The End of Netanyahu’s Unchecked Reign, The Atlantic, Sep. 19, 2019, (“Most Israeli policy would not change with a different prime minister. The basic attitudes of [all the main parties] on Iran, on Hezbollah, on Hamas, on world relations, and even on the prospects of achieving peace with the Palestinians, are all more or less in consensus. . . . [I]n terms of actionable policy, continuity would be the rule.”). For an older analysis reaching the same conclusion as the consensus first emerged into reality, see Barry Rubin, The Region: Israel’s New National Consensus, The Jerusalem Post, Jul. 19, 2009,


[93]  E.g., Corinna Barrett Lain, Judicial Supremacy v. Departmentalism Symposium: Soft Supremacy, 58 Wm. & Mary L. Rev. 1609 (2017) (“‘[S]afe seats[ ]’ . . . distort[ ] not only electoral results, but also the electoral process as a mechanism by which representatives are held accountable to the people they represent. Almost 90 percent of the House of Representatives[’] seats are safe seats today. . . . As a practical matter, representatives today do not represent the people; they represent the hardliners that form their party base.”).


[94]  See Mark E. Warren, Chapter 3: Accountability and Democracy, in The Oxford Handbook of Public Accountability (2014) (“From the perspective of accountability, [proportional representation] systems tend to be more responsive and inclusive than [single member plurality] systems; voters can maintain closer relationships with smaller parties that have more specific platforms relative to parties in SMP systems.”).


[95]  White, 412 U.S. at 765-71.