President Lyndon B. Johnson signs the Voting Rights Act of 1965

As Professor Cox’s article illuminates, however, the right to vote is considerably more elusive and conceptually difficult than most constitutional rights.3 Indeed, I tend to believe it is the most complex of all constitutional rights. Not only does the right to vote protect several different core interests, but these interests are also qualitatively distinct. Put in other terms, there is not one right to vote. There are several. Positive law, in the form of constitutional doctrine, recognizes this fact in practice, though with limited ability to articulate that fact incisively. And constitutional doctrine is right to recognize this fact: there are good normative grounds for treating the right to vote as protecting a number of qualitatively distinct interests. But for this very reason, discussion of “the right to vote”—in judicial decisions, academic commentary, and public discourse—is likely to be slippery and confusing. The general language of “the right to vote” elides the question of which right to vote (or better, which set of interests the right to vote protects) is, or ought to be, at stake in particular contexts. Under the general label of “the right to vote,” practical actors, such as judges, are likely to move back and forth between protecting qualitatively distinct interests. Analogies to other constitutional rights might be apt when the right to vote functions to protect certain interests, but completely misguided when the right to vote instead functions to protect other, qualitatively distinct, interests.

“The Temporal Dimension of Voting Rights” explores, elaborates, and nicely even adds to these complexities. Professor Cox begins by suggesting there are distinct “theories of voting rights” that can be “grouped into two categories.”4 The first views these rights as individualistic, as he puts it, in that one can identify harms to the right to vote without looking beyond the treatment of the individual voter. The second views these rights as collective or group or aggregate rights, in that harms to the right to vote can be identified only by looking at how the system of aggregating votes affects the distribution of political representation and power as between various groups.

I would put the starting point somewhat differently. It is not that there are distinct theories of voting rights. It is that the right to vote protects several distinct interests. The right to vote does protect the expressive interest in equal political standing that inheres to each citizen, taken one by one. But it also protects, as a matter of positive law, the interests groups of citizens have in systems of election and representation that distribute political power “fairly” or “appropriately” as between these various groups. Theorists like myself, who emphasize this second set of interests, do not deny that the right to vote also protects the individualistic interest in the act of casting a ballot itself—that is, that the right to vote protects the basic act of political participation. We believe that, in addition to that interest, the right to vote also protects an interest in the processes and institutional structures through which votes are aggregated. We emphasize these aggregate interests because in well-established, mature democracies, most of the actual conflicts that arise tend to be over aggregation, not over the individual act of participation. It is more plausible to imagine that those who believe the right to vote protects individualistic interests believe the right protects only those interests; indeed, Justices Thomas and Scalia appear to have precisely that view.5 But that is a rare view. Constitutional doctrine and the Voting Rights Act have recognized for several decades that the right to vote can be violated when election structures dilute the voting power of particular groups through the way in which those structures aggregate individual votes. Nor does Professor Cox cite any theorist who believes the right to vote protects only individualistic interests. Thus, it is not that there are competing theories of voting rights. It is that the right to vote protects several qualitatively distinct interests that citizens properly have, some of which are individualistic, some of which are more aggregative and group oriented.

Professor Cox’s project is to explore the aggregative interest the right to vote protects. “Temporal Dimension” is one of a series of Professor Cox’s incisive articles that deepen our understanding of this aggregative interest. Academic theory and judicial practice have long been familiar with the fact that, inherent in any doctrine or law that protects voters against having their voting power diluted, is the need to recognize such an aggregative interest. An individual voter can be disfranchised, but no vote can be diluted in and of itself. The concept of vote dilution requires examining the aggregative effect across defined groups of voters of election structures and practices. Professor Cox creatively generalizes from this recognized fact to expose two other dimensions along which “the right to vote” might also be conceived to protect or recognize aggregative interests the democratic system ought to acknowledge.

The first is the dimension of institutional aggregation, which we might picture as emerging from the fact that representative institutions are built up from a series of institutional building blocks. Thus, the United States Congress is constituted first by apportioning a specific number of representatives to each State delegation, then by electing individual members to each of those delegations from individual single-member election districts. If the right to vote should protect not just the formal right to cast a ballot for members of Congress, but the way those votes are aggregated across various defined groups, then we have multiple institutional possibilities for defining the relevant baseline for this aggregation. As Cox has pointed out, if our concern is the risk of partisan gerrymandering, should the baseline for a “fair” aggregation system be an individual State districting system?6 Or should that baseline be the partisan distribution of seats and votes on a nationwide basis, so that the Congress as a whole reflects a “fair” system of aggregation? Thus, if Texas’ congressional seats are aggressively gerrymandered to be “excessively” Republican, while California’s are “excessively” Democratic, should that constitute two instances of troubling partisan vote dilution? Or none, if the two states effectively wash out each other’s manipulations in the composition of Congress as a whole? The second dimension of possible aggregative concern is the one Cox focuses on in “Temporal Dimension.” We can be concerned with how votes are aggregated not just across groups or across institutions but also across time. If election structures diminish the power of some defined group at time A, but give that group enhanced power at time B, should this too be seen as a wash? Can diminishments in the way election structures aggregate the voting power of some group be compensated at some later moment in time? Should it matter whether the state imposes this diminished power on the group or whether the group chooses to forego some of its potential voting power at time A in order to have more power at time B?

In investigating the provocative questions Cox’s work opens, I believe it helpful to distinguish between ex post, judicial uses of his insights into the right to vote as an aggregative right and ex ante, legislative uses. When these issues arise in litigation, courts are asked to look back at a legislatively-created election structure and decide whether the Constitution imposes a unique aggregation system that some particular election structure—like grossly malapportioned election districts—violates. When these issues arise in a lawmaking context instead (whether in a legislative or a direct democracy process), lawmakers are asked to decide which of many particular aggregation systems might best realize various democratic values in specific contexts.

All of Cox’s applied examples arise in the context of ex post, judicially developed constitutional doctrine; he focuses primarily on racial vote dilution and partisan gerrymandering claims. In those contexts, he asks how courts should (or perhaps implicitly do, to some extent) take account of the potential for institutional and temporal aggregation in deciding these voting rights claims. Perhaps framing these issues in terms of “the right” to vote encourages this kind of judicialization of the analysis. But I believe the most provocative implications of Cox’s analysis lie not in ex post, judicial applications, but in ex ante, legislative ones.

For pragmatic, institutional, and even cognitive reasons, courts are inclined to narrow, not expand, the frames within which they settle legal disputes. This is true even of public law disputes, and even of ones that involve claims of right that could, in theoretical terms, be conceived in highly expansive terms—such as expanding the aggregative interest the right to vote protects to an aggregative interest across institutions and time, as well as across groups. One institutional reason is that courts decide cases one by one. Their ability to foresee what will happen in the next case, let alone to control in advance the outcome of that case, is limited. If courts were in the business of constraining partisan gerrymandering, it is highly unlikely they would be willing to accept a pro-Republican gerrymander in Texas because a pro-Democratic gerrymander is in existence in California. What happens should the legislature in California decide, for example, to redistrict after the court’s decision? Courts are rarely in the position of being able to apply synoptic rationality to problems that spill across institutions, states, and time. Similarly, to the extent courts might be tempted to aggregate voting rights over time, the fact that the court that sits at time A cannot control what the court that sits at time B might do would surely diminish this temptation. To a significant context, Cox himself recognizes that institutional constraints of these sorts will limit the extent to which courts will or can act on his conceptual expansion of the right to vote.

For these and related reasons, the most intriguing perspective from which to build on Cox’s insights is the ex ante, legislative one. In academic theory—and to a lesser extent, in political practice—at least three examples already exist that can be viewed as illustrating such potential applications. First, and most familiar, are systems of voting like cumulative voting (“CV”); such systems can be viewed as enabling individuals to aggregate their vote across any number of different groups, with the choice of the relevant group to be made by groups of voters themselves.7 When each voter has, for example, five votes to cast, that voter can express the intensity or his or her preference for a particular package of candidate and issues by concentrating most or all of those five votes on a particular candidate. To the extent other voters do likewise, those voters are aggregating their political power on the basis of that particular candidate/issues. We can understand cumulative voting as a system that responds to an aggregative interest in voting across groups. The state responds to such an aggregative interest when it intentionally designs election districts to concentrate a particular group of voters in particular districts in order to constitute them as the majority in that district. Unlike districting of this sort, in which the state determines and assigns the relevant aggregative group interest, CV systems enable individual voters to decide for themselves—election by election—how to define their aggregative voting interests.

A second example reflects the different possibilities that open up once we recognize that voting can be aggregated across institutional domains. This is a way to understand Professor Gerald Frug’s proposal that voters be given the choice of which local government unit to vote in with which they have certain minimal contacts.8 As he puts his novel proposal:

Consider a plan, for example, in which everyone gets five votes that they can cast in whatever local elections they feel affect their interest (“local” still being defined by the traditional territorial boundaries of city, suburb, or neighborhood). They can define their interests differently in different elections, and any form of connection that they think expresses an aspect of themselves at the moment will be treated as adequate. Under such an electoral system, mayors, city council members, and neighborhood representatives in the regional legislature would have a constituency made up not only of residents but of workers, shoppers, property owners in neighboring jurisdictions, the homeless, and so forth. People are unlikely to vote in a jurisdiction they do not care about, but there are a host of possible motives for voting (racial integration, racial solidarity, redistribution of wealth, desire for gratification, etc.). Indeed, there is no reason to think that the constituency would be limited solely to those who live in the region.

Though it is difficult to imagine a court concluding that the right to vote requires these options, it is easier to envision legislative bodies concluding, at least in some circumstances, that the right to vote should recognize an aggregative interest of this sort across conventional institutional boundaries.
A third example, again of the ex ante, legislative form, suggests a way in which the temporal dimension of the aggregative interest in voting might be recognized. As Professor Cox notes, recent academic work explores the idea of “storable votes.”9 First conceived by Professor Alessandra Casella, this idea envisions a system in which each voter (or legislator) in a jurisdiction is given an initial stock of votes. The voter can decide on which issues at various points in time to spend his or her votes. Much like legislators engage in implicit or explicit logrolling, a system of storable votes would enable individual voters to reflect their intensity of preferences over different issues. Voters would store their votes for the issues that mattered to them most and on which they believed their vote could make the most difference. Storable voting is a form of cumulative voting, but extended over time. Though numerous practical difficulties would have to be confronted before we can imagine instituting a system of storable votes, such a system is perhaps the best illustration of how democratic institutional design might take account of the fact that we can have reason to be just as concerned with the way voting power is aggregated over time as we are with how it is aggregated between groups at any one moment in time.

The right to vote is rich, complex, and exceptionally consequential. As soon as we recognize that it does and should protect both the individual interest in participation and the aggregative interest in how election structures bring together votes across groups of voters, we have moved into terrain far from that covered by more conventional constitutional rights. As Professor Cox’s essay identifies, this aggregative interest is yet more complex than has often been recognized. In honoring the right to vote, we can have reasons to be concerned with how votes are aggregated, not just across defined groups, but across institutions, jurisdictional boundaries, and even time. Deepening our understanding of this aggregative interest should help stimulate further creative possibilities for how democratic systems might be designed to best promote the various interests at which the right to vote should be understood to aim.

Preferred citation: Richard H. Pildes, What Kind of Right is "the Right to Vote"?, 93 Va. L. Rev. In Brief 45 (2007), http://www.virginialawreview.org/inbrief/2007/04/23/pildes.pdf.

[Back to this issue]

The Temporal Dimension of Voting Rights | Essay by Adam B. Cox
What Kind of Right is "the Right to Vote"? | Response by Richard H. Pildes