Naomi White resides outside Window Rock, Arizona, an area within the Navajo Nation so rural that the Postal Service does not provide home delivery.1 Because White’s voter-registration application bore a physical address that was “too obscure,” the Apache County Recorder, the agency charged with election administration for the county, could not assign her to a voting precinct, considered her to be an inactive voter, and did not allow her to vote by absentee ballot.2 As a result, White, who is Navajo, was kept from voting in at least two elections in 2012.3
Agnes Laughter, another member of the Navajo Tribe, resides in Chilchinbeto, a community in the Navajo Nation 170 miles northeast of Flagstaff, Arizona.4 Laughter speaks only Navajo, does not read or write, and does not have an original birth certificate.5 When the state of Arizona passed a law requiring that voters provide certain forms of identification, Laughter was forced to travel to Flagstaff, and was able, but only after substantial effort, to obtain a form of identification that would be accepted at the polls — but not before she was kept from voting in the 2006 elections.6
Thomas Poor Bear is a member of the Oglala Sioux Tribe and resides in Wanblee, Jackson County, South Dakota, on the Pine Ridge Reservation.7 In order to register to vote or to vote absentee in person, Poor Bear and other members of the Oglala Sioux Tribe must travel almost sixty miles round-trip to Kadoka, the county seat — even though an estimated 22% of Indian households in Jackson County do not have access to a car.8 Poor Bear had asked the County to establish a satellite elections office in Wanblee, the population center of the Jackson County portion of the Pine Ridge Reservation, but the County initially refused.9 Only after Poor Bear and other members of his tribe filed suit did the County open the satellite office.10
The difficulties that White, Laughter, and Poor Bear have experienced are by no means unique: Indians routinely face hurdles in exercising the right to vote and securing representation. Though Indians were granted federal citizenship in 1924,11 their right to vote continued to be challenged both up to the Voting Rights Act of 196512 (VRA) and afterward. And though the VRA and subsequent amendments have aided the Indian franchise,13 the law continues to be an incomplete solution. Recent developments in election law jurisprudence (including the dismantling of preclearance under sections 4 and 5 of the VRA14) reveal cracks in the enforcement foundation.
This Chapter begins by reviewing, in section A, the history of Indian voting rights and the means through which Indian disenfranchisement has been attempted. Section B discusses the VRA — the primary tool used to protect voting rights — and its limitations, especially in light of recent election law developments. Section C reviews recently proposed federal legislation aimed at protecting Indian voting rights and considers possible challenges to such legislation.
Barriers faced by Indians seeking to exercise the franchise frequently “resemble the ones confronted by blacks in the South and Latinos in the Southwest.”15 These obstacles to voting and representation, like those constructed elsewhere, have made the transition from “first generation barriers” — those explicitly aimed at denying minority voters the vote outright — to more subtle “second generation barriers.”16 This section traces that evolution, from the use by states of Indians’ “distinctive status within the American political order”17 to justify disenfranchisement, to the newer techniques of vote dilution and vote denial that prevent Indians’ full political participation.
1. The Old Vote Denial. — Before the ratification of the Fourteenth Amendment, the Constitution did not explicitly define citizenship, merely making reference to “Citizen[s] of the United States”18 and “Citizens of each State.”19 Direct references to Indians were limited to the Indian Commerce Clause20 and the Three-Fifths Clause, which, in addition to providing that slaves would be counted as three-fifths of persons for apportionment purposes, established that “Indians not taxed” would be excluded.21 This ambiguous text was “understood neither to expressly confer U.S. citizenship on Indians nor to expressly prohibit extending citizenship to Indians.”22
The Civil War Amendments addressed citizenship and voting — the Fourteenth Amendment conferred both federal and state citizenship upon “persons born or naturalized in the United States, and subject to the jurisdiction thereof,”23 though again “excluding Indians not taxed” for apportionment purposes,24 and the Fifteenth Amendment prohibited the denying of the right to vote on the basis of race.25 But even they did not secure citizenship (and the attendant right to vote) for Indians. In 1884, the Supreme Court directly addressed the issue of Indian citizenship under the Civil War Amendments, holding that Indians born within the territory of the United States were “not . . . citizen[s] of the United States under the Fourteenth Amendment” and thus had no claim to the right to vote, absent individual naturalization or collective tribal naturalization by treaty.26
This ad hoc approach to Indian citizenship would continue for another forty years. By 1924, approximately two-thirds of Indians had acquired United States citizenship,27 and accordingly had the right to vote — in law, if not always in practice. The Indian Citizenship Act28 extended federal citizenship — and the attendant protections of the Fifteenth Amendment — to the remaining one-third. But this federal proclamation hardly resolved the issue in the states.29
Barriers to voting were, or had already been, erected. Many states employed facially neutral measures, such as poll taxes or literacy tests, intended to avoid the proscriptions of the Fifteenth Amendment30 — techniques mirroring those deployed against African American voters throughout the Jim Crow South.31 Further, drawing on Indians’ “unique status of citizenship at four levels of government” (federal, state, local, and tribal)32 and the complex history out of which that status arises, states deployed distinct methods of disenfranchising Indians33: First, mirroring the Three-Fifths Clause and the Fourteenth Amendment, some states explicitly disenfranchised “Indians not taxed.”34 Others passed statutes defining residency to exclude Indians living on reservations.35 Additionally, some states imposed tribal-relation limitations, extending the franchise only to Indians who had terminated their tribal relations and were deemed sufficiently “civilized.”36 Finally, finding support in Chief Justice Marshall’s pronouncement that the relationship of Indians “to the United States resembles that of a ward to his guardian,”37 states disenfranchised Indians on account of their alleged under-guardianship status.38
These explicit barriers to Indian voting would fade. The Arizona Supreme Court overruled its prior interpretation of the “guardianship” qualification to voting in 1948.39 Other states with taxation, tribal-relationship, guardianship, and other qualifications removed those barriers as well,40 leaving Utah and Maine as the last states with formal exclusions;41 those restrictions would fall by the end of the 1960s.42
2. Vote Dilution. — But even without explicit legal regimes preventing Indians from voting, states and local jurisdictions could nonetheless reduce, or “dilute,” the power of Indian votes that were successfully cast by manipulating political geography. State and local officials have applied numerous redistricting techniques — including the adoption of at-large voting systems, malapportionment, changing the size of representative bodies, and outright gerrymandering — to ensure that even if Indians could nominally vote, their votes would not translate into meaningful representation.
A jurisdiction’s representative body can choose one of two representation systems: under an at-large system, each representative is elected by voters in the entire jurisdiction; under a district-based system, the jurisdiction is divided into districts, each of which elects a specified number of representatives.43 But “because of the numerical advantage nonminority voters hold in many municipalities,” minority voting power may be “submerge[d]” in at-large elections in a way that would not happen in a district-based system.44 For example, Fremont County, Wyoming, employed an at-large system for its county commission between 1986 and 2010,45 making it difficult for Indians to “attain[] representation on the five-member board”46 even though the county had become more than 21% Indian by 2005.47
Even if a jurisdiction adopts a district-based election system, it could nonetheless “malapportion” its districts — that is, create districts with substantially disparate numbers of people, in contravention of the “one person, one vote” principle established by the Supreme Court.48 In a malapportioned area, though each district has the same amount of representation, residents in overpopulated districts are underrepresented;49 this imbalance often translates into the underrepresentation of certain demographic groups in the representative body.50 Indeed, malapportionment has been used to the detriment of Indian representation.51 In an extreme example from the 1970s, Apache County, Arizona, was malapportioned to such an extent that the largest district, which was 88% Indian, was fifteen times the population of the smallest district, which was 4% Indian52 — a disparity well above the 10% threshold at which a plan becomes constitutionally questionable.53
But even with fair apportionment, other fundamental changes to the representative body can diminish Indian representation. Charles Mix County, South Dakota, provides an example of malapportionment applied to the detriment of Indian voters that was remedied through litigation.54 After the conclusion of the Charles Mix County litigation — and the election of the County’s first Indian county commissioner — the County unsuccessfully attempted to increase the number of commissioners from three to five.55 If the increase had been implemented, “Native American voters [could have] elect[ed] their candidate of choice in only one of five districts, as opposed to one in three districts”; because “there [was] no reasonable probability that Native American voters could elect their candidate of choice” in another district, the increase in commission size would effect a reduction in Indian representation.56
And if all else fails, jurisdictions may turn to traditional gerrymandering practices, either by “cracking” — the separation of Indian communities across multiple districts — or “packing” — the consolidation of Indian communities into a single district such that Indian representation in the overall representative body is limited.57
3. The New Vote Denial. — In place of explicit bars to voting and in conjunction with vote-dilutive mechanisms, states often deploy measures aimed at making each stage of the voting process — from registration, to the securing of voter identification, to access to the physical places and mechanisms for voting — more difficult. These restrictions are not explicitly directed at Indian voters, but fundamentally, increasing the cost of voting imposes the greatest burdens on communities and groups least able to bear them. Indians are far from the only minority group disadvantaged by these various practices, but nonetheless, three factors especially affecting Indians should prompt particular concern: geography, language, and what this Chapter terms “concurrent citizenship” — the fact that an Indian voter is simultaneously a member of her tribe and a citizen of the United States and the state and local jurisdictions in which she resides.
(a) Physical Geography. — For many Indians, geography impedes each step of the voting process. Consider Naomi White’s difficulties in registering to vote. Registration in person may be difficult given geographic distance, and though states must provide for voter registration by mail,58 “[r]egistration forms have been rejected for failing to have a proper street address, even though there is no address numbering system in many rural areas.”59 Similarly, when attempting to obtain voter identification,60 “Native Americans, especially those living on rural reservations, may not be able to provide proof of residence [required to secure identification] because many tribal communities do not have street addresses.”61 State facilities providing identification services may also be distant from Indian communities — one of the difficulties that Agnes Laughter faced62 — and otherwise inaccessible due to a lack of viable transportation options.63
And even if an Indian voter successfully registers and procures the requisite identification, she may still lack access to the physical places and mechanisms for voting. Polling places are often far from Indian communities: some communities are more than 100 miles from the nearest polling place by road, and other communities are not connected by road to their polling places at all.64
Early voting and absentee voting can ameliorate this remoteness, but only partially. Early voting locations may be remote too, as Thomas Poor Bear experienced,65 and in some instances just as remote as election-day polling places.66 Absentee voting is viable only assuming reliable mail service — an assumption that often does not hold in Indian country.67 As courts have recognized, while absentee voting may generally improve participation, problems “with stable housing arrangements, poverty and transience” in Indian communities, among other socioeconomic problems, actually “make[] mail balloting more difficult for tribal members.”68
The actions taken by some state elections officials when challenged on the increased reliance on absentee voting confirm this understanding. Notably, Alaska unsuccessfully attempted in 2008 to withdraw physical polling stations from several Native villages and to consolidate those voting precincts with neighboring ones, whose polling stations were far from the Native villages and not accessible by road.69 The State further intended to designate the consolidated precincts for permanent absentee voting without first submitting that proposal to the Department of Justice (DOJ), as required by the VRA, for analysis of discriminatory impact.70 When the DOJ ultimately discovered what the State had attempted, the State withdrew both sets of changes rather than address the DOJ’s concerns71 — quite possibly a tacit acknowledgement that absentee voting is an imperfect substitute for in-person voting.
The problems of remoteness do not affect Indians alone, but Indians do live in rural areas at a substantially higher rate than the U.S. population overall.72 Further, the burdens imposed by geographic distance may be more onerous on Indians living in rural areas compared to the overall population, given lower rates of vehicle ownership specifically and more dire socioeconomic circumstances overall.73
Accordingly, decisions regarding how to conduct registration, what types of identification a voter must present (and where, when, and how she may obtain it), and where to situate early-voting locations and polling places become particularly important. But state and local officials may simply fail to act when these burdens fall on Indian voters rather than non-Indian voters. Particularly given the lack of Indian representation in many jurisdictions, county elections officials may simply be unaware, for example, of the lack of utility to many would-be Indian voters of distant early-voting locations.74 Or, more invidiously, perhaps officials are aware of the problems, but willfully opt not to take action.75
(b) Language. — Voters with limited English proficiency face substantial challenges in registering to vote, securing identification, and accessing the physical mechanisms for voting. Registration and securing identification often involve extensive dealings with government bureaucracy, which limited English proficiency may make especially difficult. Judge Posner, commenting on a voter-identification law, referred skeptics to a twelve-page appendix, cataloguing the expanse of paperwork that voters attempting to secure identification faced, “for disillusionment.”76
And though language support in Indian languages is often provided,77 some native languages such as Navajo and Zuni are “historically unwritten” and do not fit comfortably with the (English) written document–based process of voter registration, securing identification, and physical voting.78 Indeed, some things may literally be lost in translation — “[f]or instance, no word exists in the Navajo or Zuni languages for Republican or Democrat”;79 glossaries of election terms are sometimes needed.80
(c) Concurrent Citizenship. — Finally, the dual status of Indians as members of their tribes and as citizens of the United States presents challenges. Consider, again, voter-identification requirements. Many of the states that require voter identification do not accept tribal identification,81 an express rejection of the legitimacy of tribal government documents that may well be perceived as a challenge to tribal sovereignty.82 Being required to obtain state-issued identification in light of that denigration of tribal sovereignty may uncomfortably echo the self-termination voting qualifications imposed earlier in history.
Another issue arises when tribal elections and federal and state elections are scheduled simultaneously, as is the case in the Navajo Nation83 and for the Northern Cheyenne.84 While this approach “has worked well” for some tribes in increasing turnout,85 a lack of coordination between the two sets of elections can make it difficult for an Indian voter to participate in both the tribal election and the federal and state elections. Within the Navajo Nation, for example, tribal elections are conducted at chapter houses, around which Navajo political life is often organized,86 with chapters serving as, among other things, the equivalent of voting precincts.87 But chapter boundaries and voting precinct boundaries for federal and state elections are hardly coterminous,88 compounding difficulties imposed by physical geography. Polling places for federal and state elections are often not sited at chapter houses at all.89 A voter might be assigned to vote in two different places, and may have to expend considerable effort and resources traveling between them, making participation in both elections practically impossible for at least some voters.
The particular vulnerabilities of Indian voting rights and the various means through which states can inhibit Indian voting and suppress Indian representation necessitate a robust legal regime to protect those rights. The Voting Rights Act of 1965 — although passed with the disenfranchisement of African American voters across the Jim Crow South firmly in mind — has been the single most important tool in protecting Indian voting rights.90 While the VRA still holds plenty of promise, its efficacy has been limited by both inherent enforcement difficulties and doctrinal developments.
1. Section 5. — Under section 5,91 jurisdictions with a history of voting discrimination are subject to “preclearance,” under which they must submit changes to voting procedures to the DOJ.92 The DOJ reviews changes for either a “discriminatory purpose” or a “discriminatory effect.”93 Most importantly, under section 5, putative changes are reviewed before they threaten minority voting rights. For example, Alaska’s proposed 2008 precinct scrambling likely would have disenfranchised a substantial number of Alaska Native voters in villages that would have lost their physical polling places; those voters would have had to travel by air to vote at a polling station.94 Section 5 preclearance prevented these changes from occurring.
However, the Supreme Court’s decision in Shelby County v. Holder95 invalidated section 4(b),96 which provided the coverage formula that determined which jurisdictions would be subject to section 5 preclearance, thereby rendering section 5 inoperative.97 Neither the Court nor the D.C. Circuit (nor the dissenters in either court) mentioned Indian voting rights at all, but the case will substantially impact Indian voters. For one, four jurisdictions with substantial numbers of Indian voters — the entirety of the states of Alaska and Arizona and two jurisdictions in South Dakota (Oglala Lakota County98 and Todd County) — are no longer subject to preclearance under section 5.99 And, because voting changes are often made on a statewide basis,100 the withdrawal of preclearance coverage over the two South Dakota counties will likely have repercussions statewide, impacting an even greater number of Indian voters.
2. Section 2. — Section 2, as amended, outlaws voting qualifications and restrictions that “result[] in a denial or abridgement of the right . . . to vote on account of race or color.”101 Section 2 claims are of two varieties: vote dilution (that the law makes the right to vote less meaningful) and vote denial (that the law denies the right to vote outright).102 Though vote dilution claims have a longer lineage and a relatively clear doctrinal framework,103 they have little applicability outside the redistricting context — vote dilution presupposes a more-or-less unencumbered right to vote, after all. By contrast, vote denial litigation was rarely brought prior to Shelby County, a “paucity” that may “stem[] from the effectiveness of the now-defunct Section 5 preclearance requirements that stopped would-be vote denial from occurring.”104 The Courts of Appeal have taken varied approaches in considering these new section 2 claims.
The Fourth, Fifth, and Sixth Circuits have adopted a two-part disparate impact test.105 A voting procedure violates section 2 if (1) it disproportionately burdens a protected class and (2) that burden is at least partially “caused by or linked to social and historical conditions that have or currently produce discrimination.”106 None of these circuits have applied the test in the context of Indian voting rights, but their reasoning nonetheless applies. For example, the “unfortunate history of official discrimination [against African Americans] in voting and other areas”107 found in North Carolina applies to Indians in many states. Similarly, the statistical disparities found between black and white voters in North Carolina108 are also often present between Indian and white voters.109 By contrast, the Seventh Circuit has adopted a more stringent standard: even though minority voters disproportionately lack ready access to the documents required to secure voter identification, and therefore “must file more paperwork” than white voters, this “disparate outcome” did not amount to “a ‘denial’ of anything.”110 What mattered was that “everyone [had] the same opportunity to get a qualifying photo ID.”111 Under the Seventh Circuit’s test, the difficulties faced by Agnes Laughter, for instance, likely would not amount to a section 2 violation.
But even where the law may be doctrinally favorable,112 the substantial enforcement costs of section 2 impose a practical limitation on the protection it provides in both the vote dilution and vote denial contexts.113 “Litigation . . . is complex, time-consuming, and heavily dependent on access to sophisticated counsel.”114 The meandering path of a recent section 2 case is illustrative: in Cottier v. City of Martin,115 a group of Oglala Sioux plaintiffs brought suit in 2002 alleging vote dilution in the city’s at-large election system.116 The district court ruled against the plaintiffs, a ruling that was reversed by the Eighth Circuit and remanded,117 with the defendant’s petition for rehearing en banc subsequently denied.118 On remand, the district court then found in favor of the plaintiffs,119 a ruling that was affirmed by a panel of the Eighth Circuit.120 But this time, the Eighth Circuit would grant rehearing en banc (thereby vacating the panel opinion),121 and the en banc court would finally rule against the plaintiffs, directing the lower court to dismiss the case — more than eight years after the suit was initiated.122
Given these high enforcement costs, the existence of “below-threshold” section 2 violations — which may be especially likely when the communities being disenfranchised are less populous and more dispersed, as Indian communities are123 — coupled with the fact-specific, individualized inquiry required in each section 2 suit,124 renders section 2’s protection reactive and incomplete at best.125 And lingering in the background is the question of section 2’s continued constitutionality following Shelby County.126
3. Section 3. — Recognizing the limitations of section 2, scholars have argued for expanded application of the VRA’s other preclearance provision, section 3.127 Under this section, upon finding that a jurisdiction has violated the voting protections of the Fourteenth or Fifteenth Amendments, a court may “bail-in” the jurisdiction and require that it submit voting changes to the DOJ for preclearance.128 Section 3 has been used successfully in Indian country, with several jurisdictions having been bailed into coverage129: for example, Charles Mix County’s attempt to change the size of its county commission was discovered and prevented through preclearance review.130 Charles Mix County was not a jurisdiction subject to coverage under section 4(b), but had been bailed into coverage following the conclusion of an earlier litigation.131
While section 3, once applied, confers much of the protection that section 5 formerly did, its trigger threshold of a constitutional violation is often difficult to establish: City of Mobile v. Bolden132 requires a finding of discriminatory intent.133 (Indeed, many jurisdictions are bailed into coverage under section 3 as a result of consent decrees.134) As a result, assuming a group of plaintiffs even has sufficient resources to litigate a section 3 claim and the defendants are unwilling to enter into a consent decree, establishing a violation (and thereby attaining the protections of preclearance) is substantially more difficult than establishing a violation under section 2.135 Accordingly, though section 3 affords more robust protection, its protective scope is even more limited than that of section 2.
4. Section 203. — In contrast to the general provisions of sections 2, 3, and 5, section 203136 concerns a specific voting problem, that “citizens of language minorities have been effectively excluded from participation in the electoral process.”137 Like section 5, section 203 imposes affirmative obligations, unlike the simply proscriptive command of section 2. Jurisdictions with sizeable language-minority communities, as determined through a coverage formula,138 cannot provide voting materials only in English and must “provide them in the language of the applicable minority group as well”139 — thereby allowing greater access by voters who may have limited English proficiency.140
And, also unlike the general approach of sections 2, 3, and 5, section 203 directly addresses the specific language problems that Indian and Alaska Native voters face. First, Indian communities are explicitly included within the coverage formula: “a political subdivision that contains all or any part of an Indian reservation” is covered if “more than 5 percent of the American Indian or Alaska Native citizens of voting age within the Indian reservation are members of a single language minority and are limited-English proficient.”141 Second, the section acknowledges that “in the case of Alaskan natives and American Indians, if the predominant language is historically unwritten,” oral language support must be provided.142 Several Alaska Native voters recently brought suit alleging that the state of Alaska failed to satisfy its obligations under section 203, ultimately securing through settlement the translation of voting materials into two Alaska Native languages (including several dialects of one) and the placement of bilingual outreach workers in Alaska Native villages in numerous areas, among other forms of relief.143
Indeed, section 203, when enforced, has generally been viewed as effective: scholars have noted that its provisions “are not costly[,] can be efficiently implemented,” and “play a critical role in offering language minority citizens an equal opportunity to participate.”144 However, incomplete enforcement of section 203 remains an issue: “most of the cases in this area have been filed by the Department of Justice,”145 the enforcement resources of which are necessarily constrained. And finally, to the same extent that section 5 itself (especially as applied to state and local elections) could be unconstitutional on federalism grounds,146 so too could section 203, as it exacts many of the same “federalism costs” imposed by section 5’s “federal intrusion into sensitive areas of state and local policymaking.”147
Given the still-present barriers to Indian voting and representation, the current enforcement regime is inadequate.148 The VRA in its current form has done and can do only a partial job of vindicating Indian voting rights. As a starting point, scholars have called for federal legislation, not only to repair the VRA more broadly149 but also to address Indian voting rights specifically.150
The DOJ has drafted the Tribal Equal Access to Voting Act of 2015151 (TEAVA), much of which was incorporated into and expanded upon in the Native American Voting Rights Act of 2015152 (NAVRA) that was introduced in the Senate. The bills’ substantive provisions overlap considerably — both bills would cover local jurisdictions that encompass Indian Reservations and Alaska Native areas153 and would include, among others: (1) a preclearance provision prescribing federal review of changes affecting voter registration sites, early voting locations, and election-day polling stations on Indian reservations; and (2) a consultation provision requiring state and local jurisdictions that overlap with Indian reservations to consult with tribes when locating polling stations (including early voting stations).154 NAVRA also contains a tribal-identification provision under which states must accept tribal identification in satisfaction of identification requirements.155
Legislation like NAVRA has significant promise. NAVRA, like section 203 of the VRA, explicitly contemplates the difficulties faced by Indian voters. Many of NAVRA’s key provisions are tailored toward the burdens imposed by geography and concurrent citizenship that fall disproportionately or exclusively on Indians156 and are aimed at the greatest gaps in the current enforcement regime.157 And, by contemplating direct tribal involvement in elections administration, the legislation has flexibility to fit diverse tribal circumstances.158
Congress took no action on the Native Voting Rights Act of 2014 (which died accordingly at the end of the 113th Congress) and has taken no action on NAVRA since its introduction. Some commentators suggest that Congress may be more likely to act on protecting Indian voting rights because Indian rights are involved.159 However, Indian voting rights may be an exception to the exception: voting-related disputes have become highly partisan in nature,160 and legislation like NAVRA is likely to be perceived as having partisan effects.161 But even setting aside political viability questions, such legislation would likely be challenged as exceeding Congress’s authority, just as the VRA has been. This section analyzes three possible sources of congressional authority to enact legislation like NAVRA: Congress’s power under the Elections Clause, its enforcement power under the Fifteenth Amendment, and its “plenary” power over Indian affairs.
1. Elections Clause Authority. — The Elections Clause grants Congress the power to “at any time by Law make or alter” state regulations governing the “Times, Places and Manner of holding Elections for Senators and Representatives”162 — “comprehensive words embrac[ing] authority to provide a complete code for congressional elections.”163 As the Court explained in Arizona v. Inter Tribal Council of Arizona, Inc.,164 “federalism concerns . . . are somewhat weaker” when Congress legislates using its Elections Clause authority;165 “the States’ role in regulating congressional elections . . . has always existed subject to the express qualification that it ‘terminates according to federal law.’”166 But the federal limits of this congressional power are apparent in the plain text of the clause too, which accordingly cannot sustain legislation like NAVRA as applied to state and local elections.
Admittedly, elections for federal, state, and local offices are often held jointly. But this logistical coincidence far from implies that protection afforded in federal elections will extend to state and local elections. Indeed, following Inter Tribal Council, some states have severed federal election processes from state and local ones. Both Arizona and Kansas implemented two-tier registration systems before the 2014 elections: voters who met the minimum federal registration qualifications but not the additional state prerequisites were permitted to vote for federal offices only.167 While there are differences between establishing separate registration systems and, for example, establishing separate physical polling places, states nonetheless can take (and have taken) steps to keep federal restrictions on federal elections from affecting their conduct of state and local elections.168 Accordingly, some other basis for congressional action — extending to the state and local level, where racial discrimination is not only more rampant but also less likely to be addressed through litigation169 — is needed to ensure that the full protective potential of legislation like NAVRA can be realized.
2. Enforcement Power Under the Fifteenth Amendment. — Congress has the power to enact “appropriate legislation” to enforce the Fifteenth Amendment,170 the same power that was used to enact the VRA.171 NAVRA’s preclearance and consultation provisions resemble sections 4(b) and 5 of the VRA in many respects, and challenges to NAVRA would likely resemble those in Shelby County and its predecessor, Northwest Austin Municipal Utility District No. One v. Holder172 (NAMUDNO), in which the Court noted (but did not answer) “serious . . . questions”173 as to section 5’s constitutionality.
But NAVRA’s preclearance and consultation provisions differ from the VRA in several key respects. First, NAVRA’s preclearance provision is facially neutral in that it applies to all states and does not require a coverage formula.174 It does not “single[] out” a set of states as VRA section 4(b) problematically did,175 and so avoids a direct Shelby County hit. Second, the scope of actions subject to preclearance is substantially narrower than under the VRA: only certain changes to the placement and availability of polling places are subject to preclearance, rather than changes to “any . . . standard, practice, or procedure with respect to voting” under section 5.176 To the extent that Shelby County implies that the more rigorous congruence-and-proportionality test set forth in City of Boerne v. Flores177 provides the metric by which congressional action taken under the Fifteenth Amendment will be gauged,178 the narrower scope of NAVRA’s preclearance provision is less of an intrusion into states’ “sovereign authority over their election systems,”179 suggesting greater congruence and proportionality than the VRA.180 Third, preclearance under NAVRA differs from section 5 in that NAVRA does not have a sunset provision. This lack of a time limitation might suggest less congruence and proportionality,181 though the difference may be negligible given that members of the Court have expressed skepticism as to whether the VRA can be fairly considered temporary at all.182 To the extent that section 5 preclearance is itself constitutionally vulnerable, NAVRA preclearance may stand on firmer constitutional ground.
The comparison between NAVRA’s consultation provision and VRA preclearance is more direct in that the consultation provision identifies states “whose territory contains all or part of an Indian reservation,” upon which it imposes an affirmative obligation to place election sites at the request of tribes.183 The provision does “divide the States” with reference to geographic overlap with an Indian reservation, but the distinction arguably does not serve as a proxy for areas with discrimination as VRA section 4(b) fatally did.184 Rather, the consultation provision expressly contemplates coordination between jurisdictions and tribes, which requires the presence of a tribe. Arguably, then, the coverage formula serves primarily as a proxy for areas with tribal presence. Alternatively, even if NAVRA’s coverage formula is a proxy for discrimination, the formula is distinguishable from section 4(b) in that it would be defined with reference to 2010 Census designations,185 conditions far more “current” than those used by section 4(b).186 Opponents of legislation like NAVRA are likely to further argue that the presence of a reservation does not, by itself, establish the “‘pervasive,’ ‘flagrant,’ ‘widespread,’ and ‘rampant’ discrimination” that the Shelby County Court suggested is required to justify the distinction in the section 5 preclearance context,187 but NAVRA’s consultation provision is also less intrusive than section 5 preclearance. While there is no shortage of evidence of continued discrimination against Indian voters, the obligations of the consultation provision are much narrower than those of section 5: states are only obligated to establish a single polling place per tribe as a starting point,188 and many reservations are already allocated at least one polling place by state and local elections authorities.189 Therefore, just as the narrower scope of NAVRA preclearance should help to sustain its constitutionality under City of Boerne, so too should the consultation provision’s narrow obligations.
3. “Plenary” Power over Indian Affairs. — Congress also possesses “broad general powers to legislate in respect to Indian tribes,” often described as “plenary,”190 that is “employed as a shorthand for general federal authority to legislate on health, safety, and morals within Indian country.”191 If legislation like NAVRA can be considered an exercise of this power, the difficult questions regarding the bounds of Congress’s Fifteenth Amendment authority left after NAMUDNO and Shelby County can likely be avoided: as a practical matter, the Court has shown great deference to Congress when it legislates with respect to Indian tribes.192 But the legislation must fit within Congress’s power in the first place, and challengers are likely to argue at least three reasons that it does not.
First, legislation like NAVRA would be primarily oriented at protecting the voting rights of individual Indians. That is, NAVRA involves itself in the relationship between individual Indians and the states of which they are citizens, rather than legislating with respect to tribes.193 For one, this line appears illusory to begin with. Many programs, such as section 184 housing-loan guarantees,194 do not fall clearly on either side; NAVRA could equally be construed as ensuring that tribal interests are represented at the state level. Indeed, the individual-nature argument is undercut by both the consultation provision, which explicitly contemplates a role for tribes and imposes certain responsibilities on them,195 and the voter-identification provision, which requires that states recognize some degree of tribal sovereignty in the validity of identification issued by tribes.196
Second, Congress’s obligations in protecting voting rights can hardly be described as “unique” to Indians, as Morton v. Mancari197 could be read to require.198 Mancari set a deferential standard of review, holding that congressional action need only be “tied rationally to the fulfillment of Congress’ unique obligation toward the Indians” in order to be constitutional,199 whereas the protections of the Fifteenth Amendment extend to all citizens.200 But NAVRA could be framed as a protection of Indian rights motivated by concern for tribal-state relationships, and therefore draws a “political” distinction like that in Mancari.201 Given that states and their policies have significant impacts on Indians202 — not only in states with criminal jurisdiction over Indian reservations under Public Law 280203 but also in non–Public Law 280 states through state involvement in areas such as Indian gaming under the Indian Gaming Regulatory Act204 (IGRA) — protecting Indian voting rights is arguably part of fulfilling Congress’s obligation to Indians.
Finally, opponents of legislation like NAVRA could argue that such legislation exceeds the bounds of Congress’s plenary power because it infringes the states’ authority to conduct their own elections.205 That is, even if federalism concerns present less of an issue when Congress is exercising its plenary power,206 they may nonetheless help to define the boundaries of the power itself. Consider Seminole Tribe of Florida v. Florida,207 in which the Court held that the Indian Commerce Clause did not grant Congress the power to abrogate state sovereign immunity208 in an opinion imbued with federalist themes.209 However, the state interest that NAVRA affects — the ability of a state to conduct its own elections — is arguably less weighty than the abrogation of sovereign immunity contemplated by IGRA. Accordingly, even if Seminole Tribe indeed implies that federalism helps delimit the bounds of Congress’s plenary power, NAVRA’s status is far from preordained by that case.
Despite these possible arguments against NAVRA’s constitutionality, as a practical matter, the Court has shown great deference to Congress when it claims to be legislating pursuant to its plenary power; Congress’s most significant actions taken pursuant to that power — Public Law 280, ICWA, and IGRA, among others — have not been seriously challenged as unconstitutional.210 Similarly, NAVRA may very well be sustainable as an exercise of Congress’s plenary power.
Half a century after the Voting Rights Act of 1965 — and a full one after the Indian Citizenship Act — Indians still face a host of “second generation barriers” to voting and representation. While the Voting Rights Act has made significant progress in protecting Indian voting rights, its limitations have become all the more obvious recently. In its absence, something — possibly renewed congressional action in the form of legislation like NAVRA — is needed to fill the gap. At the very least, to the extent that voting rights will become increasingly reliant on private protection in this new doctrinal environment,211 the vulnerability of the Indian franchise must not be forgotten. This Chapter hopes, at minimum, to have brought attention to this oft-overlooked corner of the voting-rights world.