Davidson v. City of Cranston, RI

LYNCH, Circuit Judge. The City of Cranston appeals from an injunction, entered by the district court, forbidding the City from holding elections based on its 2012 Redistricting Plan and ordering it to prepare a new redistricting plan within- thirty days. The district court held that the inclusion in the Redistricting Plan of 3,433 inmates of the Adult Correctional Institutions (“ACI”) in the population count of the City’s Ward Six, the ward in which the ACI is physically located, dilutes the votes of voters in the City’s other five wards in violation of the Equal Protection Clause of the Fourteenth Amendment. The question presented is whether the Constitution permits Cranston to count the ACI prisoners as residents of Ward Six. We issued a stay to preserve the status quo ante in anticipation of the September 13, 2016 primaries and the November 8, 2016 general election. We now hold that the methodology and logic of the Supreme Court’s decision in Evenwel v. Abbott, — U.S. -, 136 S.Ct. 1120, 194 L.Ed.2d 291 (2016), require us to reverse the district court and instruct it to enter summary judgment in favor of the City. L. The Rhode Island Constitution specifies that state legislative districts “shall be constituted on the basis of population and ... shall be as nearly equal in population ... as possible.” R.I. Const, art. VII, § 1; id. art. VIII, § 1. Similarly, Cranston’s charter, ratified by the Rhode Island General Assembly in 1963, see 1963 R.I. Pub. Laws 550, “divide[s Cranston] into six wards in such a manner that ... all wards shall contain as nearly as possible an equal number of inhabitants as determined by the most recent federal decennial census,” Cranston, R.I., City Charter § 2.03(b). “Each ward elects one representative to *138 the City Council and one to the School Committee,” and all six wards collectively elect “three at-large city councilors and one at-large school committee member.” Davidson v. City of Cranston, — F.Supp.3d -, -, No. 14-91L, 2016 WL 3008194, at *1 (D.R.I. May 24, 2016). The U.S. Census “serves as a linchpin of the federal statistical system.” Dep’t of Commerce v. U.S. House of Representatives, 525 U.S. 316, 341, 119 S.Ct. 765, 142 L.Ed.2d 797 (1999) (citation omitted). Since 1790, the Census has produced its total-population counts by counting each person where he or she “usually resides,” see Act of Mar. 1, 1790, § 5, 1 Stat. 101, 103, and from 1850 to the present the Census has continually refined its “usual residence” rule for determining where to count persons. Currently, “usual residence” is defined as “the place where [persons] live and sleep most of the time.” 2020 Decennial Census Residence Rule and Residence Situations, 80 Fed. Reg. 28950 (May 20, 2015). The 2010 Census data used by Cranston in its 2012 Redistricting Plan, the most recent such plan, included in its population count for the City 3,433 inmates of the ACI. The ACI, “which is located on a state-operated campus,” Davidson, — F.Supp.3d at -, 2016 WL 3008194, at *2, is Rhode Island’s sole state prison. Although the ACI makes “most requests for police services [to] the State Police, which maintains an office at the ACI,” “the Cranston police occasionally deliver a prospective inmate to the prison.” Id. at -, 2016 WL 3008194, *2. The ACI also depends on Cranston’s roads and sewage system, as well as on the City’s fire department for emergency services. The City’s population in the 2010 Census was 80,387, and each of the City’s six wards includes approximately 13,500 persons, with a “total maximum deviation among the population of the six wards [of] less than ten percent.” Id. at -, 2016 WL 3008194, *1. The 3,433 ACI inmates were counted by the City as part of the 13,642 members of Ward. Six. If the inmates were not included, Ward Six would contain only 10,209 persons, and the maximum deviation among the population of the wards would be approximately thirty-five percent. “[D]emographic experts retained by the parties” testified that “153 or 155 [ACI] prisoners came from Cranston at the time of the Census,” and that “[eighteen of those had pre-incarceration addresses located in Ward Six.” Id. The plaintiffs’ demographic expert also testified that “the median length of stay for those serving a sentence at the ACI is 99 days,” and that “[t]he median stay for those awaiting tidal is three days.” Id. at -, 2016 WL 3008194, *2. Those inmates at the ACI not imprisoned for felonies may vote by absentee ballot in their pre-incarceration communities, provided that they meet that community’s absentee-ballot requirements. The Rhode Island Constitution forbids felons to vote while incarcerated, R.I. Const, art. II, § 1, but under Rhode Island law, non-felon inmates may vote at their “fixed and established domicile,” the location of which is unaffected by their incarceration, 17 R.I. Gen. Laws § 17-l-3.1(a), (a)(2). Taking into account testimony that “approximately 37% of the [ACI] population is serving a felony sentence,” the district court estimated that only “six [or] seven inmates ... could be eligible to vote in Ward Six.” 1 *139 Davidson, — F.Supp.3d at -, 2016 WL 3008194, at *2. In February 2014, four residents of Cranston and the American Civil Liberties Union of Rhode Island filed a complaint against the City under 42 U.S.C., § 1983 for declaratory and injunctive relief alleging that Cranston’s 2012 Redistricting Plan violates the Equal Protection Clause of the Fourteenth Amendment. The plaintiffs argued that the inclusion of the ACI inmates in Ward 6 “inflates the voting strength and political influence of the residents in Ward 6 and dilutes the voting strength and political influence of Plaintiffs and other persons residing outside of Ward 6,” thereby violating the equal protection principle of “one person, one vote.” 2 The City filed a motion to dismiss the complaint in March 2014, which the district court denied in September 2014. Davidson v. City of Cranston, 42 F.Supp.3d 325 (D.R.I. 2014). The parties then filed cross motions for summary judgment in July and August of 2015. In May 2016, the district court denied the City’s motion for summary judgment and granted summary judgment to the plaintiffs. It held that the City’s inclusion of the ACI inmates in its Redistricting Plan violated the principle of “one person, one vote” as consistently articulated by the Supreme Court, notwithstanding its recent decision in Evenwel v. Abbott. Davidson, — F.Supp.3d at -, 2016 WL 3008194, at *3-4. Rejecting the City’s argument “that Evenwel stands simply for the constitutional propriety of drawing district lines based on Census population data,” the district court instead stressed “the Supreme Court’s emphasis on the conceptual basis of representational equality.” Id. at -, 2016 WL 3008194, *4. The district court concluded that “[t]he inmates at the ACI share none of the characteristics of the [historically nonvoting] constituencies [such as women, children, slaves, taxpaying Indians, and non-landholding men] described by the Supreme Court” and found by the Court to deserve representation in apportionment. *140 Id. The district court found that the inmates have no interest in Cranston’s public schools, receive few services from the City, and have no contact with Cranston’s elected officials. Id. The court further emphasized that the “inmates are different from other groups of non-voting residents of Cranston,” including “college students and military personnel.”' Id. Unlike those non-voting residents, many of the inmates are forbidden by law from voting in Cran-ston. And though college students “are most certainly affected by municipal regulations,” the court concluded that the ACI inmates have no stake in the local political process. Id. The court noted that, “were the Cranston City Council to enact any ordinance bearing on the treatment of inmates or other conditions at the ACI, it would no doubt be preempted by state law and therefore unenforceable.” Id. These distinctions, the court reasoned, rendered Evenwel’s general approval of districting based on Census data inapplicable to the inclusion of prisoners in redistricting population counts. Id. The district court found support for its holding in Calvin v. Jefferson County Board of Commissioners, 172 F.Supp.3d 1292, No. 4:15CV131-MW/CAS, 2016 WL 1122884 (N.D. Fla. Mar. 19, 2016), a case decided shortly before Evenwel. Davidson, — F.Supp.3d at -, 2016 WL 3008194, at *5. In Calvin, the court held that the Equal Protection Clause barred Jefferson County from including prisoners from the Jefferson Correctional Institution, a state prison, in its redistricting population count. 172 F.Supp.3d at 1325, 2016 WL 1122884, at *28. The Calvin court reached this conclusion because the prisoners “comprise a (1) large number of (2) nonvoters who (3) lack a meaningful representational nexus with the [County] Boards, and ... [are] (4) packed into a small subset of legislative districts.” Id. at 1315, 2016 WL 1122884, *19. As a result, the inclusion of the prisoners unconstitutionally diluted the voting power of those in other County districts. Id. at 1323, 2016 WL 1122884, *26. The district court found that Calvin’s reasoning applied with full force to Cran-ston’s similar inclusion of the ACI prisoners in its Redistricting Plan. Davidson, — F.Supp.3d at -, 2016 WL 3008194, at *5i It granted summary judgment to the plaintiffs; entered declaratory judgment on behalf of the plaintiffs; enjoined the City, the City Council, and the City’s Board of Canvassers “from holding further elections under the current ward districting plan”; and ordered the City Council to “propose a [new] districting plan that complies with this Order” within thirty days. Id. at 11, 2011 WL 4346673, *6. The court vacated its declaratory judgment in a separate order. Cranston then filed this timely appeal. II. Before addressing the merits of the City’s appeal, we must ask whether we have jurisdiction to hear it. The parties agree that we have jurisdiction, but partly dispute its source and the resulting standard of review. The City argues that we have jurisdiction pursuant both to 28 U.S.C. § 1291, which grants us jurisdiction over “appeals from all final decisions of the district courts of the United States,” and to 28 U.S.C. ■§ 1292(a)(1), which grants us jurisdiction over “[interlocutory orders of the district courts of the United States ... granting, continuing, modifying, refusing or dissolving injunctions, or refusing to dissolve or modify injunctions, except where a direct review may be had in the Supreme Court.” The City argues that the district court’s “order was final for all practical purposes because, based on the District Court’s findings of fact and conclu *141 sions of law, the City’s apportionment was deemed unconstitutional,” and that accordingly the proper standard is the de novo review we give to “cross motions for summary judgment.” See Segrets, Inc. v. Gillman Knitwear Co., 207 F.3d 56, 61 (1st Cir. 2000) (standard of review for appeal from summary judgment is de novo). The plaintiffs argue that because “the district court entered but then vacated a judgment in a separate document .., there is no final judgment allowing appellate jurisdiction under 28 U.S.C. § 1291.” See Fed. R. Civ. P. 58(a). Instead, the plaintiffs contend that our only avenue for review is our jurisdiction over interlocutory appeals under § 1292. The plaintiffs contend that review under § 1292 triggers a deferential abuse of discretion standard. This disagreement is less significant than it seems. Whether or not the district court’s judgment could be treated as “final” for purposes of § 1291, we are satisfied that the injunctive relief the district court granted allows us to review Cranston’s appeal under § 1292. See Small v. Wageman, 291 F.2d 734, 735 n.1 (1st Cir. 1961) (“Since the appeal is from an order of the District ■ Court granting an injunction, this court clearly has appellate jurisdiction under Title 28 U.S.C. § 1292(a)(1).”). While the plaintiffs take the position that “[a]n appeal from interlocutory relief is reviewed for abuse of discretion,” this statement is overbroad. The cited cases actually stand for the narrower proposition that the abuse of discretion standard applies, to a district court’s decision on a request for a preliminary injunction. This appeal arises from a permanent injunction that necessarily rests upon an actual adjudication of the merits. That adjudication, in turn, was entered summarily under Rule 56, rather than after a trial. Cranston’s challenge is to the merits of the district court’s summary judgment ruling. 3 Accordingly, our review is de novo. See Domínguez-Cruz v. Suttle Caribe, Inc., 202 F.3d 424, 428 (1st Cir. 2000) (“Our review of the entry of summary judgment is de novo.”); Casas Office Machs., Inc. v. Mita Copystar Am., Inc., 42 F.3d 668, 678 (1st Cir. 1994) (stating, in an interlocutory appeal, that a permanent injunction could “stand only if the court properly awarded summary judgment,” and applying a de novo standard of review). Further, the issue presented is one of law, engendering - de novo review. Esso Standard Oil Co. v. Lopez-Freytes, 522 F.3d 136, 142 (1st Cir. 2008). III. It is • true that Evenwel did not decide the precise question before us. Nevertheless, we hold that its methodology and logic compel us to hold in favor of Cranston. Evenwel dictates that we look at constitutional history, precedent, and settled practice. 136 S.Ct. at 1126. Doing so leads us to find the inclusion of the ACI prisoners in Ward Six constitutionally permissible. In particular, Evenwel did not disturb Supreme Court precedent that apportionment claims involving only minor deviations normally require a showing of invidious discrimination, which has not even been alleged here. Without such a showing of discrimination, Evenwel reinforces that federal courts must give deference to decisions by local election authorities related to apportionment. Finally, the *142 Evenwel Court gave general approval to the use of total-population data from the Census in apportionment, which is what Cranston used here. Application of these principles requires us to uphold the constitutionality of Cranston’s decision to include the ACI inmates in Ward Six. Evenwel rejected an equal protection challenge to the use of total-population data from the 2010 Census to redraw Texas’s State Senate districts. Id. at 1125. The resulting districting map’s “maximum total-population deviation [among districts was] 8.04%, safely within the presumptively permissible 10% range.” Id.; see also Brown v. Thomson, 462 U.S. 835, 842-43, 103 S.Ct. 2690, 77 L.Ed.2d 214 (1983) (establishing ten-percent range). “But measured by a voter-population baseline — eligible voters or registered voters — the map’s maximum population deviation exceeded] 40%.” Evenwel, 136 S.Ct. at 1125. Plaintiffs from “Texas Senate districts ... with particularly large eligible- and registered-voter populations” sued, “[c]ontending that basing apportionment on total population dilute[d] their votes in relation to voters in other Senate districts, in violation of the one-person, one-vote principle of the Equal Protection Clause.” Id. The basic theory of them complaint was that the Constitution’s one-person, one-vote principle, first announced and applied to state governments in Reynolds v. Sims, 377 U.S. 533, 560-61, 84 S.Ct. 1362, 12 L.Ed.2d 506 (1964), and later applied to local governments in Avery v. Midland County, 390 U.S. 474, 484-86, 88 S.Ct. 1114, 20 L.Ed.2d 45 (1968), requires that political bodies use voter population, rather than total population, in apportionment. Evenwel, 136 S.Ct. at 1125-26. A three-judge district-court panel had dismissed the complaint, holding that “[djecisions of [the Supreme] Court ... permit jurisdictions to use any neutral, nondiscriminatory population baseline, including total population, when drawing state and local legislative districts.” Id. at 1126. The Supreme Court affirmed. The Court examined “constitutional history,” id. at 1127-30, precedent, id. at 1130-32, and “settled practice,” id. at 1132. After reviewing historical materials, the Court concluded that “it remains beyond doubt that the principle of representational equality figured prominently in the decision to count people, whether or not they qualify as voters.” Id. at 1129. Likewise, the Court’s precedents “suggested[ ] repeatedly[ ] that districting based on total population serves both the State’s interest in preventing vote dilution and its interest in ensuring equality of representation.” Id. at 1131. The Court observed that “[a]dopting voter-eligible apportionment as constitutional command would upset a well-functioning approach to districting that all 50 States and countless local jurisdictions have followed for decades, even centuries.” Id. at 1132. Moreover, the Court rejected the more general notion that nonvoters are not a significant population for apportionment: Nonvoters have an important stake in many policy debates — children, them parents, even their grandparents, for example, have a stake in a strong public-education system — and in receiving constituent services, such as help navigating public-benefits bureaucracies. By ensuring that each representative is subject to requests and suggestions from the same number of constituents, total-population apportionment promotes equitable and effective representation. Id. Several principles emerge from Even-wel. First, the Court did not disturb precedents holding that, where Reynolds’s requirements of population-based apportionment are met, a plaintiff usually must *143 show invidious discrimination to make out an apportionment claim under the Equal Protection Clause. The Court has made it clear elsewhere that: . Where the requirements of Reynolds v. Sims are met, apportionment schemes ... will constitute an invidious discrimination only if it can be shown that “designedly or otherwise, a multi-member constituency apportionment scheme, under the circumstances of a particular case, would operate to minimize or cancel out the voting strength of racial or political elements of the voting population.” Burns v. Richardson, 384 U.S. 73, 88, 86 S.Ct. 1286, 16 L.Ed.2d 376 (1966) (quoting Fortson v. Dorsey, 379 U.S. 433, 439, 85 S.Ct. 498, 13 L.Ed.2d 401 (1965)); see also Gaffney v. Cummings, 412 U.S. 735, 754, 93 S.Ct. 2321, 37 L.Ed.2d 298 (1973) (“As we have indicated, for example, multimem-ber districts may be vulnerable [to constitutional challenge], if racial or political groups have been fenced out of the political process and their voting strength invidiously minimized.”). Here, thére is no showing, nor even a claim, that either racial or political groups have been fenced out of the process in Cranston. The Court has further emphasized in the context of state-level apportionment that “minor deviations from mathematical equality among ... districts are insufficient to make out a prima facie case of invidious discrimination,” Brown, 462 U.S. at 842, 103 S.Ct. 2690 (quoting Gaffney, 412 U.S. at 745, 93 S.Ct. 2321), and that “as a general matter, ... an apportionment plan with a maximum population deviation under 10% falls within this category of minor deviations,” id. Again, the plaintiffs have made no claim that Cran-ston’s redistricting works any invidious discrimination. Second, Evenwel reinforced the principle established by earlier Supreme Court decisions that courts should give wide latitude to political decisions related to apportionment that work no invidious discrimination. It has long been constitutionally acceptable, but by no means required, to exclude non-voting persons such as “aliens, transients, short-term or temporary residents, or persons denied the vote for conviction of crime [from] the apportionment base,” Burns, 384 U.S. at 92, 86 S.Ct. 1286, so long as the apportionment scheme does not involve invidious discrimination, id. at 89, 86 S.Ct. 1286. The Court has repeatedly cautioned that such decisions, absent any showing of discrimination, “involve[ ] choices about the nature of representation with which we have been shown no constitutionally founded reason to interfere.” Id. at 92, 86 S.Ct. 1286; see also Brown, 462 U.S. at 847-48, 103 S.Ct. 2690 (“Particularly where there is no ‘taint of arbitrariness or discrimination,’ substantial deference is to be accorded the political decisions of the people of a State acting through their elected representatives.” (citation omitted) (quoting Roman v. Sincock, 377 U.S. 695, 710, 84 S.Ct. 1449, 12 L.Ed.2d 620 (1964))). Cranston’s Redistricting Plan, accordingly, is entitled to deference. Third, Evenwel approved the status quo of using total population from the Census for apportionment. 4 The Court underlined *144 that point when it observed that “in the overwhelming majority of cases, jurisdictions have equalized total population, as measured by the decennial census. Today, all States use total-population numbers from the census when designing congressional and state-legislative districts, and only seven States adjust those census numbers in any meaningful way.” 136 S.Ct. at 1124. The Court further noted that only four states (California, Delaware, Maryland, and New York) “exclude inmates who were domiciled out-of-state prior to incarceration.” Id. at 1124 n.3 (citing Cal. Elec. Code § 21003(a)(5); Del. Code Ann. tit. 29, § 804A; Md. Code Ann., State Gov’t § 2-2A-01; and N.Y. Legis. Law § 83-m(13)(b)). It is implausible that the Court would have observed that the majority of states use unadjusted total population (including , prisoners) from the Census for apportionment, upheld the constitutionality of apportionment by total population as a general proposition, and yet implied that the inclusion of prisoners in total .population for apportionment, without any showing of discrimination, is constitutionally suspect. The more natural reading of Ev-enwel is that the use of total population from the Census for apportionment is the constitutional default, but certain deviations are permissible, such as the exclusion of non-permanent residents, inmates, or non-citizen immigrants. See Evenwel, 136 S.Ct. at 1124 n.3, 1132-33; see also Burns, 384 U.S. at 92, 86 S.Ct. 1286 (recognizing that these decisions belong to state legislatures). Those permissible deviations are just that — optional, but not the norm. The norm, as practiced by the large majority of states, is to district based on total-population data from the Census, which includes prisoners. Evenwel held this approach to be “plainly permissible.” 136 S.Ct. at 1126. For the past half-century, Cranston has chosen to use this “plainly permissible” method for its districting. We also note the risks inherent in the plaintiffs’ theory. The district court’s reading of Evenwel would turn one of the arguably permissible adjustments to total-population data Evenwel described briefly in a footnote — the exclusion of prisoners with domiciles out of state — into a constitutional requirement even for in-state prisoners. Plaintiffs’ analysis invites federal courts to engage in what have long been recognized as paradigmatically political decisions, best left to local officials, about the inclusion of various categories of residents in the apportionment process. We decline that invitation. The decision whether to include or exclude the ACI prisoners in (Cranston’s apportionment is one for the political process. This conclusion becomes more obvious when one considers the unusual nature of the plaintiffs’ vote-dilution claim. The plaintiffs claim that the overwhelming majority of Cran-ston’s population (more than 66,000 of its approximately 80,000 residents) has incrementally diluted its own voting power by inflating the voting power of voters in Ward Six. That majority, which controls at least five of Cranston’s nine city councilors, may reverse its own decision if it sees fit. Under the logic of Evenwel and the Supreme Court’s earlier apportionment jurisprudence, Cranston’s 2012 Redistricting Plan easily passes constitutional muster. As mandated by the City’s charter, which mirrors the total-population apportionment required by Rhode Island’s constitution, the Redistricting Plan is based on total population from the Census. It contains a maximum population deviation of less than ten percent. And there is no evidence that *145 the . plan works any invidious discrimination. Cranston’s longstanding tradition of districting by total population based on the Census aligns with the practice of the large majority., of states,. which Evenwel endorsed. 5 ; The plaintiffs advance other arguments in support of affirmance, but they fare no better than the district court’s reading of Evenwel. The plaintiffs contend that the City’s Redistricting Plan violates the Equal Protection Clause because “Supreme Court precedent establishes that jurisdictions may not blindly or conclusively rely upon Census numbers when drawing their districts, but rather must look to the realities on the ground' when seeking to achieve representational equality.” But the cases they cite for this proposition involve facts easily distinguishable from those of this appeal. Plaintiffs cite to Evans v. Cornman, 398 U.S. 419, 90 S.Ct. 1752, 26 L.Ed.2d 370 (1970), which involved an unconstitutional exclusion from Maryland’s voting rolls of persons who lived on a federal enclave in Maryland. The case does not speak to a city’s inclusion of inmates residing in a state-operated prison facility located within the city. In Mahan v. Howell, 410 U.S. 315, 93 S.Ct. 979, 35 L.Ed.2d 320, modified in part, 411 U.S. 922, 93 S.Ct. 1475, 36 L.Ed.2d 316 (1973), the Virginia state legislature divided the cities of Virginia Beach and Norfolk into three districts each calculated to have an equal population and a single senatorial representative. In calculating the supposedly equal districts, however, the state relied on Census data-to assign to one district roughly 18,000 military personnel who lived in the' adjoining districts,- thereby diluting the votes of the military personnel. 6 - In those “unusual, if not- unique, circumstances,” the Court affirmed a decision requiring reapportionment in order to eliminate “discriminatory treatment” of the military personnel. Id. at 331-32, 93 S.Ct. 979. Here, plaintiffs advance no claim that the state legislature has diluted their votes in a manner that implicates a comparable consideration of discriminatory treatment. There has been no allegation that the Census has mistakenly assigned the ACI inmates to a place that was not their residence at the time the Census was conducted, nor has there been any allegation that the assignment resulted in “discriminatory treatment” of the inmates or any other party. 7 The inclusion of the prisoners in the 2010 Census data for the City affords a presumptively valid reason for including them in the City’s Redistricting Plan. Nothing argued by the plaintiffs or found by the district court casts doubt on that presumptive validity. Ultimately, the plaintiffs’ arguments and the reasoning of the district court are hard *146 to distinguish from the “voter population” argument rejected in Evenwel. This is most obvious when the plaintiffs and the district court emphasize that although the maximum deviation of total population in Cranston’s wards is less than ten percent when one' includes the ACI inmates in Ward Six, that figure leaps to thirty-five percent if one excludes the ACI prisoners in Ward Six. See Davidson, — F.Supp.3d at -, 2016 WL 3008194, at *1. Like the disparities in voter population complained of by the Evenwel plaintiffs, this argument begs the question. The Constitution does not require Cranston to exclude the ACI inmates from its apportionment process, and it gives the federal courts no power to interfere with Cranston’s decision to include them. IV. The order of the district court is reversed, and the case is remanded with instructions to enter summary judgment for the City of Cranston. Footnotes [1] To be clear, this case is not about the right of the ACI inmates to vote. Cranston argues that the district court "reached the erroneous legal conclusion that [state law] forbids ACI inmates from registering to vote in Cranston.” The City contends that, like other persons, the *139 ACI inmates may "establish a new voting domicile” in Cranston under the general standards set by Rhode Island law. See R.I. Const, art. II, § 1 ("Every citizen of the United States of the age of eighteen years or over who has had residence and home in this state for thirty days next preceding the time of voting, who has resided thirty days in the town or city from which such citizen desires to vote, and whose name shall be registered at least thirty days next preceding the time of voting as provided by law, shall have the right to vote for all offices to be elected...17 R.I. Gen. Laws § 17-1-3.1(a) (“The determinant of one’s domicile is that person's factual physical presence' in the voting district on a regular basis incorporating an intention to reside for an indefinite period.”); In re Op. of the Justices, 65 R.I. 451, 16 A.2d 331, 332 (1940). This contention, if true, would in turn cast doubt on the district court’s low estimate of eligible voters. We need not resolve this issue. Even assuming that the district court’s estimate of voter-eligible ACI inmates is accurate, its conclusion that the Constitution requires Cranston to exclude the ACI inmates still does not follow. [2] The plaintiffs' vote-dilution claim is distinct from a different claim advanced by amici curiae NAACP et al., who have argued that Cranston’s inclusion of the ACI prisoners in Ward Six is an example of "prison-based gerrymandering, which ... significantly and im-permissibly weakens the political power of communities of color" elsewhere in the state. This hypothetical claim is that the inclusion of minority ACI inmates in the Redistricting Plan dilutes the political power of the communities of color where the prisoners resided pre-incarceration. The claim is simply not advanced by the plaintiffs in this case, and we will not hear new claims from amici. See González-Droz v. González-Colón, 660 F.3d 1, 12 n.6 (1st Cir. 2011). To the degree amici advance a policy argument, such an argument should be addressed to the Rhode Island legislature. [3] Cranston also challenges the district court’s factual findings concerning the ACI inmates’ "representational nexus” to the City. Because we find Cranston's challenge to the district court’s legal analysis dispositive, we do not reach its challenge to the district court’s factual findings. [4] Several past directors of the United States Census Bureau have submitted an amicus brief arguing that “the usual residence rule and the Census Bureau’s determination of where to count persons for census purposes are not designed to answer the question of where those persons should be counted for redistricting purposes,” particularly in the case of incarcerated individuals. That may well be the case. But it does not follow that a town errs in itself relying on census data to *144 determine population in the absence of any unusual circumstances not presented here. [5] To our knowledge, the Supreme Court has never adopted a "representational nexus” analysis. ■ ■ [6] The Census data assigned all personnel to the location of their ship'berths. Id. at 330 n.11, 93 S.Ct. 979. [7] Likewise, in Hartung v. Bradbury, 332 Or. 570, 33 P.3d 972, 986-87 (2001) (en banc), the Oregon Supreme Court held that the Oregon Secretary of State had incorrectly relied on Census data that assigned the inmates of a federal prison- to the wrong Census block after the inaccuracy had been brought to his attention. The court’s holding on this challenge was articulated 'primarily1 on state-law grounds, and the court, "stress[ed] that this holding is limited to the particular circumstances of this' case. We do not suggest that the Secretary of State always must question census data or that the Secretary of State must investigate simply on the allegation that the Census Bureau made an error.” Id. at 987 n.26.