In Bond v. Floyd,1 the United States Supreme Court held that members of the Georgia Assembly could not deny civil rights activist Julian Bond his oath of office based on his antiwar statements.2 Bond, duly elected by his constituency, enjoyed “the widest latitude to express [his] views on issues of policy.”3 Bond’s right to speak was not merely an individual right; rather, his freedom of speech enabled his constituents to “be represented in governmental debates by the person they have elected to represent them.”4
Long viewed in a doctrinal silo, Bond in fact dovetails with a maturing opprobrium of the partisan gerrymander. For it seems odd to forbid the state to silence a representative of the people but to permit the state to deprive the people of representation in the first place through the partisan gerrymander. If the First Amendment secured Bond’s speech from censure both in his individual and representative capacity, it makes little sense to permit the state, by use of the partisan gerrymander, to do at an earlier juncture in the electoral process what it could not do after Bond was elected. No less peculiar is a jurisprudence that would forbid viewpoint discrimination against a legislator such as Bond, but would allow viewpoint discrimination against citizens who share Bond’s ideological commitments and wish to be assembled in a district to support him or someone like him. If Bond had been denied his oath of office for being a Democrat instead of for expressing antiwar views, his and his constituents’ First Amendment rights would have been no less assaulted.
In Whitford v. Nichol,5 Democratic plaintiffs alleged that the State of Wisconsin had gerrymandered its state legislative districts to punish Democratic voters for their party affiliation and the expression of their political views.6 In sustaining the plaintiffs’ complaints against defendants’ motions to dismiss and for summary judgment, however, the three-judge federal district court devoted no discussion to these First Amendments claims because the plaintiffs did not focus on them.7 Instead, the plaintiffs focused heavily on their Equal Protection claim, which was supported by evidence of an “efficiency gap” in Wisconsin’s redistricting. According to plaintiffs, the efficiency gap compares the total number of wasted votes for each party.8 A vote is wasted either when it is cast for a losing candidate or when it is cast for a winning candidate who did not need the vote to prevail. The difference between the parties’ wasted votes among all districts, divided by all votes cast, is the efficiency gap.9 According to plaintiffs, Wisconsin Republicans created an extreme efficiency gap that would skew redistricting in their favor for the life of the current redistricting plan.10
In Shapiro v. McManus,11 the United States Supreme Court held that a single judge’s dismissal on the merits of a redistricting case in which plaintiffs properly requested a three-judge district court was reversible error.12 Plaintiffs in Shapiro, registered Republicans, brought a First Amendment claim against the state of Maryland alleging its congressional redistricting burdened their freedom of association.13 Without reaching the merits of plaintiffs’ claim, a unanimous Supreme Court reversed on statutory grounds, finding the single judge was required to refer the case to a three-judge panel pursuant to 28 U.S.C. 2284(a).14 In significant dicta, however, the Court concluded that plaintiffs’ First Amendment claim was not “insubstantial” for federal jurisdictional purposes. The Court cited to Justice Kennedy’s concurring opinion in Vieth v. Jubelirer,15 in which he suggested that the First Amendment “may offer a sounder and more prudential basis for intervention” in partisan gerrymandering claims than the Equal Protection Clause.16 Because plaintiff’s First Amendment claim was “based on a legal theory put forward by a Justice of [the Supreme] Court and uncontradicted by the majority in any of [the Court’s] cases,” there was a substantial federal question presented by the Maryland plaintiffs’ gerrymandering claim.17
On remand, a three-judge district court denied the state of Maryland’s motion to dismiss the plaintiffs’ second amended complaint for failure to state a claim.18 Finding that the First Amendment worked in conjunction with other constitutional protections to prevent a state from diminishing citizens’ representational interests, the court held that plaintiffs state a constitutional violation if, applying a traditional First Amendment framework, they demonstrate that the state retaliated against them in the redistricting process for the expression of their views (i.e., how they voted in past elections) or their membership in a particular political party. In this context, actionable retaliation is established where plaintiffs allege a specific intent to injure them because of their partisan affiliation or voting behavior; a tangible, concrete injury such as vote dilution resulted; and that but for the specific intent to harm the plaintiffs, the injury would not have occurred.19 “Because there is no redistricting exception to this well-established First Amendment jurisprudence,”20 the plaintiffs in Shaprio were allowed to proceed with their suit.
Unlike Whitford, Shapiro supplies the contours of a First Amendment challenge to a partisan gerrymander. Yet in upholding plaintiffs’ Equal Protection claim against summary judgment, the court in Whitford set forth parameters that are also relevant to a First Amendment challenge to partisan gerrymandering. Namely, the court cautioned that judicial intervention should be circumspect and any test used should eschew mass invalidation of districting plans throughout the country.21 With that admonition in mind, this Article argues against the framework adopted by the district court in Shapiro and instead treats partisan gerrymandering as a form of content discrimination similar to Georgia’s attempt to deny the oath of office to Julian Bond. This type of content discrimination, however, is admittedly ubiquitous in redistricting.22 This Article therefore proposes a First Amendment test that examines the interaction of aggressively gerrymandered districts, as evidenced by the efficiency gap in Whitford, with the political polarization of a state. Where this interaction is such that the dominant party has achieved not just a partisan gerrymander but rather what Professor Lani Guinier has called a “deliberative gerrymander,”23 the basic right of voters to free thought, speech and debate is violated.24 Although Guinier writes of the deliberative gerrymander in the context of the legislative isolation of black elected officials,25 within a politically polarized electorate and legislative body, a partisan minority might endure a similar marginalization by the dominant party.26 Julian Bond’s exclusion from the Georgia Assembly constitutes the most extreme form of deliberative gerrymandering, but a spectrum of other behaviors can also manifest the phenomenon.
Part I of this Article situates the partisan gerrymander within First Amendment jurisprudence, arguing specifically the relevance of Bond v. Floyd. Part II explicates the parameters of a First Amendment challenge to partisan gerrymandering, distinguishing such a challenge from the new Equal Protection claim offered in Whitford. Part III applies the First Amendment test developed in Parts I and II to the facts of Whitford. Part IV contrasts the deliberative gerrymander approach to evaluating the constitutionality of a partisan gerrymander to the framework adopted in Shapiro.
I. THE EXPRESSIVE PURPOSES OF REPRESENTATION
This section begins with a curation of the scholarship that has drawn the link between the First Amendment and electoral structures and rules. It then situates Bond v. Floyd within the context of a group of cases that illuminate the proposition that elected officials have First Amendment rights of free expression that are derived from the Frist Amendment rights of their constituents. If, as Bond holds, the First Amendment forbids the state to discriminate against an elected official and his constituents based on the content of their beliefs and expression, it necessarily restrains the state from content discrimination in the drawing of electoral districts. The only difference between these two forms of content discrimination is the point in time at which they occur.
Drawing on the Supreme Court’s First Amendment jurisprudence, legal scholars have posited that the creation of electoral districts to the disadvantage of discrete voting blocs can violate a group’s right to free association27 and free expression.28The Supreme Court, however, has largely favored the Equal Protection Clause over the First Amendment in voting rights jurisprudence.29 The Court’s avoidance of the First Amendment in the voting context coexists uncomfortably with its concerns about viewpoint discrimination in the speech context because states can and do engage in such discrimination in formulating electoral rules.30 Rooted in concerns of democratic legitimacy, the First Amendment’s prohibition on viewpoint discrimination “reflects the broad concept that government cannot regulate speech based on its content, and regulations must be neutral as to both viewpoint and subject matter.”31
As did the plaintiffs in Whitford v. Nichol, in Vieth v. Jubelirer, the plaintiffs pled but did not pursue a First Amendment claim in challenging Pennsylvania’s congressional redistricting as an unconstitutional political gerrymander.32 In rejecting the plurality’s conclusion that partisan gerrymandering claims are nonjusticiable, however, Justice Kennedy picked up where the plaintiffs left off. Relying in part on the Court’s patronage cases which ruled unconstitutional consideration of partisan affiliation for government employment, Kennedy maintained that the First Amendment would be violated if a redistricting had the purpose and effect of burdening a group of voters’ representational interests.33
Legal commentators largely greeted Kennedy’s turn to the First Amendment with skepticism.34 Critics viewed Kennedy’s First Amendment approach as a nostrum that offered no clearer a standard for determining an unconstitutional partisan gerrymander than the Equal Protection claim that failed in Vieth.35 Supporters of Kennedy’s First Amendment turn did little to grow his rudimentary theory.36 Yet Bond v. Floyd, not invoked by Kennedy, is perhaps the most imbricated fit between the First Amendment and the partisan gerrymander. To be sure, its application requires an inference that a state’s constitutional disability to exclude an elected representative based on the content of his ideas extends to putative representatives and voters during the redistricting process. This analogy, however, is less a stretch than equating the First Amendment harm of the partisan gerrymander to voters with those of employees denied work by the state because of their political affiliation. Bond is about constitutional limits placed on political factions that seek to deny opponents representational opportunity based on their viewpoint. This aptly describes the partisan gerrymander as well.
B. The Relevance of Bond
In reflecting on his victory in Bond v. Floyd, Julian Bond celebrated the Court’s decision as “larger than a victory for the First Amendment. It was a reaffirmation of my constituents’ rights to free choice in casting their votes.”37 Indeed, it would require three elections in which Bond was overwhelmingly supported by his constituents before Bond was seated in the Georgia House of Representatives.
The speech to which members of the Georgia House objected was in fact not Bond’s. Rather, the Student Nonviolent Coordinating Committee (SNCC), of which Bond was the Communications Director, issued a statement critical of the United States’ involvement in Vietnam while failing to protect the rights of black citizens at home.38 SNCC’s statement read in part: “We take note of the fact that 16 per cent of the draftees from this country are Negroes called on to stifle the liberation of Viet Nam, to preserve a ‘democracy’ which does not exist for them at home. We ask, where is the draft for the freedom fight in the United States?”39
SNCC had issued the statement after Bond’s election to the Georgia Assembly but before he had taken his oath of office. During an interview with a local Atlanta radio station, Bond endorsed the SNCC statement.40 Thereafter, seventy-five members of the Georgia House of Representatives filed petitions challenging Bond’s right to be seated.41 A special committee of the Georgia House conducted a hearing at which Bond again asserted his agreement with the SNCC statement “without reservation.”42 The special committee agreed with the allegations of the challenge petitions that Bond’s embrace of the SNCC statement and his subsequent remarks demonstrated that he does not and would not support the Constitutions of the United States and Georgia and that “he gives aid and comfort to the enemies of the United States . . . .”43 Believing that Bond could not take the oath of office with the requisite “sincerity,” the Georgia House voted 184 to twelve not to allow Bond to be seated.44
Bond and two constituents challenged Georgia’s actions on First Amendment grounds.45 Writing for a unanimous Supreme Court, Chief Justice Earl Warren sustained Bond’s claim. The Court rejected Georgia’s argument that the oath provisions of the state and federal constitutions imbued it with authority to determine Bond’s earnestness in taking the oath of office, for “such a power could be utilized to restrict the right of legislators to dissent from national or state policy or that of a majority of their colleagues under the guise of judging their loyalty to the Constitution.”46
The Court also vivisected Georgia’s claim that while it could not constitutionally penalize Bond’s speech if Bond were a mere “citizen-critic” of the government, it could demand a higher standard of fealty to government policy from an elected official.47 Georgia, according to the Court, had it exactly backwards: “The manifest function of the First Amendment in a representative government requires that legislators be given the widest latitude to express their views on issues of policy.”48 The Court explained that the functional purposes for this breadth of latitude were to inform voters of politicians’ positions so that voters could make informed decisions and to ensure that voters “may be represented in governmental debates by the person they have elected to represent them.”49 Although the Court did not address the standing of the two constituents who joined Bond’s First Amendment action,50 the functional purposes enumerated by the Court clearly indicate that the rights vindicated by Bond were not merely personal to him.
Prior to Bond v. Floyd, the Supreme Court had intimated in general terms that the exercise of creating electoral units is one of empowering the voices of constituents and their representatives. Thus, Baker v. Carr51 granted standing to citizens in one district who had been disfavored vis-à-vis citizens in less densely populated districts because the former have “an interest in maintaining the effectiveness of their votes.”52 Similarly, Reynolds v. Sims53 codified one-person, one-vote on an impliedly expressive premise: “Legislators represent people, not trees or acres. Legislators are elected by voters, not farms or cities or economic interests. As long as ours is a representative form of government, and our legislatures are those instruments of government elected directly by and directly representative of the people, the right to elect legislators in a free and unimpaired fashion is a bedrock of our political system.”54
Neither Baker nor Reynolds, of course, was decided under the First Amendment; they are Equal Protection cases. Yet Bond bears an important if largely unknown relationship to these cases because the legislative district from which Bond was elected resulted from one-person, one-vote reapportionment litigation.55 One-person, one-vote jurisprudence helped to break a rural stranglehold of legislatures and to add diverse (and yes, controversial) voices to them like Bond’s. More to the point, the concept of representation in both Baker and Reynolds embodies the free speech of the representative—and by extension his constituents—both as a practical and precedential matter.
First, the United States Constitution itself presupposes free speech as an ordinary function of elected representatives in the Speech and Debate Clause, which insulates Representatives and Senators from suit “for any Speech or Debate in either House.”56 Legislative immunity “insures that legislators are free to represent the interests of their constituents without fear that they will be later called to task in the courts for that representation.”57 The free expression of legislators, while not unbounded, is absolutely protected where such expression is essential to the deliberative processes of the legislature.58
Second, elected officials, like other government employees, do not forfeit their free speech rights as a condition of their government employment.59 Although the Supreme Court has held that public employees have no protected rights of expression when they speak pursuant to their official duties,60 most courts have not extended this restriction to the speech of elected officials.61 Nor could they without overruling Bond’s central tenet: “The manifest function of the First Amendment in a representative government requires that legislators be given the widest latitude to express their views on issues of policy.”62
Third, the free speech rights of an elected representative derive from those who elect him or her. In recently rejecting the Arizona legislature’s attempt to invalidate congressional redistricting performed by an independent commission established pursuant to a citizens’ initiative, the United States Supreme Court understood the Elections Clause of Article I of the Constitution to be premised on the core idea that “all political power flows from the people.”63 Arizona had argued that the federal Constitution’s command that the manner of electing House members be determined by “the legislature” of each state64 meant that the independent commission lacked authority to perform redistricting even though Arizona’s constitution expressively provided for the citizens’ initiative as a means of law-making. The Court disallowed the Arizona legislature’s attempt to divest its citizens of law-making authority and repeatedly invoked “the animating principle of our Constitution that the people themselves are the originating source of all the powers of government.”65
If legislative authority derives from the people, then the expressive rights attendant to the exercise of such authority has the same source. Thus, the infringement on Julian Bond’s right of expression was likewise an infringement upon those of his constituents for whom he spoke. If it was impermissible for Georgia to punish Bond’s antiwar speech once he was elected to the state assembly, there is simply no sound reason to allow Georgia to discriminate against Bond and like-minded voters in the state-controlled processes that antedate an election, namely redistricting. The partisan gerrymander does precisely this, however. This Article next sets forth the parameters of the First Amendment’s prohibition on the partisan gerrymander.
II. PARTISAN GERRYMANDERING, THE DELIBERATIVE GERRYMANDER AND THE FIRST AMENDMENT
Partisan gerrymandering almost always requires viewpoint discrimination against some candidates and voters.66 It does not follow, however, that the practice is always constitutional or irremediable. In Vieth v. Jubelirer, Justice Scalia and a plurality of the Court proceeded from the implicit premise that the sheer ubiquity of the practice of partisan gerrymandering adorned it with presumptive legitimacy.67 Yet the ubiquity-equals-validity argument in the gerrymandering context is as untenable as the Court found it to be in the campaign finance context, where the myriad and longstanding prohibitions on the use of corporate treasuries to fund candidate elections did not impede the Court from finding that such prohibitions violate the First Amendment.68 Drawing further from the campaign finance context, the Court has concluded that governmental attempts to equalize campaign resources by limiting the expression of a candidate or party are repugnant to the First Amendment.69 If this is so, then the First Amendment must similarly disapprove of governmental efforts to limit expression through the partisan gerrymander, the goal of which is not the laudable aim of equality but rather the suspect objective of enshrining political inequality.
Concerns with the manageability of such a principle—namely, whether it would enmesh the courts in the redistricting arena interminably—rest on a misapprehension of the nature of the First Amendment harms of the partisan gerrymander. Suppose Julian Bond was the candidate of the Anti-War Party and this faction commanded significant support in Georgia in the early 1960s. Under Vieth, of course, the Anti-War Party is not entitled to control any legislative districts because, as it is now gainsaid, the Constitution does not compel proportional representation.70 But where the state of Georgia sets out to do to the Anti-War Party what the Georgia House of Representatives actually did to Julian Bond—exclude it for holding and expressing a particular viewpoint—the injury inflicted is not merely a numerical disadvantage but a deliberative one. That is, the state has sought to exclude or limit the expression of an identifiable viewpoint in the essential process self-government. Certainly proof that the dominant party has intentionally made it more difficult for the Anti-War Party to elect its candidates is highly relevant to the First Amendment claim. The First Amendment claim, however, may also be amenable to other evidence of a discernible hostility to the party’s viewpoint accompanied by proof of specific efforts to cabin the party’s voice at stages of the political process other than reapportionment.
Upon reading Justice Scalia’s plurality opinion in Vieth, one could be lulled into thinking that such proof is rare to non-exist. Citing no political science literature or data, Scalia offered a portrait of the American electorate in which:
[A] person’s politics is rarely as readily discernible—and never as permanently discernible—as a person’s race. Political affiliation is not an immutable characteristic, but may shift from one election to the next; and even within a given election, not all voters follow the party line. We dare say (and hope) that the political party which puts forward an utterly incompetent candidate will lose even in its registration stronghold. These facts make it impossible to assess the effects of partisan gerrymandering, to fashion a standard for evaluating a violation, and finally to craft a remedy.71
Contrary to these musings, party identification continues to be one of the most important predictors of voting behavior.72 More to the point, the American electorate has been growing more polarized over time, with the Republican Party becoming more conservative and the Democrats becoming more liberal.73 This polarization is reflected in a number of measures of the electorate, such as the dissonance in the approval ratings of the president between Democrats and Republicans, a gap which has become inelastic.74 The sources of contemporary polarization are diffuse, but according to two leading scholars of Congress and the political process, the polarization has been asymmetric: it has ensued largely from the Republican Party’s lurch to the right.75 Asymmetric polarization is relevant in evaluating one party’s incentive and intent to suppress the expression of another. In their best-seller, It’s Even Worse Than It Looks, Thomas Mann and Norman Ornstein meticulously document the roots of present-day political dysfunction in the United States, noting,
The center of gravity within the Republican Party has shifted sharply to the right . . . . The post-McGovern Democratic Party, while losing the bulk of its conservative Dixiecrat contingent, has retained a more diverse constituency base, and since the Clinton presidency, has hewed center-left, with an emphasis on the center, on issues ranging from welfare reform to health policy.76
Mann and Ornstein have observed a worsening of polarization since the publication of their book in 2012. They recently argued that the current populist wave driving the insurgent candidacy of Donald Trump is the result of the Republican Party and its leadership’s “rejection of the norms and civic culture underlying our constitutional system . . . .”77 That rejection manifests itself in “[c]ontinuing to fan the flames of hatred of Obama” and “pursu[ing] a strategy of unified opposition to every Obama policy and initiative, including those they had recently supported . . . .”78 Consistent with these findings, Republicans express significantly greater antipathy for Democratic policies than vice-versa.79
From a deliberative governance standpoint, there is now a “huge” decline in Congress in the number of members of one party willing to vote with the opposite party, particularly the number of Republicans willing to vote for Democratic legislation.80 But it is not merely the bipartisanship of the roll-call vote that has suffered. A variation on Bond, members of the minority party in the House of Representatives are now excluded from key junctures of the law-making process, such as conference committees to reconcile House and Senate bills.81 Vanquished from the deliberative process in the House is the “duty-to-deal” that is the hallmark of deliberative democracy.82 This coarsened state of affairs was recently thrown into sharp relief when House Democrats conducted an unprecedented sit-in on the House floor in an attempt to compel a vote on gun control legislation in the wake of a mass shooting at a gay club in Orlando, Florida.83 While not rising to the level of immutability referenced in Scalia’s comparison of race to politics in Vieth, partisan polarization has interacted with the partisan gerrymander to produce something more insidious than the ordinary incidents of minority political status, at least at the federal level.
A concomitant feature of increased polarization is decreased ticket-splitting. Ticket-splitting describes a discursive preference by a voter whereby he votes for candidates of different parties for different offices within the same election.84 The plurality in Vieth argued that ticket-splitting made ascertaining which party was the majority party, and hence which party was entitled to a majority of congressional seats, impossible. Justice Scalia opined, again without reference to social science data, “[T]o think that majority status in statewide races establishes majority status for district contests, one would have to believe that the only factor determining voting behavior at all levels is political affiliation. That is assuredly not true.”85 Vieth was decided in 2004. According to political scientist Gary C. Jacobson, the decline in ticket-splitting was well underway even as the Vieth plurality based its decision on judicial hypothesizing:
In the 1970s, a quarter of the House and Senate electorates reported voting a split ticket; by the 2000s, the average incidence of ticket splitting had fallen to 16% in House elections, 13% in the Senate elections. The rates for 2012 were 11.4% and 10.9%, respectively. Reflecting these individual-level changes, the proportion of House districts delivering split verdicts—preferring the president of one party, the House candidate of the other—has fallen dramatically since the 1970s . . . reaching a remarkably low 6% in 2012. The same trend occurred in Senate elections . . . although split outcomes remain considerably more common in statewide elections because states tend to have greater political heterogeneity and a more even partisan balance than congressional districts. In 2012, only six states delivered split verdicts, and after the election, only 21 senators represented states lost by their presidential candidate, both lows for the period covered.86
Political polarization and its attendant decline in ticket-splitting at once heighten the danger of the partisan gerrymander to free thought and expression while also rendering the corrective to the gerrymander more judicially manageable by making the actual partisan distribution in a given state easier to identify. Polarization endangers First Amendment freedoms because it decreases the prospects for deliberation and compromise, emboldening a gerrymandered majority to ignore and even punish the views of a gerrymandered minority.87 The decline in ticket-splitting wrought by polarization is an ironic potential curative to these anti-democratic tendencies, however, because it allows a court to more reliably use statewide results to envision what a fair legislative map may look like.88
The political polarization evident at the federal level extends to the states. Measuring the average ideological distance between the median Democrat and Republican in a given state legislature, nearly half the states were more polarized than Congress in 2013.89 Similar to Mann and Ornstein’s findings, Professors Boris Shor and Nolan McCarty find that Republican state legislators are polarizing more rapidly than Democrats,90 but they find significant variation across states. The more competitive the two major parties are within a state, the more polarized their legislative behavior.91
Focusing on the effects of a somewhat different kind of polarization more than two decades ago, Professor Lani Guinier criticized voting rights jurisprudence as insufficiently attentive to the post-electoral structures that impede the effectiveness of black legislators who gain access to power through the creation of majority-minority districts.92 Specifically, Guinier was concerned with the “deliberative gerrymander” which in effect transferred to the legislature the same racial polarization that prevented the election of blacks absent the creation of majority-minority districts. To Guinier, “Black electoral visibility is useless if district-based electoral arrangements gerrymander legislative decision-making and reproduce in the legislature a mirror image of a racially skewed electorate.”93
Guinier does not apply this concept to the First Amendment, or to the partisan gerrymander as such. Yet it is a highly germane framework for evaluating when a partisan gerrymander infringes rights of free thought and expression. There is an infringement when the partisan gerrymander is evidence of a broader purpose to exclude the views of the opposing party from important aspects of the deliberative process of self-government, as does the deliberative gerrymander. This is certainly a reasonable interpretation of Justice Kennedy’s concurrence in Vieth, in which he wrote: “The inquiry is not whether political classifications were used. The inquiry is whether political classifications were used to burden a group’s representational rights.”94 Although Bond v. Floyd concerned the literal exclusion of a representative to burden representational rights, nothing in Kennedy’s concurrence or, for that matter, in reason, suggests that this is the only way to unconstitutionally burden such rights.95
The test outlined above does not displace but instead supplements social science evidence of extreme gerrymandering such as the efficiency gap proffered by the plaintiffs in Whitford or decreased ticket-splitting that renders statewide results more indicative of the true partisan breakdown of a state. Such evidence is foundational to a free expression claim since it supplies a reasonable baseline against which to compare outcomes. Unlike an Equal Protection claim, however, a First Amendment argument focused on the hostility of the dominant party to views of its opponents lends itself to a more searching analysis of other types of evidence, such as partisan polarization. The degree of political polarization in a jurisdiction or legislative body, and the parties’ respective roles in that polarization, will naturally bear on questions of intent and effect.
III. APPLYING THE DELIBERATIVE GERRYMANDER TEST TO WHITFORD V. NICHOL
By 2014, only five state legislatures were more politically polarized than Wisconsin; Wisconsin was more politically polarized than Congress.96 That divide is mirrored by a racial and geographic divide in which the majority-minority city of Milwaukee is surrounded by predominantly white suburbs whose minority population is in the single digits.97 Metropolitan Milwaukee is the most polarized portion of a highly polarized state.98 The counties surrounding Milwaukee are among the most Republican of anywhere outside the south.99 By other metrics of polarization, Wisconsin, and Metropolitan Milwaukee in particular, stands out. The divide between Democrats and Republicans in perceptions of the president’s and the state governor’s performance is greater in Wisconsin than most other states.100 In Metropolitan Milwaukee, President Obama enjoyed a ninety-two percent approval rating among Democrats versus an eight percent approval rating among Republicans.101 Conversely, ninety-two percent of Republicans in the area approved of Republican Gov. Scott Walker’s performance, while only eight percent of Democrats did.102
These figures not only paint the backdrop for a gerrymander that the Whitford plaintiffs contend excessively wastes Democratic votes, they also are evidence of motive and intent to suppress the plaintiffs’ expression. Wisconsin Republicans acted on this motive on more than one occasion recently. In passing a restrictive voter photo I.D. law in the absence of any evidence of in-person voter fraud, for instance, there is strong evidence that the state sought to discourage voting by Democrats.103 Similarly, in a rapid two-week period, Wisconsin’s Republican-controlled legislature criminalized the mandatory collection of union dues by private employers, a step that will reduce the resources available to unions, a largely Democratic constituency, to support their preferred candidates in Wisconsin elections.104
At the same time Gov. Walker was signing legislation to weaken Democrats’ sources of campaign contributions, Republicans in the Wisconsin legislature passed legislation that made it easier for members of their party to raise money by, for instance, allowing corporate donations to political parties.105 In these examples, the very substance of the bills—to say nothing of questions of parliamentary procedure—are directed at the ability of Democrats not merely to get elected but to meaningfully participate in the political process at all. If Justice Kennedy’s concurrence in Vieth means anything, it must be intended to reach circumstances such as those in Wisconsin, where the excesses of hyper-partisan polarization have gone beyond the incidental partisan disadvantage of the gerrymander to injuries even more deleterious to the democratic process, namely the punishment and suppression of political beliefs and speech.
Were the plaintiffs in Whitford to compile (1) a record of legislation the foreseeable and intended consequences of which is to harm the ability of Democrats to participate in the political process for no other reason than their identification as Democrats106 and (2) a record of actions evincing an effort on the part of Republicans to exclude Democratic representatives from the deliberative processes of self-governance,107 this evidence would go beyond a mere showing that Republicans employed political classifications. It would powerfully imbue evidence of an extreme efficiency gap with an inference that the gerrymander was a vehicle to quash dissenting viewpoints. Indeed, Wisconsin appears to have become a model of the schismatic politics that the Court in Davis v. Bandemer108 held would justify invalidation of the partisan gerrymander.109 Writing for a plurality, Justice White insisted that plaintiffs demonstrate not merely reduced electoral opportunity but also a diminishment of their ability to influence the political process as a whole.110 The latter, in turn, required a demonstration of a lack of responsiveness by the elected representative to the interests of those who voted against him.111
Bond v. Floyd makes clear that the Constitution places limits on the conduct of political factions to seek advantage at the expense of their opponents. Although Wisconsin has stopped short of outright exclusion of Democrats from the state legislature, the unrestrained partisanship afoot inside the legislature flirts with virtual exclusion and, when combined with the extremity of Republicans’ gerrymander, likely crosses the constitutional boundary set by Bond.
IV. CONTRASTING SHAPIRO V. MCMANUS WITH THE DELIBERATIVE GERRYMANDER APPROACH
Although it is not as polarized as Wisconsin, Maryland is a highly polarized state and is more polarized than Congress.112 Unlike Wisconsin, Maryland has not been the subject of national news for one party’s efforts to remove the other from the process of self-government. And unlike the plaintiffs in Whitford, the plaintiffs in Shapiro are alleging a congressional gerrymander rather than a state legislative gerrymander. Nevertheless, the deliberative gerrymander approach applied to Wisconsin is fully applicable to the First Amendment claim in Shapiro. As in Whitford, the inquiry would be whether the partisan gerrymander is symptomatic of a more troubling process exclusion by the party in control or is an ordinary incident of being in the minority.113 The purpose of this focus is both to remain faithful to Bandemer’s formulation of the harm of the gerrymander and to limit judicial intervention into the inherently political process of redistricting. The traditional First Amendment framework applied by the district court in Shapiro fails to achieve these ends.
In rejecting the state of Maryland’s motion to dismiss and adopting a traditional First Amendment framework, the district court in Shapiro sought to ensure “that courts will not needlessly intervene in what is quintessentially a political process.”114 Consider, however, the ease with which most political gerrymanders would offend the elements of the speech retaliation claim set forth by court:
- Specific Intent: This element requires plaintiffs to demonstrate more than the mere fact that the legislature was aware of the likely political impact of its redistricting plan and adopted it notwithstanding its adverse effects on the plaintiffs.115 Instead, there must be a specific intent to discriminate against the plaintiffs for their past voting behavior.116 Absent a plan that seeks proportional representation of the parties,117 it is difficult to imagine a political party in control of a legislature that sets out to benefit itself in the redistricting process without intentionally setting out to disadvantage its opponent. Indeed, the more politically polarized a state legislature, the less likely such a result.
- Vote Dilution: Partisan gerrymandering claims are usually vote dilution claims.118 That is, the plaintiffs believe that their party has been given less its due in the redistricting process. The conundrum for partisan gerrymandering plaintiffs, however, is the question of what a non-dilutive plan looks like. The efficiency gap employed in Whitford and adapted to the deliberative gerrymander approach essays a resolution to this vexing problem. The court in Shapiro, however, simply incants a tautology: “whether the State’s intentional dilution of the weight of certain citizens’ vote by reason of their views has actually altered the outcome of an election.”119 Since this formulation simply begs the question of what weight the plaintiffs’ vote was entitled to in the first place, and since all successful gerrymanders alter election outcomes, the court’s injury requirement fails to provide meaningful parameters for the judiciary.
- Causation: The Shapiro formulation requires plaintiffs to demonstrate that their votes would not have been diluted but for the specific intent to disadvantage them because of their past voting behavior. This element is uncontroversial inasmuch as one would expect a partisan gerrymander to fulfill the intent of the map drawers unless something goes awry. But given this expectation, it is difficult to see how most partisan gerrymanders would not fail under the traditional First Amendment analysis used in Shapiro.
Perhaps that is the ultimate objective of plaintiffs. The Shapiro plaintiffs were joined in their opposition to Maryland’s motion to dismiss by several public interest amici, including Common Cause. Following the victory in Shapiro, Common Cause filed suit against the state of North Carolina, alleging that North Carolina’s latest congressional redistricting plan violated the First Amendment.120 Because Republican map drawers in North Carolina openly admitted that the purpose of their plan was to achieve a partisan gerrymander, the suit brings into sharp relief the fundamental question of whether a state legislature may ever constitutionally disadvantage adherents to a particular political party. The executive director of the North Carolina Chapter of Common Cause made clear the objective of Common Cause v. Rucho: “We believe our case can finally make clear that gerrymandering of any kind violates the constitutional rights of North Carolina voters.”121
Rucho arises from a prior litigation, Harris v. McCrory, in which a three-judge federal district court in North Carolina found the state’s congressional redistricting plan to be an unconstitutional racial gerrymander and ordered the state to create a contingent redistricting plan forthwith.122 Plaintiffs in Harris objected to the contingent plan as an impermissible partisan gerrymander, citing, as do the Rucho plaintiffs, the admission of the plan’s drafters that they intended to disadvantage Democrats.123 The district court overruled plaintiffs’ objection, holding that it was bound by the plurality holding in Vieth v. Jubelirer that the absence of manageable standards renders partisan gerrymandering claims nonjusiticiable.124
Plaintiffs in Harris have appealed the district court’s ruling to the Supreme Court on Fourteenth Amendment and First Amendment grounds.125 One is hard-pressed to conjure a better analogy to Julian Bond’s exclusion from the Georgia legislature than the political polarization that currently engulfs North Carolina and feeds its exclusionary political practices. A traditional First Amendment framework that focuses only on the fact of numerical disadvantage neglects to capture the myriad ways that a faction might seek to engage in viewpoint or associational discrimination. The totality of such actions can constitute a deliberative gerrymander that excludes opposing factions from the process of self-governance.
The deliberative gerrymander approach advanced is this Article is a line-drawing exercise that attempts to determine when partisanship has run amuck in the redistricting process. If all gerrymandering violates free expression, as suggested by some of the pending litigation, such line-drawing is obviated. Yet courts have not shown an inclination to cleanse the redistricting process so purely. As Whitford, Shapiro, Rucho, and Harris progress to the Supreme Court, perhaps clarity (and greater democracy) will finally be at hand.
* Distinguished Research Professor, DePaul College of Law. J.D., 1989, New York University; A.B., 1986, Brown University. I am indebted to Terrie Sullivan for her superb research assistance on this project.
**Cite as Terry Smith, Bond v. Floyd and Expressive Proscriptions On the Partisan Gerrymander, 2016 Wis. L. Rev. Forward 122, http://wisconsinlawreview.org/bond-v-floyd-and-expressive-proscriptions-on-the-partisan-gerrymander.