Opinion
DOCKET # 11-CV-5632 (RR)(GEL)(DLI)(RLM)
03-19-2012
MARK A. FAVORS, HOWARD LEIB, LILLIE H. GALAN, EDWARD A. MULRAINE, WARREN SCHREIBER, and WEYMAN A. CAREY, Plaintiffs, DONNA KAYE DRAYTON, EDWIN ELLIS, AIDA FORREST, GENE A. JOHNSON, JOY WOOLLEY, SHEILA WRIGHT, LINDA LEE, SHING CHOR CHUNG, JULIA YANG, JUNG HO HONG, JUAN RAMOS, NICK CHAVARRIA, GRACIELA HEYMANN SANDRA MARTINEZ, EDWIN ROLDAN, MANOLIN TIRADO, LINDA ROSE, EVERET MILLS, ANTHONY HOFFMAN, KIM THOMPSON-WEREKOH, CARLOTTA BISHOP, CAROL RINZLER, GEORGE STAMATIADES, JOSEPHINE RODRIGUEZ, and SCOTT AUSTER, Intervenor-Plaintiffs, v. ANDREW M. CUOMO, as Governor of the State of New York, ROBERT J. DUFFY, as President of the Senate of the State of New York, DEAN G. SKELOS, as Majority Leader and President Pro Tempore of the Senate of the State of New York, SHELDON SILVER, as Speaker of the Assembly of the State of New York, JOHN L. SAMPSON, as Minority Leader of the Senate of the State of New York, BRIAN M. KOLB, as Minority Leader of the Assembly of the State of New York, NEW YORK STATE LEGISLATIVE TASK FORCE ON DEMOGRAPHIC RESEARCH AND REAPPORTIONMENT ("LATFOR"), JOHN J. McENENY, as Member of LATFOR, ROBERT OAKS, as Member of LATFOR, ROMAN HEDGES, as Member of LATFOR, MICHAEL F. NOZZOLIO, as Member of LATFOR, MARTIN MALAVE DILAN, as Member of LATFOR, and WELQUIS R. LOPEZ, as Member of LATFOR, Defendants.
OPINION AND ORDER
REENA RAGGI, United States Circuit Judge,
GERARD E. LYNCH, United States Circuit Judge,
DORA L. IRIZARRY, United States District Judge:
This three-judge court was convened on February 14, 2012, pursuant to 28 U.S.C. § 2284(a), to address plaintiffs' complaint that defendants' failure to redraw New York's state and federal congressional districts consistent with the results of the 2010 Census deprives them of the ability to vote in upcoming elections in accordance with rights guaranteed by the federal and state constitutions, see U.S. Const. art. I, § 2; N.Y. Const. art. III, §§ 4, 5, and the Voting Rights Act of 1965, see 42 U.S.C. §§ 1973-1973aa-6. Like the census that triggers it, this argument is now raised in federal courts at predictable ten-year intervals. See Rodriguez v. Pataki, No. 02-cv-618, 2002 WL 1058054 (S.D.N.Y. May 24, 2002); Puerto Rican Legal Def. & Educ. Fund, Inc. v. Gantt, 796 F. Supp. 681 (E.D.N.Y. 1992); Flateau v. Anderson, 537 F. Supp. 257 (S.D.N.Y. 1982). In the past, judicial creation of a congressional redistricting plan has spurred the New York legislature to produce its own plan just in time to avoid implementation of the judicial plan. See, e.g., Rodriguez v. Pataki, 308 F. Supp. 2d 346, 357-58 (S.D.N.Y.) (describing state legislature's enactment of congressional redistricting plan shortly after court adoption of special master plan), aff'd, 125 S. Ct. 627 (2004). This time is different. With less than 24 hours until the scheduled March 20, 2012 start of the petitioning process for the June 26, 2012 congressional primaries, the New York legislature has not delineated congressional districts for the state. Accordingly, the court declares New York to be without a congressional redistricting plan that conforms to the requirements of federal law, and it hereby orders defendants to implement the redistricting plan attached as Appendix 1 to this opinion ("Ordered Plan").
Plaintiffs Mark A. Favors, Howard Leib, Lillie H. Galan, Edward A. Mulraine, Warren Schreiber, and Weyman A. Carey ("Favors Plaintiffs") are registered voters in the State of New York; Leib is also a prospective State Senate candidate. Four sets of individuals intervened as plaintiffs pursuant to Fed. R. Civ. P. 24: (1) Donna Kaye Drayton, Edwin Ellis, Aida Forrest, Gene A. Johnson, Joy Woolley, and Sheila Wright ("Drayton Intervenors"); (2) Linda Lee, Shing Chor Chung, Julia Yang, and Jung Ho Hong ("Lee Intervenors"); (3) Juan Ramos, Nick Chavarria, Graciela Heymann, Sandra Martinez, Edwin Roldan, and Manolin Tirado ("Ramos Intervenors"); and (4) Linda Rose, Everet Mills, Anthony Hoffman, Kim Thompson-Werekoh, Carlotta Bishop, Carol Rinzler, George Stamatiades, Josephine Rodriguez, and Scott Auster ("Rose Intervenors").
Defendants, all sued in their official capacities, are Andrew M. Cuomo, as Governor of the State of New York; Eric T. Schneiderman, as Attorney General of the State of New York; Robert J. Duffy, as President of the State Senate; Dean G. Skelos, as Majority Leader and President Pro Tempore of the State Senate; Sheldon Silver, as Speaker of the State Assembly; John L. Sampson, as Minority Leader of the State Senate; Brian M. Kolb, as Minority Leader of the State Assembly; the New York State Legislative Task Force on Demographic Research and Reapportionment ("LATFOR"); John J. McEneny, as a member of LATFOR; Robert Oaks, as a member of LATFOR; Roman Hedges, as a member of LATFOR; Michael F. Nozzolio, as a member of LATFOR; Martin Malave Dilan, as a member of LATFOR; and Welquis R. Lopez, as a member of LATFOR.
Since filing, plaintiffs have withdrawn a claim that defendants' redistricting failure also violates the New York Prisoner Reallocation Law, see N.Y. Corr. Law § 71(8), making it unnecessary to discuss that claim further. Plaintiffs have also withdrawn their remaining claims against Attorney General Schneiderman.
Insofar as plaintiffs also challenge the defendants' failure to draw district lines for New York State Senate and Assembly districts, the parties are directed to appear before the court for a status conference on Wednesday, March 21, 2012, at 3:00 p.m. in the Ceremonial Courtroom to discuss what, if any, further proceedings are necessary in light of the enactment of a legislative redistricting plan for state offices on March 15, 2012.
I. The Undisputed Merits of Plaintiffs' Claim That New York Lacks a Constitutional Congressional Redistricting Plan
Defendants do not seriously dispute plaintiffs' claim that New York is without a constitutional congressional redistricting plan for the 2012 elections. Nor could they. As a result of the relative decline in New York's population reflected in the 2010 Census, the number of congressional districts allotted to the state is reduced from 29 to 27. See Kristin D. Burnett, U.S. Census Bureau, 2010 Census Briefs: Congressional Apportionment 2 (Nov. 2011), available at http://www.census.gov/prod/cen2010/briefs/c2010br-08.pdf; see generally U.S. Const. art. I, § 2, cl. 3; 2 U.S.C. § 2a. Thus, New York cannot operate under its existing congressional districting plan. Rather, it must redraw congressional district lines in order to have representatives seated in the 113th Congress. Further, the state must do so in a way that both (1) conforms to the constitutional mandate that "as nearly as is practicable one man's vote in a congressional election is to be worth as much as another's," Wesberry v. Sanders, 376 U.S. 1, 7-8 (1964), and (2) adheres to the constitutional prohibition against both intentional and excessive uses of race or ethnicity in redistricting, see Miller v. Johnson, 515 U.S. 900, 916 (1995) (prohibiting use of race or ethnicity as "predominant factor" motivating decision to place significant number of voters within or without particular district); City of Mobile v. Bolden, 446 U.S. 55, 66 (1980) (holding that redistricting cannot purposefully discriminate against racial group by diluting its vote). Federal law also obligates New York to effect redistricting consistent with the Voting Rights Act, particularly Section 2, which ensures against minority vote dilution, see 42 U.S.C. § 1973, and Section 5, which forbids retrogression in the electoral position of minorities in covered jurisdictions, see id. § 1973c, here including New York, Kings, and Bronx Counties, see 28 C.F.R., pt. 51, App.
To the extent some defendants' Answers dispute the merits of plaintiffs' claim, no defendant has pursued the challenge before this court, and we deem it waived by each defendant's failure to object to the magistrate judge's findings that New York currently has two too many districts, and that those districts are not equally apportioned in light of 2010 census data. See Report and Recommendation at 14 (Mar. 12, 2012), Dkt. Entry 223.
See 2 U.S.C. § 2a(c) (requiring at-large election of representatives if state has more districts than apportioned). Defendants do not propose to hold statewide congressional elections.
No such plan being in place, plaintiffs are entitled to both a declaratory judgment in their favor and relief in the form of a judicially ordered congressional redistricting plan.
II. The Ordered Redistricting Plan
In ordering defendants to implement the attached redistricting plan, we adopt the March 12, 2012 Report of Magistrate Judge Roanne L. Mann in its entirety and the redistricting plan recommended therein, see Report and Recommendation (Mar. 12, 2012), Dkt. Entry 223 ("Report" or "Recommended Plan"). The court's Ordered Plan modifies the Recommended Plan only to the extent noted in the margin. We write here to discuss the process and legal principles informing development of the Ordered Plan, and the court's reasons for rejecting certain objections or complaints about the Recommended Plan from parties and interested members of the public.
In response to submissions from the parties and the public, the court makes the following four changes to the Recommended Plan:
(1) The Brooklyn waterfront extending from the Brooklyn Bridge to the Brooklyn Battery Tunnel is placed in the same district, Ordered District 7, as it was under the existing plan. Two blocks in Sunset Park, Brooklyn, are then moved from Recommended District 7 to Ordered District 10 to achieve constitutionally mandated population equality.
(2) Wyoming County is united in Ordered District 27. The split of Ontario County, which was already split under the Recommended Plan, is reconfigured to ensure that Ordered District 23 achieves population equality.
(3) A zero population census block is moved from Recommended District 23 to Ordered District 27 to unify the Town of Canandaigua.
(4) A census block containing two people is moved from Recommended District 25 to Ordered District 27 to unify the Town of Hamlin. The split of the Town of Clarkson is shifted from the northeast to the southwest in order to achieve population equality.
Maps reflecting these changes are attached as Appendix 2 to the opinion.
A. The Process Employed To Develop the Ordered Plan
1. Defendants' Ripeness Challenge
Rather than challenge the merits of plaintiffs' claim before the three-judge panel or the magistrate judge, defendants questioned its ripeness, moving for dismissal on the ground that state inaction had not yet reached the point where a court could recognize a violation of federal law. We rejected this argument in an electronic order on February 21, 2012, supported by an opinion filed on March 8, 2012. See Order Denying Motions to Dismiss (Mar. 8, 2012), Dkt. Entry 219. With the petitioning process for the state's congressional elections set to begin on March 20, 2012, and with defendants conceding that no new congressional district plan was imminent, plaintiffs' claim was plainly ripe. Not only did the existing plan—providing for 29 congressional districts that do not comport with either the 2010 Census or the constitutional mandate of "one person, one vote"—clearly violate federal law, but also the court's ability to provide the necessary remedy, a constitutional redistricting plan, in time for the March 20 petitioning process faced significant time challenges. See id. at 7-14.
In this order, the court also summarily rejected defendants' challenge to plaintiffs' standing. See Order Denying Motions to Dismiss at 14-17 (Mar. 8, 2012), Dkt. Entry 219.
The March 20 date is a product of litigation in the Northern District of New York challenging New York's compliance with the Uniformed and Overseas Citizens Absentee Voting Act of 1986, see 42 U.S.C. §§ 1973ff-1973ff-7, as amended by the Military and Overseas Voter Empowerment Act, Pub. L. No. 111-84, subtitle H, §§ 575-589, 123 Stat. 2190, 2318-2335 (2009). On January 27, 2012, Judge Gary L. Sharpe ordered that New York's date for its non-presidential federal primary elections be moved to June 26, 2012, in order to bring the state into compliance. See United States v. New York, No. 1:10-cv-1214 (GLS/RFT), 2012 WL 254263, at *3 (N.D.N.Y. Jan. 27, 2012). Judge Sharpe then adjusted the rest of New York's election calendar accordingly, setting March 20, 2012, as the first day candidates may collect designating petitions in order to appear on the primary ballot. See United States v. New York, No. 1:10-cv-1214 (N.D.N.Y. Feb. 9, 2012), ECF No. 64 (adopting schedule including March 20, 2012 date for commencement of petition gathering); see also N.Y. Elec. Law § 6-134(4) (requiring designating petitions to be filed within 37-day period); id. § 6-158(4) (setting "eighth Thursday preceding the primary election" as deadline for filing designating petitions).
Defendants acknowledge that petitions must be obtained in the district that the prospective candidate seeks to represent. See Feb. 27, 2012 Hr'g Tr. at 31. Thus, districts must be delineated by the start of the petitioning process.
Under ideal circumstances, a court would require at least two months to devise a statewide redistricting plan, "one month for the drawing of the plan and an additional month for hearings and potential modifications to it." Nathaniel Persily, When Judges Carve Democracies: A Primer on Court-Drawn Redistricting Plans, 73 Geo. Wash. L. Rev. 1131, 1148 (2005). This court has had to develop the Ordered Plan in half that time.
2. The Magistrate Judge's Report
In order to provide a timely remedy, on February 27, the court referred the task of devising a recommended plan for redrawing New York's congressional districts to Magistrate Judge Mann with instructions to issue a report and recommendation to the court on March 12, 2012. The court further authorized the retention of Dr. Nathaniel Persily as a redistricting expert to assist Magistrate Judge Mann and this court in fashioning redistricting relief. See Order of Referral to Magistrate Judge at 3-4 (Feb. 28, 2012), Dkt. Entry 133.
A formal appointment order was entered on February 28, 2012. See Order of Referral to Magistrate Judge (Feb. 28, 2012), Dkt. Entry 133.
Dr. Persily, Charles Keller Beekman Professor of Law and Professor of Political Science at Columbia University School of Law, and the author of the article referenced in note 6, supra, is a well recognized expert on redistricting and voting rights. He served as a court-appointed expert in New York's 2002 redistricting, and has served in a similar capacity in connection with redistricting challenges in Connecticut, Georgia, and Maryland. See Aff. of Professor Nathaniel Persily, J.D., Ph.D., App. K (Mar. 12, 2012), Dkt. Entry 223, Attach. 12. Several parties urged and no party opposed Dr. Persily's retention in this case.
Exerting efforts that have been aptly characterized as "Herculean," Magistrate Judge Mann filed a detailed report and plan recommendation on March 12, supported by Dr. Persily's equally detailed affidavit and accompanying exhibits. The Report is remarkable in several respects. First, and most obviously, it provides this court in two weeks' time with what defendants have been unable—or unwilling—to provide New York State voters in more than a year: a redistricting plan for the state's congressional districts. In doing so, the Report cogently sets forth controlling principles of law, the challenging choices implicated in any redistricting assignment, and the magistrate judge's reasons for making the choices reflected in the Recommended Plan. Second, the Report discusses the commendable process employed by the magistrate judge to develop the Recommended Plan, which afforded the parties and interested members of the public frequent opportunities to be heard. See Report at 8-12. Third, the Report recommends a redistricting plan that is exemplary in satisfying each and every standard set forth in this court's referral order.
Mar. 15, 2012 Hr'g Tr. at 44 (statement of Jackson Chin, counsel for Ramos Intervenors); id. at 59 (statement of Richard Mancino, counsel for Favors Plaintiffs); Dominican American National Roundtable Objection at 3 (Mar. 13, 2012), Dkt. Entry 240, Ex. 7 ("DANR Objection").
3. Adoption of the Report and Recommended Plan
This court is nevertheless required to review the Recommended Plan de novo and to decide for itself what redistricting plan is necessary to ensure compliance with controlling law. See 28 U.S.C. § 636(b)(1); Fed. R. Civ. P. 53(f). Toward that end, we have carefully reviewed not only the magistrate judge's Report and Recommended Plan, but also all filings in the case, including all submitted plans (partial or statewide), as well as all objections and comments to the Recommended Plan presented by any party or interested person either in writing or at a public hearing conducted on March 15, 2012. The court has consulted further with Dr. Persily to ensure its full understanding of the demographics informing the Recommended Plan and implicated in changes urged by the parties or members of the public. Upon such careful and independent review, we adopt the magistrate judge's Report and Recommended Plan, with only minor and uncontroversial adjustments as indicated supra note 5.
Any suggestion that the magistrate judge failed to consider partial-plan submissions by the parties or public, see DANR Objection at 2-3, finds no support in the record, for reasons discussed infra at 39. In any event, this court has itself reviewed all plan submissions in reaching its final decision. In considering partial plans, the court, like the magistrate judge, is mindful that its task is not simply to accommodate the particular interest addressed in any plan but to effect a redistricting plan for the entire state that, first and foremost, satisfies the mandates of federal constitutional and statutory law.
The quality and fairness of the Recommended Plan is attested by the scarcity of objection from the highly engaged parties to this litigation, who include the entire executive Intervenors and the Senate Majority Defendants submitted objections to portions of the Report and Recommended Plan. See Drayton Intervenors' Letter (Mar. 14, 2012), Dkt. Entry 226; Ramos Intevenors' Letter (Mar. 14, 2012), Dkt. Entry 228; Senate Majority Defs.' Objections (Mar. 14, 2012), Dkt. Entry 231. The Lee and Rose Intervenors filed letters stating that they had no objections to the Report and Recommended Plan, see Lee Intervenors' Letter (Mar. 14, 2012), Dkt. Entry 227; Rose Intervenors' Response (Mar. 14, 2012), Dkt. Entry 229, and the Plaintiffs filed a memorandum of law in support of the Recommended Plan, see Pls.' Mem. in Support (Mar. 14, 2012), Dkt. Entry 232. The Governor, Assembly Majority, Senate Minority, and Assembly Minority Defendants filed no responses to the Report and Recommended Plan.
Although the magistrate judge and the court each invited members of the public to submit comments by easilyaccessible electronic means or in person at public hearings, only a small number of negative comments were received, and only a handful of New York's many elected officials suggested changes. The court received approximately twenty written submissions from the public supporting or opposing the Report and Recommended Plan. See Public Submissions to the Court (Mar. 19, 2012), Dkt. Entry 240. Approximately thirty members of the public spoke at the public hearing in support of or in opposition to the Recommended Plan. As set forth infra in Part II.B, only a small number of these comments opposing the plan require extended discussion.
4. Legal Principles Informing the Court's Decision
In ordering defendants to implement the court's redistricting plan, we are guided by principles of law that all parties agree are accurately set forth in the magistrate judge's Report. See Report at 13-19 (discussing constitutional and statutory redistricting standards);
and legislative leadership of New York State. Of the parties, only the Drayton and Ramos id. at 21-38 (discussing traditional redistricting factors). Because we adopt the Report, we do not repeat that discussion here. Nevertheless, in light of apparent confusion by some parties and members of the public at the March 15, 2012 hearing as to the court's obligations with respect to redistricting principles that impose legal mandates as compared with principles that afford some discretion, a preliminary discussion is useful to our subsequent analysis of specific objections.
a. The Constitutional Mandate of "One Person, One Vote"
At the first tier of redistricting analysis, the controlling principle is constitutional and mandatory: Article I, Section 2 requires that congressional election districts conform to the principle of "one person, one vote." Abrams v. Johnson, 521 U.S. 74, 98 (1997); Wesberry v. Sanders, 376 U.S. 1, 7-8 (1964); see also Reynolds v. Sims, 377 U.S. 533, 561-64, 568 (1964) (holding that state election districts must be apportioned by equal population). To satisfy this mandate, the population of each of New York's 27 new congressional districts must be within one person of the target number of 717,707 persons. See Report at 14; Aff. of Professor Nathaniel Persily, J.D., Ph.D. ¶¶ 103-04 (Mar. 12, 2008), Dkt. Entry 223, Attach. 1 ("Persily Aff."); see also Abrams v. Johnson, 521 U.S. at 98 (holding that "[a] court-ordered plan should ordinarily achieve the goal of population equality with little more than de minimis variation" (internal quotation marks omitted)). Thus, in considering arguments urging that persons be moved from one district to another in furtherance of one or more of the traditional redistricting factors discussed infra Part II.A.4.c, this court is constitutionally mandated to replace any persons moved with the same number of persons drawn from another district. As should be obvious, this means that, with the exception of changes that can be effected with a simple population swap between two districts, see, e.g., supra n.5 (discussing some such population swaps), most changes will trigger a ripple effect through multiple districts with serious constitutional implications for the entire redistricting plan.
b. Constitutional and Statutory Prohibition of Discriminatory
Redistricting
The second tier of redistricting analysis is of equal importance to the first in that it too is constitutional and mandatory: a redistricting plan cannot intentionally discriminate against a racial or ethnic group. See City of Mobile v. Bolden, 446 U.S. at 66; White v. Regester, 412 U.S. 755, 765-66 (1973); Gomillion v. Lightfoot, 364 U.S. 339, 341-42, 346 (1960). At the same time, race or ethnicity cannot be used as the "predominant factor" in deciding whether to put significant numbers of persons within or without a particular district. Miller v. Johnson, 515 U.S. at 916; see Shaw v. Reno, 509 U.S. 630, 642-49 (1993). The prohibition on discrimination is reinforced by the Voting Rights Act, which proscribes both minority vote dilution in Section 2, see 42 U.S.C. § 1973, and retrogression of existing minority strength in jurisdictions covered under Section 5, see id. § 1973c. Magistrate Judge Mann accurately discusses in some detail the law relevant to these principles, and thus, we do not repeat that discussion here. See Report at 13-19; see also Rodriguez v. Pataki, 2002 WL 1058054, at *4 (setting forth same legal standards in reviewing special master plan for 2002 New York redistricting). Further, the appendices attached to Dr. Persily's affidavit document the care taken in the Recommended Plan—and, now, the Ordered Plan—to safeguard against minority vote dilution and retrogression. See Persily Aff., Apps. A-J. Here too, then, we consider arguments urging this court to move persons from one district to another mindful of our obligation not to make any changes that could cause vote dilution or retrogression.
c. Traditional Redistricting Factors
At the third tier of redistricting analysis, a court considers, to the extent possible, traditional principles that generally inform legislative redistricting. This process contemplates the exercise of discretion.
In our referral to the magistrate judge, this court identified four traditional redistricting factors warranting consideration: (1) district compactness, (2) contiguity, (3) respect for political subdivisions, and (4) preservation of communities of interest. See Order of Referral to Magistrate Judge at 3 (Feb. 28, 2012), Dkt. Entry 133. At the same time, we stated that the magistrate judge's discretion to weigh these factors also extended to other factors that she might identify as reasonable, consistent with otherwise controlling law. See id. Two further traditional redistricting factors urged by the parties and members of the public are (5) maintaining the cores of existing districts and (6) protecting incumbency. See Karcher v. Daggett, 462 U.S. 725, 740 (1983) (recognizing these two factors as traditional considerations in legislative redistricting); Diaz v. Silver, 978 F. Supp. 96, 105, 123 (E.D.N.Y.) (stating that maintaining cores of existing districts is traditional redistricting factor, and that it is not improper for legislature to consider incumbency in enacting redistricting plan), aff'd, 522 U.S. 801 (1997).
To the extent the Senate Majority Defendants characterized the protection of core districts as the "primary" factor in redistricting analysis, Mar. 15, 2012 Hr'g Tr. at 16, 20, they sensibly withdrew that argument, conceding that traditional redistricting factors are necessarily subordinate to the two identified constitutional and statutory principles that must be satisfied by any redistricting plan, see id. at 20-21.
In fact, the magistrate judge's Report carefully addresses each of these six factors, discussing both the factor's relevancy to redistricting and the concerns associated with its application. See Report at 21-38. The magistrate judge accords some weight to each of these factors, with the single exception of incumbency protection. See id. In all respects pertaining to traditional redistricting factors, we adopt both the Report's reasoning and its conclusions as our own. We add only a few observations relevant to our future discussion of particular objections to the weight assigned to some of the traditional factors. These demonstrate that the noted traditional factors are not all of a kind.
First, only the first three traditional redistricting factors—compactness, contiguity, and respect for political subdivisions—can claim the sanction of enacted New York law. The State Constitution requires that State Senate and Assembly districts be contiguous and "in as compact form as practicable." N.Y. Const. art. III, §§ 4, 5 (emphasis added). It further requires that such districts consist of contiguous territory, and limits the division of counties, towns, and city blocks in forming state legislative districts. See id. Where state policy is thus reflected in law, there is sound reason for a court to accord that policy some weight, even when devising a redistricting plan for the federal rather than state legislature, subject, of course, to the superior demands of federal law, see U.S. Const. art. VI, cl. 2. As detailed in the magistrate judge's Report, the Recommended Plan is scrupulous in ensuring district contiguity; further, it achieves compactness and avoids splitting political subdivisions better than the existing plan. See Report at 23-24 (noting that Recommended Plan achieves compactness and contiguity in all districts, and splits six fewer counties and five fewer towns than existing plan); Persily Aff. ¶¶ 134-35 (stating that Recommended Plan keeps together 42 of 62 counties, and 894 of 970 towns, both of which improve on existing congressional districts). Indeed, the Ordered Plan has itself modified the Recommended Plan to unite one more county and three more towns, changes that could be effected without undue disruption to the overall plan's compliance with federal law. See supra n. 5. Thus, the Recommended Plan keeps 43 New York counties and 897 towns whole.
Second, the remaining factor identified in our referral order, the preservation of communities of interest, has no comparable pedigree in enacted state law. While the preservation of communities of interest has been recognized as "a legitimate goal in creating a district plan," Diaz v. Silver, 978 F. Supp. at 123, we observe, as did the magistrate judge, that this factor can more easily draw the court into political debates than factors such as compactness, contiguity, and respect for political subdivisions. These last three factors are more susceptible to neutral analysis, with a court's options and choices often evident on a map. But the identification of a "community of interest," a necessary first step to "preservation," requires insights that cannot be obtained from maps or even census figures. Such insights require an understanding of the community at issue, which can often be acquired only through direct and extensive experience with the day-to-day lives of an area's residents. Legislators are expected to have such understanding and experience. Judges are not. Thus, even if legislators routinely seek to preserve their constituents' communities of interest in a new redistricting plan, courts are understandably inclined to accord redistricting weight only to the preservation of obviously established and compact communities of interest. The Recommended Plan does this by respecting "certain widely recognized, geographically defined communities." Report at 27 (quoting Persily Aff. ¶ 137).
In noting that these factors are susceptible to objective evaluation, we do not suggest that they are necessarily easy to apply. See, e.g., Persily Aff. ¶ 130 & App. D (noting and applying eight different mathematical models to assess Recommended Districts' compactness).
Third, the remaining factors at issue—maintaining the cores of prior districts and incumbency protection—similarly risk drawing the court into political disputes. This is not to denigrate such factors, but simply to recognize that assigning them weight is a process that frequently requires political tradeoffs, a task for which the legislature—by virtue of insight, process, and ballot accountability—is better suited than the judiciary. Indeed, that conclusion is only reinforced here, where the court was obliged to create a congressional redistricting plan for as populous and diverse a state as New York in only a few weeks. To satisfy the mandates of federal law, the court's focus has necessarily been on census information. Nevertheless, like the magistrate judge, this court has made every effort also to consider the range of traditional redistricting factors, but it has done so cautiously, mindful of its obvious inability to acquire the sort of comprehensive insights that allow a legislature to balance the competing political concerns implicated in preserving various communities of interest, maintaining the cores of existing districts, and protecting incumbents.
To be sure, the legislature may itself wish to insulate its own redistricting efforts from complaints of partisanship by employing an independent commission. This is the subject of current debate in New York. See, e.g., Danny Hakim, John Eligon & Thomas Kaplan, Cuomo, Admitting Setbacks, Says He Asked for the Moon, N.Y. Times, Mar. 15, 2012, at A20 (discussing mixed reactions to compromise on state redistricting reform). We express no view on this question. We note only that, to the extent political concerns are implicit in certain redistricting factors that legislatures have permissibly employed in the past, a court may reasonably decide to accord those factors less weight than others better suited to neutral determination.
Of course, when the legislature itself weighs these factors in enacting a redistricting plan, a federal court reviewing that plan must confine itself to correcting error of federal law, without displacing otherwise "legitimate state policy judgments." Perry v. Perez, 132 S. Ct. 934, 941 (2012). That, however, is not this case. The court was obliged to create a new redistricting plan because the state has failed to do so. In these circumstances, the court owes no comparable deference to the outdated policy judgments of a now unconstitutional plan. See id. at 943 (stating that where there is "no recently enacted state plan," a court may be "compelled to design an interim map based on its own notion of the public good"). Indeed, when, as here, a new plan is required to contract the number of congressional districts, a court could only guess at which communities of interest the legislature would preserve or not, which districts the legislature would maintain or sacrifice, and which incumbents it would protect and which it would not.
Moreover, in creating a redistricting plan for a state that has none, a court's consideration of the traditional factors informing legislatively enacted plans does not require it to assign specific weight to any particular factor, or consistently to assign the same weight to a factor in delineating each district. Rather, the court must exercise discretion. Indeed, even if a court were generally inclined to accord significant weight to factors such as compactness or respect for political subdivisions, it might sometimes have to subordinate those factors to satisfy the population requirement of "one person, one vote," or to avoid proscribed minority voter dilution or retrogression. Similarly, a court may generally value the preservation of communities of interest while recognizing the particular challenges of "defining and accommodating" such communities of interest "in a region as diverse and dynamic as New York City and its environs." Report at 26. Further, as the court had repeated occasion to observe at the public hearing, the delineation of communities of interest is cabined by constitutional limits on racial stereotyping. See Miller v. Johnson, 515 U.S. at 920 (recognizing that communities of interest may be based on racial or ethnic makeup so long as "action is directed toward some common thread of relevant interests," but holding that, when a "State assumes from a group of voters' race that they 'think alike, share the same political interests, and will prefer the same candidates at the polls,' it engages in racial stereotyping at odds with equal protection mandates" (quoting Shaw v. Reno, 509 U.S. at 647)).
As Dr. Persily observed:
Respecting communities of interest is both an essential and slippery consideration in redistricting processes. In one respect, redistricting is about representation of communities. Communities that are split between districts often view their voice as diminished. In another respect, arguments based on communities of interest can often be pretexts for incumbency or partisan-related considerations. Moreover, community boundaries are inherently amorphous, contested, shifting and conflicting. By respecting one community's boundaries or some advocates' conception of their community, a redistricting plan might conflict with other advocates' conception of their community or with another community's boundaries.Report at 25 n.16 (quoting Persily Aff. ¶¶ 52-53).
Insofar as Magistrate Judge Mann assigned no weight to protecting incumbents, we do not understand her to have concluded that she was legally proscribed from considering this factor, but only that she opted to assign it no weight for reasons explained in her Report. We adopt this reasoning and exercise our own discretion in the same manner. We note only that, in creating a redistricting plan that eliminates two congressional districts, it is impossible to protect all incumbents, thus presenting the judiciary with political choices. See Report at 37-38. To insulate the Ordered Plan from any complaint of actual or apparent partisan bias, we choose to assign no weight to incumbency protection. In urging that we do otherwise, objecting parties and interested persons suggest that some incumbents' decisions not to stand for reelection may obviate the need for the court to make any such political choice. We are not persuaded. The present intentions of congressional incumbents to seek other offices or decline to run for reelection are always subject to change and possibly contingent on expectations about the districts the court might draw. In any event, those intentions are not part of any evidentiary record in this case, being known to the court at best through press reports or suggestions by parties and members of the public that are speculative and unreliable. Even in the absence of this evidentiary defect, the argument assumes that it would be possible to devise a redistricting plan that eliminated only districts represented by members not pursuing reelection while still satisfying all requirements of federal law. Our ability to devise such a plan is not apparent on the record. In any event, we are not inclined to upset the carefully crafted Recommended Plan on the eve of the petitioning process in order to further the most political of the traditional redistricting factors, particularly as there is no district residency requirement for congressional candidates. See id. at 36 (citing U.S. Const. art. I, § 2).
In reaching this conclusion, we do not overlook the importance of incumbent seniority. See Diaz v. Silver, 978 F. Supp. at 123 ("[T]he powerful role that seniority plays in the functioning of Congress makes incumbency an important and legitimate factor for a legislature to consider."). Like the magistrate judge, however, we do not think this concern requires that incumbency protection be given the weight urged in a court-devised plan. The legislature, of course, is free to give this factor greater weight in any redistricting plan it chooses to adopt.
Having reviewed the full record for ourselves and having carefully considered the objections and comments of all parties with respect to the Recommended Plan, we decide de novo to accord the identified traditional redistricting factors the same weight as the magistrate judge. See Report at 21-38. Any party or person disappointed with the weight assigned by this court to any of the traditional redistricting factors may certainly pursue this concern with their representatives in the New York legislature, which body has the authority to write a new plan for the next congressional election if it is dissatisfied with the Ordered Plan. Contrary to a concern expressed by one member of the public, nothing requires the Ordered Plan to remain in place for a decade. That result will occur only if the legislature chooses to leave the Ordered Plan in place.
We now address particular objections to the Recommended Plan that merit discussion.
B. Objections to the Recommended Plan
In discussing objections to the Recommended Plan, we start with those raised by the parties and proceed to others raised by public commentators.
1. Senate Majority Defendants
The Senate Majority Defendants object to the principles applied in creating the Recommended Plan and raise specific objections to particular Recommended Districts. See Senate Majority Defs.' Objections (Mar. 14, 2012), Dkt. Entry 231 ("Senate Majority Defs.' Objections"). Because we have already discussed general principles, here we address only specific objections and incorporate our previous discussion as necessary.
a. Recommended Districts 1, 2, and 3 (Long Island)
The Senate Majority Defendants complain that the Recommended Plan creates Long Island districts that do not "respect[] the cores of current districts and the communities of interest that have formed around them." Senate Majority Defs.' Objections at 4 (quoting Rodriguez v. Pataki, No. 02-Civ. 618 (RMB), 2002 WL 1058054, at *6 (S.D.N.Y. May 24, 2002) (internal quotation marks omitted)). They object to Recommended Districts 2 and 3, which are generally oriented from east to west to create districts along parts of the South and North Shores of Long Island, respectively, because, they contend, "Districts have traditionally run north to south across Long Island." Id.
As we have already observed, and as the Senate Majority Defendants concede, the preservation of elements of former districts so as to include in a newly formed district a large percentage of a previous district's population, see Persily Aff. ¶¶ 141-42, is not a stated policy of New York. See supra at 14-16; Mar. 15, 2012 Hr'g Tr. at 30. Nevertheless, because core preservation has been recognized as a legitimate consideration in a redistricting plan, see, e.g., Karcher v. Daggett, 462 U.S. at 740-41, a court may consider this factor, along with compactness, contiguity, respect for political subdivisions, and maintenance of communities of interest, in drawing district lines, see Abrams v. Johnson, 521 U.S. at 84, 98-100. Because the record plainly demonstrates such consideration by the magistrate judge, the Senate Majority Defendants do not contend that the Recommended Plan violates any legal principle. They simply urge the court to exercise its discretion to give core preservation greater weight in creating districts on Long Island. Having carefully considered their arguments, we conclude that the objection provides insufficient reason to alter the Recommended Plan, which seems to us better to balance the appropriate considerations.
As Professor Persily notes, only 7 of the 27 new districts fail to include at least half of a prior district's population, and nearly half (13) of the new districts contain at least 70% of a prior district's population. Persily Aff. ^ 142 & App. E. The question here is not whether core preservation is an appropriate factor (it is), or whether the Recommended Plan takes it into account (it does). The question is whether an objection based on that factor alone, without taking into account other counterbalancing factors, warrants rejecting this particular aspect of the Recommended Plan.
Inspection of the Senate Majority Defendants' own proposed Long Island districts reveals that they are significantly less compact than the Recommended Districts. Their proposed districts take highly unusual shapes. Indeed, it is far from clear that the Senate Majority Defendants' proposed districts run north to south across Long Island, or are even primarily north to south in orientation. Defendants' proposed District 2 forms a rough L shape; it is impossible to draw a north-south line through it that touches both shores. Their proposed District 3, meanwhile, covers all of the northern and southern shorelines of Nassau County (and parts of the southern shore of Suffolk) but narrows to an anemic mile-wide corridor from Hicksville to Greenvale. The Recommended Plan's more compact geography is sufficient reason in itself to reject the Senate Majority Defendants' objection. Moreover, districts that center on the northern and southern shores of Long Island—such as those of the Recommended Plan— follow widely understood social and geographical distinctions between Long Island's "North Shore" and "South Shore" communities and thereby reflect communities of interest that have no counterpart in the Senate Majority Defendants' plan. While defendants argue that preserving the cores of former districts is a more objectively measurable goal than recognizing communities of interest, the argument is not convincing with respect to well-established and geographically compact communities such as the North and South Shores of Long Island. In any event, for reasons already discussed, we think that deciding how much of a "core" to preserve, as against other redistricting considerations, is itself a highly subjective—and potentially partisan—endeavor. As with protecting incumbents, a factor to which we give no weight, judicial competence and neutrality signal caution in assigning considerable weight to factors best resolved by the political branches.Finally, the Senate Majority Defendants complain that Smithtown is split between Recommended Districts 1 and 2, and should instead be placed wholly in Recommended District 1. This warrants no change to the Recommended Plan. The Senate Majority Defendants have not offered a solution for how we would unite Smithtown's 80,000 residents into a single district, see Persily Aff., App. J at 23, a task that would require swapping thousands of people between Recommended Districts 1 and 2. Such a population swap would compromise both Recommended Districts' compactness and preservation of existing cores. We are especially reluctant to make such a change because, as is, Recommended District 1 maintains 96.81% of existing district 1, see Persily Aff., App. E, thus achieving the preservation that defendants separately urge for Recommended Districts 2 and 3.
Indeed, a glance at the map shows that the Senate Majority Defendants exaggerate the extent to which the existing districts are oriented in a simple north-south direction. Their shapes are considerably more complex, resembling an "L" or an inverted "T."
Indeed, it is not easy to distinguish core preservation from incumbent protection in this case. While the two goals may be distinguishable, it is telling that the Senate Majority Defendants themselves have previously equated them, see Skelos Defs.' Letter Regarding Incumbency (Feb. 29, 2012), Dkt. Entry 145 ("Preserving the cores of existing districts— sometimes also referred to as incumbency protection—is a well-established, traditional districting principle in New York.") (citations omitted)), and noteworthy that maps reflecting these defendants' proposed districts label them most prominently with the names of the incumbent representatives for whom they have apparently been designed. We are thus wary of giving "core preservation" the weight urged.
For these reasons, we conclude that this objection provides no persuasive reason to alter the district lines in the Recommended Plan.
b. Recommended District 5 (Long Island/Queens)
The Senate Majority Defendants raise a number of objections to Recommended District 5: (1) it pairs incumbents and fails to preserve cores of prior districts; (2) it fragments (along with several other districts) Jewish communities; and (3) it needlessly crosses the Nassau County border. The first objection is unpersuasive for reasons already fully discussed. The second objection is addressed in our discussion of specific issues raised by public commentators on behalf of the affected Jewish communities. See infra Part II.B.3.
With respect to the county border, the Senate Majority Defendants argue that crossing the county line renders Recommended District 5 non-compact. This contention is without merit. Some piercing of the Queens-Nassau line is necessary to achieve the constitutionally mandated population of 717,707 people. After the 2010 Census, Suffolk and Nassau counties together contain just below four districts' worth of population. See Persily Aff. ¶ 69. Thus, a breach of the Queens-Nassau border is mathematically inevitable. Indeed, the Senate Majority Defendants' own plan also breaches the border in its proposed District 4, though it does so farther north. We therefore reject this argument.
To the extent that defendants contend that the district is drawn along racial lines, we note that a substantial African American community straddles the county border in this area in an entirely compact unit, and that to divide that community, in preference to other ways of solving the inevitable breach of the county line, might well justly draw objection as a deliberate fragmentation of a compact minority community.
c. Remaining Objections
Because the Senate Majority Defendants' remaining arguments are less detailed, we consider them together.
First, persisting in their argument for preserving the cores of prior districts, defendants argue that Recommended Districts 8 and 11 should exchange the Marlboro Housing Development for Coney Island and a small part of Midwood. The specific plan offered by defendants to accomplish this change does not suggest an even population swap and, thus, is unconstitutional as proposed. Moreover, any change to move the Marlboro Housing Development would detract from the compactness of Recommended District 8 by creating a "finger" reaching up from Coney Island. Given the weakness of the core-preservation argument, and the difficulties with the defendants' proposed alternative, we decline to alter the Recommended Plan's entirely logical border between Districts 8 and 11.
Second, the Senate Majority Defendants argue that Recommended District 19, which covers much of the state's eastern border with Connecticut, Massachusetts, and Vermont, fails to preserve the core of existing district 20. But defendants' proposal for this district is far less compact than Recommended District 19, perhaps because their attempt to preserve the oddly shaped existing district 20 results in another oddly shaped district. As with Long Island's Recommended Districts 2 and 3, discussed supra in Part II.B.1.a, we place greater weight on compactness and, therefore, reject defendants' core-preservation objection to Recommended District 19.
Finally, the Senate Majority Defendants propose small changes to the boundaries of Recommended Districts 23, 25, and 27 to unite towns and counties in the western part of the state. These suggestions are logical and modest, and we have adopted them. See supra n.5.
2. Drayton and Ramos Intervenors
The Drayton and Ramos Intervenors urge us to remove the Brooklyn neighborhoods of Greenpoint and east and central Williamsburg from Recommended District 12 and to place them in Recommended District 7. They do not contend that such a change is necessary to unite a compact community of interest. They seek only to align the Greenpoint and east and central Williamsburg communities with other constituents who, the intervernors submit, share common socioeconomic features and, historically, congressional representation. See Drayton Intervenors' Letter (Mar. 14, 2012), Dkt. Entry 226; Ramos Intervenors' Letter (Mar. 14, 2012), Dkt. Entry 228. We decline to adopt this suggested change.
First, these intervenors offer no record evidence that the Greenpoint and Williamsburg communities have more in common with the communities of Recommended District 7 than with those of Recommended District 12. Although intervenors posit that Greenpoint and Williamsburg residents lack affinity with the distant wealthier enclaves of Manhattan in Recommended District 12, they do not contend that Greenpoint and Williamsburg residents lack commonality with other neighborhoods in the Recommended District, such as Long Island City and Astoria to their immediate north. Nor do these intervenors demonstrate Greenpoint and Williamsburg's commonality with distant neighborhoods they would join in Recommended District 7, such as Chinatown and Woodhaven. Insofar as intervenors express a need for any representative of Greenpoint and Williamsburg to be attentive to ongoing environmental clean-up projects in Newtown Creek, there is no record basis for concluding that such attention could not be secured if these neighborhoods were included in Recommended District 12, which spans both shores of the East River, the waterway into which the creek's pollution would continue to be dispersed if not remedied.
Second, the Drayton and Ramos Intervenors do not contend that their suggested change will improve the compactness, contiguity, or respect for political subdivisions of either Recommended District 7 or 12. Indeed, the urged change could well upset the balance of those traditional redistricting factors in surrounding districts, whose boundaries would have to be redrawn to accommodate transplanting Greenpoint's and Williamsburg's sizeable populations, representing tens of thousands of people, from Recommended District 12 to Recommended District 7.
Third, the transfer of such a sizeable population could require population adjustments to as many as six surrounding districts to maintain the equal populations mandated by the Constitution. Moreover, these surrounding districts are within jurisdictions covered under Section 5 of the Voting Rights Act and, in the case of Recommended Districts 8 and 9, are majority-minority districts under Section 2. Thus, moving Greenpoint and Williamsburg would entail a complex rearrangement of the surrounding districts to ensure that there is no retrogression in minority voting strength and that majority-minority districts are maintained. In these circumstances, we conclude that federal law mandates caution about adopting the urged discretionary change.
Because the totality of these concerns outweighs the arguments advanced by the Drayton and Ramos Intervenors, we maintain Recommended Districts 7 and 12 without change in the Ordered Plan.
3. Preserving a Jewish Community
Turning now to objections and comments from members of the public, we begin with a concern raised by the Orthodox Alliance for Liberty ("Orthodox Alliance"), which describes itself as "an alliance of Orthodox Jewish grass roots advocacy groups based in the New York City area." Orthodox Alliance Letter to District Court at 1 (Mar. 13, 2012), Dkt. Entry 240, Ex. 17. The Orthodox Alliance argues that congressional district lines, primarily but not exclusively in South Brooklyn, historically have caused Orthodox Jewish communities in the New York City area to be "egregiously" under-represented in Congress. Id. The Orthodox Alliance maintains that past plans have "divided into oblivion" and "gerrymandered into political irrelevance" these communities by splitting them among too many separate congressional districts, Orthodox Alliance Letter to Magistrate Judge Mann at 2 (Mar. 2, 2012), Dkt. Entry 222; and that the Recommended Plan "aggravates the injustice" by breaking up "the large Orthodox Jewish neighborhood of Flatbush" and "subsum[ing]" it "in neighboring African American communities," and by splitting Orthodox communities in Queens County and in Nassau County among multiple districts, Orthodox Letter to District Court at 2 (Mar. 13, 2012), Dkt. Entry 240, Ex. 17. Thus, the Orthodox Alliance submitted its own plan to the magistrate judge and now objects to the Recommended Plan for failing to adopt its proposal. It asks this court to redraw a number of congressional districts in Kings, Queens, and Nassau Counties to unify Orthodox Jewish neighborhoods into as few congressional districts as possible.
At the March 15 hearing, a representative of the Orthodox Alliance and several other members of the public spoke in support of the objection, or urged other changes to the Recommended Plan in furtherance of the Alliance's goal of unifying Orthodox Jewish communities within fewer congressional districts. In addition, two speakers argued that the court should also avoid splitting Russian Jewish neighborhoods across districts and splitting Russian Jewish neighborhoods from Orthodox Jewish neighborhoods. We have carefully considered all of these arguments as well as the Orthodox Alliance proposal, and we conclude that we cannot adopt the requested changes.
We have considerable sympathy for the concerns expressed by these commentators about the Recommended Plan and can easily take judicial notice of the fact that a number of distinctive Jewish communities exist in the areas in question, that these communities are significant and growing in population, and that, in some configuration, these communities (and particularly the Orthodox communities) share significant commonalities of interest. Nevertheless, as explained in more detail below, the record before us does not contain adequate information either to evaluate fully the objectors' claims, or to attempt to map the contours of the neighborhoods to which they refer with the precision necessary for redistricting purposes. Even assuming for the sake of argument the theoretical possibility of drawing lines on a map that would incorporate more of these communities into a contiguous district than does the Recommended Plan, the difficulties of doing so under the circumstances confronting us make it impossible either to adopt the plan proposed by the Orthodox Alliance or to revise the Recommended Plan ourselves in a way that would satisfy the concerns of these objectors and still be consistent with constitutional and statutory requirements.
First, and most fundamentally, the plan submitted by the Orthodox Alliance cannot be adopted because—as some of the plan's proponents conceded at the public hearing—its proposed districts vary too widely from the population numbers required to satisfy the Constitution's requirement of "one person, one vote." See, e.g., Mar. 15, 2012 Hr'g Tr. at 108-09, 154-55. Some speakers argued that the court could fix this defect in the Orthodox Alliance plan by various means. See, e.g., id. at 108-09. That turns out to be an all-but-impossible task under the conditions facing the court, for reasons we discuss in the following paragraph. In any event, the fact remains that the only actual plan presented by those raising the noted concerns is unconstitutional as submitted and, thus, must be rejected for that reason alone.
The Recommended Plan achieves populations of 717,707 or 717,708 in every district. The Orthodox Alliance plan, by contrast, would create districts with widely varying populations, from 707,204 in District 4 to 729,054 in District 8.
Second, while we have grouped together a number of commentators' concerns pertaining to the treatment of "Jewish communities," that characterization oversimplifies a number of distinct but overlapping concerns and masks significant differences in the problems presented by various objectors. Thus, while the Orthodox Alliance emphasizes the specific concerns of Orthodox Jewish communities, other speakers concentrated on the concerns of Russian Jewish communities in South Brooklyn. Still others proposed larger concerns about Jewish voters generally, and some commentators presented combinations of concerns. The proposed solutions also varied. One public official noted that the Orthodox population was insufficient to constitute a majority in any single district. He suggested that the Recommended Plan adequately concentrated what he characterized as about half of the Orthodox residents of South Brooklyn (those residing in Borough Park) in Recommended District 10, but he argued that more easterly residents were divided among several districts and could beneficially be unified. See id. at 183-86. Still other speakers proposed linking Orthodox communities with Russian Jewish communities; with other groups of ethnically or religiously Jewish residents in other parts of Queens, Brooklyn, or Nassau; or with other groups that purportedly share conservative political values with members of the Orthodox communities. These divergences suggest the difficulties, even under ideal conditions, of identifying and satisfying the particular concerns of the objectors.
Third, the record of this case is not in any event adequate for us to evaluate fully these concerns and determine whether a plan could be created that both satisfies them and complies with constitutional and statutory mandates. While we can take judicial notice of the existence and vibrancy of a number of Orthodox communities in Brooklyn and other counties, we can similarly take notice that, as with other neighborhoods in New York, even highly distinctive, concentrated, and cohesive population groupings most often exist cheek-by-jowl with other, quite different neighborhoods. Several Orthodox areas exist, close to and separated by communities of widely divergent natures. Census data—the primary concrete information available in the record for us to perform our task—does not reflect information about religion and, thus, does not permit us to identify whether a particular census block is part of an "Orthodox Jewish" neighborhood. Such precision is necessary, however, in the drawing of district lines. Its absence here is all the more noteworthy because, at the March 15 hearing, objectors presented different, and sometimes conflicting, descriptions of the locations, boundaries, and "cores" of Orthodox Jewish and Russian Jewish neighborhoods in the New York City area. As we have repeatedly emphasized, the knowledge required to identify the boundaries of a community of interest, and the ability to make essentially political decisions about how best to do so, reside in the legislature, not in a court required to operate by the most neutral principles it can devise, using often dry mathematical data. See supra Part II.A.4.c.
Indeed, because some of the Orthodox Jewish and Russian Jewish communities identified by the Orthodox Alliance and its supporters are geographically distant from each other, unifying these communities (particularly those outside of South Brooklyn) within single districts would create significant contiguity and compactness problems.
One objector argued that Orthodox neighborhoods could be identified from census data by noting areas in which there was a significant spread between the percentage of white residents and the percentage of voting age white residents, because Orthodox families have more children than other population groups. See Mar. 15, 2012 Hr'g Tr. at 135. Even if this sociological generalization is accurate—a proposition that we cannot test on the underlying record—it is obviously impractical to perform that kind of calculation under the constraints facing us.
To illustrate the divergence between professed local knowledge and the official data available to the court, one public commentator emphasized his view that the core of a cohesive Orthodox community was "Flatbush," but agreed that what he meant by the term is "not what's on the map as Flatbush." See Mar. 15, 2012 Hr'g Tr. at 88. Indeed, his definition is quite different from the area designated as Flatbush by New York City's Community District Profiles. See N.Y. City Dep't of City Planning, Community District Profiles (Brooklyn Community District 17), http://www.nyc.gov/html/dcp/html/lucds/cdstart.shtml (last visited Mar. 18, 2012). The speaker may well be right about local usage of the term within the Orthodox community, but it is not possible for the court to identify the contours of a neighborhood, or to consider objections to the Recommended Plan's treatment of that neighborhood, under these circumstances.
Finally, even if these formidable difficulties could somehow be overcome, the proposals before us, like other objections that focus specifically on the aspirations of individual communities and seek the creation of particularized districts tailored to their interests, fail to take into account the ripple effects entailed in configuring a district to meet their priorities. Some objectors contended that the Brooklyn communities that they want to unite are located in as many as five Recommended Districts. Accommodating the objectors could require completely reconfiguring not only those districts, but a number of others bordering them. In addition to presenting constitutional and statutory challenges, such widespread changes risk entirely unpredictable effects on residents satisfied with the Recommended Plan, who might have valid unanticipated objections to the reconfiguration. Like the magistrate judge, we have carefully considered the objections presented, and the implications of the Orthodox Alliance's proposals. We conclude that there is no way to adopt those and similar proposals while complying with the constitutional mandate of "one person, one vote," the Voting Rights Act, and the redistricting principles of contiguity and compactness. Accordingly, we do not adopt these objectors' proposals, and we conclude that, on the present record, the objections do not provide sufficient reason not to adopt the Recommended Plan.
4. Dominican American National Roundtable
The court received objections, through both written submissions and oral statements made at the March 15 hearing, from members of the Dominican American community requesting that the court revise the Recommended Plan to conform to a map proposed by non-party Dominican American National Roundtable ("DANR"), which creates a Hispanic majority district that snakes through northwestern Manhattan, northern and eastern sections of the Bronx, and southward into the Corona, Jackson Heights, Woodside neighborhoods of Queens (the "DANR Plan"). See DANR Objection (Mar. 13, 2012), Dkt. Entry 240, Ex. 7. DANR argues that this proposed district better unites Hispanic communities, whereas Recommended District 13 fuses Harlem and the rest of northern Manhattan with the Kingsbridge section of the Bronx, purportedly putting together disparate African American and Hispanic communities. DANR contends that the magistrate judge erred in not considering its partial plan and claims that the creation of the Recommended Plan was unduly rushed. DANR requests that the court either adopt its proposed district or remand this matter to the magistrate judge for further proceedings.
We understand and are sympathetic to the concerns animating support for the DANR Plan, in much the same way that we understand and are sympathetic to the concerns animating arguments by members of Jewish communities seeking to have district lines drawn to enhance their electoral voice. See supra Part II.B.3. Upon careful consideration of the DANR Plan, however, the court is not persuaded to adopt the proposal.
First, DANR's claim that the Recommended Plan effectively "cracks" the Hispanic community and dilutes its vote is not supported by the record. The Recommended Plan does not retrogress and, in fact, increases the Hispanic voting age population ("VAP") percentage from 43.8% in existing district 15 to 52.7% in Recommended District 13, thereby adding a second majority-minority Hispanic congressional district in New York State. See Persily Aff., App. C. At the same time, the Recommended Plan effectively maintains the percentage of Hispanic VAP in the single majority-minority Hispanic congressional district in the existing plan, see Persily Aff., App. C (showing 65.3% Hispanic VAP in existing district 16, and 64.3% Hispanic VAP in Recommended District 15), as well as in the existing Hispanic-plurality district from which DANR's proposed district would cull voters, see id. (showing 42.1% Hispanic VAP in existing district 7, and 45.0% Hispanic VAP in Recommended District 14). These facts belie DANR's claim that the Recommended Plan simply maintains the status quo and fails to increase the ability of a growing Hispanic community to obtain political representation, or, worse, dilutes the voting power of that community.
This increase in the proportion of Hispanic voters reflects the reality of the fast-changing demographics of New York City, New York State and, indeed, the country, where Hispanics are the fastest growing demographic group. See Michael Martinez & David Ariosto, Hispanic Population Exceeds 50 Million, Firmly Nation's No. 2 Group, available at http://articles.cnn.com/2011-03-24/us/census.hispanics_1_hispanic-population-illegal-immigration-foreign-born, Mar. 24, 2011 (last visited Mar. 18, 2012).
The Recommended Plan also maintains the percentage of Hispanic VAP in the remaining existing congressional district with a Hispanic plurality. See Persily Aff., App. C (showing 41.4% Hispanic VAP in existing district 12, and 41.5% Hispanic VAP in Recommended District 7).
Second, DANR's proposed district is barely contiguous and not compact, factors that militate against changes to a carefully crafted Recommended Plan that we find strengthens, rather than dilutes, the electoral voice of the Hispanic population. But beyond these concerns, DANR proposes a district so oddly shaped that, if we were to adopt it, an inference might arise that the court had segregated large numbers of Hispanic voters into a particular congressional district because of their nationality or ethnicity. See Miller v. Johnson, 515 U.S. at 916-17; Shaw v. Reno, 509 U.S. at 644-49. Indeed, statements made at the March 15 hearing in support of the DANR Plan could be construed to urge the creation of a "Dominican District." See, e.g., Mar. 15, 2012 Hr'g Tr. at 77-78; id. at 140-42. We ascribe no improper purpose to any of the speakers, each of whom disrupted his or her daily routine and waited patiently to address the court. We emphasize only our own constitutional obligation to protect against discrimination in redistricting while at the same time avoiding the use of race or ethnicity as "the predominant factor" in delineating a district. Miller v. Johnson, 515 U.S. at 916.
Third, DANR's proposed district would be carved out of a densely populated area and potentially traverse five Recommended Districts. This would require not only a reconfiguration of these five Recommended Districts, but of other adjacent districts as well. As we have noted with respect to other proposals for changes that would have sizeable ripple effects throughout the Recommended Plan, this presents the court with major constitutional challenges in ensuring that the affected districts contain equal populations and that the population mix of these districts avoids minority vote dilution or retrogression. Neither the DANR Plan nor its supporters have demonstrated to the court how this could be done.
DANR cannot minimize our concerns with its proposal by claiming that it was denied access to the redistricting process or that the magistrate judge only considered statewide plans. The record is clearly to the contrary. In its own letter objecting to the Report and Recommendation, DANR details all the submissions it made to the magistrate judge. See DANR Objection at 1 (Mar. 13, 2012), Dkt. Entry 240, Ex. 7. Further, at the March 5, 2012 hearing before the magistrate judge, DANR's counsel, in urging adoption of its proposed partial plan, thanked Magistrate Judge Mann for "making the process open to the public, to people who frequently don't have a chance to have their voices heard." Mar. 5, 2012 Hr'g Tr. at 116 (Mar. 8, 2012), Dkt. Entry 221. The magistrate judge's consideration of the DANR Plan is evident both in her Report and in the accompanying affidavit of Dr. Persily. See Report at 2 (incorporating Dr. Persily's affidavit); Persily Aff. 149, 159 (describing submission by Voting Rights for All, which included DANR proposed partial plan, and explaining that rejected suggestions "would likely entail violations of the Constitution or VRA," among other problems).
In any event, as noted supra n.12, the court has itself carefully reviewed all plans, whether statewide or partial, submitted by any party or member of the public. This includes the DANR Plan. In short, we have not categorically rejected any plan on the ground that it was partial. Rather, we have done our best to determine whether any statewide or partial plan identifies a remediable weakness in the Recommended Plan. Nevertheless, like the magistrate judge, we note that partial plans, by virtue of their narrower focus, frequently fail to consider the ripple effect of a discrete proposal on surrounding districts. The need to reconfigure adjoining districts often presents a variety of problems, some of constitutional or statutory dimension. DANR's partial plan presents such insurmountable problems, and it is for this reason that we do not adopt it.
Nor is there merit to DANR's claim that more time is needed to craft a fair and responsible redistricting plan. To be sure, the court has had to work on a punishing schedule to achieve the Ordered Plan, but the four judicial officers who have adhered to that schedule (along with Dr. Persily) have put in the hours necessary to ensure that every aspect of the plan received thorough and careful consideration. In any event, prompt implementation of the Ordered Plan is necessary to permit the 2012 congressional elections to go forward as scheduled in an orderly manner.
The concerns identified with DANR's proposed plan are significant. Meanwhile, we are not persuaded that the Recommended Plan fails to create districts fair to Hispanic voters. Accordingly, we reject DANR's proposal and adopt the Recommended Plan.
5. Objections to Recommended District 13: Harlem and the Bronx
Voting Rights for All, representing a coalition of elected officials and community advocates, including DANR, New York County Democratic Committee Leader Keith L.T. Wright, Assemblyman Carl E. Heastie, and Hazel M. Dukes, President of the NAACP New York State Conference, have objected in writing or at the March 15 hearing to Recommended District 13 on the ground that it would not adequately protect African American voters in existing district 15. See Voting Rights for All Objection (Mar. 14, 2012), Dkt. Entry 240, Ex. 19; Keith L.T. Wright Objection (Mar. 14, 2012), Dkt. Entry 240, Ex. 20; Carl E. Heastie Objection (Mar. 14, 2012), Dkt. Entry 240, Ex. 13; Mar. 15, 2012 Hr'g Tr. at 97-100 (statement of Hazel M. Dukes). Some urge the court to adopt a district proposed by Keith Wright (the "Wright Plan"), to extend the boundaries of existing district 15, which covers Harlem, from Manhattan to the western shore of the Bronx, across the northern Bronx, to the northeastern corner of Bronx County, as well as to two Queens housing projects near the Queensboro Bridge and into Westchester County. Like DANR's proposed district, the Wright Plan is oddly shaped and lacks compactness and contiguity. As such, it is inferior to Recommended District 13.
Recommended District 13 achieves population equality while avoiding dilution of African American voting strength by joining Harlem with East or "Spanish" Harlem and adjacent areas of the Bronx. See Persily Aff. ¶ 139, App. C. This results in a higher percentage of Hispanic voters in Recommended District 13 (52.7%) than had been in existing district 15 (43.8%). See id. ¶¶ 120-21, App. C. But it does not decrease African American voting strength in Recommended District 13. On the contrary, African Americans increase their voting population share in the district to 35.7%, compared to 34.1% in existing district 15, see id., even though the district had to gain 77,834 people in order to comply with the constitutional mandate of "one person, one vote," see Persily Aff. at 30, tbl. I. The Recommended Plan thus avoids retrogression with respect to the African American population while also increasing the Hispanic population share of the district. Moreover, in contrast to other proposals received by the court, the Recommended Plan achieves both of these goals without splitting Harlem between districts.
Accordingly, the court adopts Recommended District 13 without change.
6. Remaining Objections
We have considered the remaining objections by the parties and members of the public, and we conclude that none warrants any change to the Recommended Plan.
III. Conclusion
In the face of an outdated congressional districting plan, the application of which would plainly violate the requirements of federal law, and of the New York legislature's complete abdication of its congressional redistricting duty, this court is obliged not only to recognize a violation of law but also to create a new redistricting plan to ensure against the disenfranchisement of state voters in the 2012 congressional elections.
Judicial redistricting has correctly been described as an "unwelcome obligation." Conner v. Finch, 431 U.S. 407, 415 (1977); accord Perry v. Perez, 132 S. Ct. at 940; Puerto Rican Legal Def. & Educ. Fund, Inc. v. Gantt, 796 F. Supp. at 684. That is particularly so in this case because of the extremely limited time frame within which the court has had to create the Ordered Plan. But the task is unwelcome for reasons that go beyond the practical to implicate the proper division of power within our federal republic. While congressional district lines must always be drawn to conform to federal law, the power to draw such lines is committed in the first instance to the states, not to the federal government, and is properly exercised by the most democratic branch of state government, the legislature. See U.S. Const. art. I, § 2. Indeed, when such power is duly exercised, a federal court's review authority is limited to ensuring an enacted plan's compliance with federal law, and will not extend to state policy choices. See Perry v. Perez, 132 S. Ct. at 941. But when, as here, a state completely abdicates its congressional redistricting duties, it effectively cedes state power to the federal government. Further, it transfers power that should be exercised by democratic bodies to a judiciary ill equipped to resolve competing policy arguments. Such a twin recalibration of important power balances in a federal republic is itself "unwelcome."
In prior redistricting challenges, New York has avoided such a wholesale transfer of state legislative power to the federal courts through last-minute enactments of new redistricting plans. In this case, however, New York has been willing to let even the last minute pass and to abdicate the whole of its redistricting power to a reluctant federal court. Confronted with this unwelcome failure of state government, and consistent with its obligations under federal law, the court hereby
(1) GRANTS judgment in favor of plaintiffs, declaring that New York's existing plan for delineating congressional districts fails to comply with the Constitution; and accordingly,
(2) ADOPTS the March 12, 2012 Report of Magistrate Judge Mann in its entirety and the Recommended Plan referenced in the Report with the changes indicated in note 5 of this opinion and reflected in the attached Ordered Plan;
(3) ORDERS defendants promptly (a) to implement the attached Ordered Plan so that the 2012 elections for the House of Representatives can go forward as scheduled, and (b) to confirm to the court in writing, on or before noon on Wednesday, March 21, 2012, that they have taken the steps necessary to do so; and
The parties—as well as interested members of the public—may access the block equivalency data files through a link that will be provided on the court's public website, located at www.nyed.uscourts.gov, or at the following two addresses: https://www.nyed.uscourts.gov/pub/docs/cv/11-5632/panel/final/FINAL%20COURT%20 PLAN.txt; https://www.nyed.uscourts.gov/pub/docs/cv/! 1 - 5632/panel/final/ FINAL%20COURT%20PLAN.DCC.
--------
(4) DIRECTS all parties to appear before the court for a status conference on Wednesday, March 21, 2012, at 3:00 p.m., in the Ceremonial Courtroom on the second floor of the Brooklyn Federal Courthouse, located at 225 Cadman Plaza East.
SO ORDERED.
DATED: Brooklyn, New York
March 19, 2012
________________
REENA RAGGI
United States Circuit Judge
________________
GERARD E. LYNCH
United States Circuit Judge
________________
DORA L. IRIZARRY
United States District Judge