GERRYMANDERING AND ASSOCIATION.

• Author: Tokaji, Daniel P.
• Position: Special Issue on Gerrymandering

TABLE OF CONTENTS INTRODUCTION 2160 I. THE QUEST FOR A STANDARD 2166 II. AN ALTERNATIVE PATH 2177 A. The Right of Expressive Association 2177 B. Voting as Association 2183 III. THROUGH THE THICKET 2190 A. Gerrymandering as an Associational Injury 2191 B. Applying the Voting-as-Association Standard 2197 C. Answering Objections 2206 CONCLUSION 2209 INTRODUCTION

A legal standard for partisan gerrymandering is the holy grail of election law. (1) For decades, jurists and commentators have struggled to articulate a manageable standard that will avoid entangling courts in the political thicket. The pressure for courts to put some limits on partisan gerrymandering has intensified over the past decade as mapmakers have drawn lines with increasingly deadly precision, maximizing their own party's power while minimizing that of the other major party. (2) The enhanced technological tools at their disposal can allow the dominant party to retain its dominance throughout the decade, even in strong years for the other major party. (3) This effectively allows the party that holds the pen at the start of the decade to retain power for the next ten years, even when more voters would prefer a different outcome.

Still, a fierce debate rages over whether partisan gerrymandering claims are nonjusticiable political questions, as four Justices in Vieth v. Jubelirer believed. (4) There can be no denying the difficulty or gravity of the problem. On one level, partisan gerrymandering is a problem that cries out for judicial intervention. The sophisticated means through which the dominant party can entrench itself in power deny the accountability to voters upon which democracy depends while relegating its opponent to semi-permanent minority status. (5) This is not a problem that will self-correct. So long as the dominant party knows it can retain its legislative branch dominance--and its control over the redistricting process in the next cycle--it has no incentive to make a change. If one views judicial review as a means by which to check incumbent self-entrenchment and protect political minorities, thus making democracy work better, then there is a strong case for judicial intervention. (6) That is particularly true in the era of hyperpolarization that we inhabit, which makes it extremely unlikely that the major parties will negotiate a political solution.

At the same time, there are good reasons to worry about the feasibility of a legal standard for partisan gerrymandering claims, particularly under the Equal Protection Clause. However noxious the problem of partisan gerrymandering may be, it has a long pedigree. (7) The idea that partisan gerrymandering violates the Constitution is therefore a hard sell for originalists. (8) Nor is there an established body of equal protection doctrine upon which those challenging partisan gerrymandering can rely. There is certainly some helpful language to be found in Supreme Court precedent, most notably the Court's statement in Reynolds v. Sims that the Equal Protection Clause guarantees "fair and effective representation." (9) But equal protection jurisprudence is an imperfect fit for the problem of partisan gerrymandering, given its usual requirement that either facial or intentional discrimination be established. (10) The one-person, one-vote cases were an exception, allowing an equal protection claim to be made without proof of discriminatory intent, but this one-off was justified by the relatively bright line that the Court developed through the equal population requirement. (11) Partisan gerrymandering does not lend itself to a comparably simple solution. Any standard the Court creates will involve difficult judgment calls. There are good reasons to worry about opening Pandora's box.

While most constitutional partisan gerrymandering litigation has focused on the Equal Protection Clause of the Fourteenth Amendment, Justice Anthony Kennedy's Vieth concurrence suggested that the First Amendment might provide a more promising basis for challenges. (12) Although his three paragraphs on the subject were more suggestive than directive, the four cases he cited all involve associational rights. (13) Yet the scholarly commentary on the subject--most of it negative--has focused more on freedom of speech than freedom of association. (14) Two three-judge district courts--both of whose decisions will be considered by the Court this term--have looked more favorably on First Amendment association as a basis for challenging partisan gerrymandering. (15)

This Article argues that the First Amendment expressive right of association should be understood to prohibit excessive partisan gerrymanders. The right of association has long been understood to limit the dominant political group's ability to discriminate against its rivals. (16) And the ballot is one of the primary loci at which voters, candidates, and political parties associate. (17) Accordingly, the Supreme Court has understood voting as First-Amendment-protected association in certain contexts, including ballot access and primaries.

Expressive association provides the strongest constitutional basis for challenging partisan gerrymandering for three reasons. The first is that it best captures the character of the injury, which inheres in party-based discrimination--more specifically, the dominant political party's entrenchment of itself at the expense of the rival major party and its supporters. Voting is not just an individual right but also a collective activity. (18) It is the means through which we join our voices with like-minded others for the shared purpose of electing our leaders. And political parties are the main entity through which this aggregation occurs. (19) It requires no great doctrinal innovation to recognize this reality, nor to understand voting as associational activity deserving of protection under the First Amendment. The Supreme Court has long understood elections as a locus for association among voters, candidates, and parties. (20) It has therefore cast a skeptical eye on laws that unduly burden elective association, including restrictions on parties' and candidates' ballot access. (21) Because existing First Amendment jurisprudence already accords special attention to political parties as a group through which voters associate, it provides a firmer legal basis for partisan gerrymandering claims than the Equal Protection Clause does. While equal protection has long been the go-to basis for challenging burdens on voting, it is less well suited to address party-based discrimination. There is, by contrast, a rich body of First Amendment law involving discrimination based on party affiliation, on which courts might draw in developing an association-based doctrine of partisan gerrymandering. (22)

The second reason for viewing partisan gerrymandering through the lens of expressive association is that its legal test properly focuses on effect rather than intent. The Supreme Court's associational rights cases have long employed effect-based legal standards. (23) Among the practices to which such standards have been employed are ballot access, blanket primaries, and restrictions on party endorsements. (24) In the equal protection realm, by contrast, the Court has usually required a showing of discriminatory intent (25)--with the important exception of malapportionment cases, although even in this context the legislature's reasons for departing from the one-person, one-vote rule are still relevant. (26) There are formidable reasons to prefer a legal standard that is based on effect rather than intent in redistricting cases. When a legislature or other multimember body adopts a plan, it always has multiple purposes, including partisan advantage. The racial gerrymandering cases demonstrate the practical difficulties of discerning the "predominant" intent of line-drawers. (27) Unless the Court is prepared to declare even the slightest partisan motivation impermissible--a standard that could well render every existing district unconstitutional--it will find itself in a similar conundrum on partisan gerrymandering. An effects-based test is therefore preferable, and the First Amendment right of association provides the strongest doctrinal basis for embracing such a standard.

The third reason why expressive association provides a suitable vehicle for partisan gerrymandering claims is that it provides an appropriately nuanced standard. For very good reasons, the Court-including Justice Kennedy--has been unwilling to lay down a rule that would categorically prohibit any partisan considerations in redisricting. (28) Such a rule would render virtually every existing plan unconstitutional. It is, moreover, unrealistic to believe that partisan considerations can be wholly extirpated from the process. Attempting to do so would simply drive those considerations underground, encouraging even greater subterfuge than redistricting already inspires. There may also be good reasons for drawing districts in a manner that disproportionately benefits one major party or the other. An emphasis on compact districts might advantage Republicans, for example, while an emphasis on preserving communities of interest might benefit Democrats. Any viable redistricting standard should account for the good reasons that a state might have for drawing a plan that, in the aggregate, gives one party more seats than another with a comparable share of the vote. This turns out to be a major advantage of the balancing standard that the Court has adopted in its voting-as-association cases. It allows fair consideration of both the burdens that the challenged practice has on the nondominant party and the state interests that might justify these burdens.

Before proceeding further, I should clarify the boundaries of my argument. My goal is to work within the framework of existing case law to develop a legal argument for striking down excessively partisan gerrymandering based...