IN THE UNITED STATES DISTRICT COURT FOR THE NORTHERN DISTRICT OF ILLINOIS EASTERN DIVISION CHRISTINE RADOGNO, in her official capacity as Minority Leader of the Illinois Senate, THOMAS CROSS, in his official capacity as Minority Leader of the Illinois House of Representatives, ADAM BROWN, in his official capacity as a state representative from the 101st Representative District and individually as a registered voter, VERONICA VERA, CHLOE MOORE, JOE TREVINO, ANGEL GARCIA, ELIDIA MARES, EDWIN TOLENTINO, and THE ILLINOIS REPUBLICAN PARTY, ) ) ) ) ) ) ) ) ) ) ) ) Plaintiffs, ) ) vs ) ) ILLINOIS STATE BOARD OF ELECTIONS, ) RUPERT BORGSMILLER, Executive Director of ) the Illinois State Board of Elections, HAROLD D. ) BYERS, BRYAN A. SCHNEIDER, BETTY J. ) COFFRIN, ERNEST C. GOWEN, WILLIAM F. ) McGUFFAGE, JUDITH C. RICE, CHARLES W. ) SCHOLZ, JESSE R. SMART, all named in ) their official capacities as members of the Illinois ) State Board of Elections, AFRICAN ) AMERICANS FOR LEGISLATIVE ) REDISTRICTING, and LATINO COALITION ) FOR FAIR REPRESENTATION, ) ) Defendants. ) NO. 1:11-cv-04884 Judge Elaine E. Bucklo Judge Diane S. Sykes Judge Philip P. Simon MEMORANDUM OPINION AND ORDER On July 20, 2011, a mix of citizen-voters and Republican state legislators filed a complaint alleging that the legislative redistricting plan enacted by the Illinois legislature following the 2010 census violated various state and federal laws. A handful of interested parties intervened in the action; the complaint was twice amended; and we granted, in part, successive motions to dismiss. Now faced with defendants� motion for summary judgment of the two counts remaining in the operative iteration of the complaint�count 5, alleging that Representative District (�RD�) 96 is the product of racial gerrymandering in violation of the Equal Protection Clause, and count 1, alleging that RD 23 dilutes the votes of Latino voters in violation of § 2 of the Federal Voting Rights Act of 1965�we must decide whether plaintiffs are entitled to a trial on either claim. For the reasons that follow, we conclude that they are not and grant defendants� motion accordingly. Background Our previous decisions in this case set forth its background, which we restate in summary fashion here. Illinois has 59 Legislative (or �Senate�) districts and 118 Representative (or �House�) districts. Each Senate district is composed of two House districts. Pursuant to the state constitution, the boundaries of these districts are regularly reconfigured to reflect population data obtained in the decennial federal Census. In March, April, and May of this year, redistricting committees in the House and Senate held public hearings throughout the state to receive input from citizens and interested parties on the redistricting plan. Districts maps were proposed and amended in late May, and on June 3, 2011, the General Assembly passed P.A. 97-6, which established the new map of House and Senate districts. P.A. 97-6 adopted and incorporated House Resolution 385 (�HR 385") and Senate Resolution 249 (�SR 249") for the purpose of identifying legislative intent. Both HR 385 and SR 249 provide narrative discussions of each new district. These narratives generally identify the number of residents in each district, which in every case is said to achieve the �equal-population target� (in the words of the House), or the �ideal equal population target� (the phrase used by the Senate); explain population changes revealed by the 2010 Census; describe the shape and borders of the new district; and discuss demographic, social, economic, and partisan characteristics of the communities located within each, as well as the racial composition of 2 the district�s voting-age population (the �VAP�). The narratives also identify the political subdivisions that are encompassed, in whole or in part, within each district. In many instances, they also set forth the extent to which the �core� of the previous district is preserved in the new district. Representative District 96 Representative District 96 joins, for the first time, the downstate urban centers of Springfield and Decatur. This idea was first proposed by the African-Americans for Legislative Redistricting (�AALR�), a defendant-intervenor in this action. Lawrence Hill, a co-chair of AALR, testified that the purpose of creating the new RD 96 was to combine communities of interest in the urban areas of Springfield and Decatur. In his deposition, Mr. Hill testified that he identified �common factors� in those communities, including: Poor education outcomes, poor economic outcomes for the citizens. There is a stream of commerce between those two along I-72. They advertise in those same communities, the print media does and the radio. They worship along those communities. And when I say that, I say that there are people who are from Springfield who go to Decatur and vice versa for services. So they have a similar set of interests. They have a community of interests. Deposition of Lawrence Hill, Defendants� L.R. 56.1 Statement, Exh. E, at 42:20-43:4 [DE 73-8]. Mr. Hill further testified that he discussed the idea of joining portions of Springfield and Decatur into one district with Rev. Eric Jackson, a Decatur pastor who told Mr. Hill that some of his parishioners traveled from Springfield for services. Id. at 74:23-75:6. Mr. Hill also consulted Frank McNeil, a former Alderman in Springfield and community activist, who supported the concept of joining parts of Springfield and Decatur into one district. Id. at 55:6-56:7. Current Springfield Alderman Doris Turner testified to a joint House-Senate hearing expressing support for a district that would join �the eastern parts of Springfield, communities along Interstate 72, and western Decatur.� Transcript of May 24, 2011, joint hearing, Def.�s L.R. 56.1 3 Stmt., Exh. F, at 124 [DE 73-9]. She explained her position as follows: Currently both of these communities are urban areas that continue to be represented by individuals with a very rural perspective. This dilemma speaks to our most fundamental right, appropriate representation by our government.... When concerns are not understood and appropriately addressed [by our elected officials], it leads to the disenfranchisement of an entire community. And I believe that�s the situation currently being experienced by the individuals residing in Springfield and Decatur. I think that it should also be brought to your attention that currently these two cities are joined by media markets and co-sponsor events and a lot of other things that join them together in a very real way throughout the entire year. ... This House district will bring a new focus and attention to the many issues that continue to plague this community, among them an increase in violence and declining educational outcomes, and increase voter participation in the election process. Id. at 124-25, 126. Ultimately, the General Assembly passed a version of RD 96 that �closely resembles� AALR�s proposal. SR 249, Def.�s L.R. 56.1 Stmt., Exh. D, at 86 [DE 73-7].1 Timothy Mapes, the Speaker of the House�s Chief of Staff, was primarily responsible for drawing the district lines in the 2011 redistricting plan. Aff. of Timothy Mapes, Def.�s L.R. 56.1 Stmt., Exh. L, at ¶ 4 [DE 73-19]. Mr. Mapes testified that he was not the principal drafter of RD 96, but that he reviewed its boundaries before signing off on the final version of the map. Id. at ¶ 8. In his deposition, Mr. 1 In their responses to defendants� statements of facts, plaintiffs uniformly object to any reliance on SR 249 to support statements relating to RD 96 on the ground that the senate resolution describes LD 48, �a distinct geographic area� that is not coextensive with (but obviously encompasses) RD 96. While this objection could theoretically have merit, plaintiffs do not, in fact, dispute, much less do they identify evidence to controvert, the specific facts for which defendants draw support from SR 249�indeed, they sometimes admit that SR 249 �contains the facts� defendants assert with respect to RD 96. Because, as discussed further below, it is plaintiffs who bear the ultimate burden of persuasion, their unsupported speculation that the Senate�s factual findings regarding LD 48 might not apply to the specific area of RD 96 is insufficient to raise a genuine factual dispute. 4 Mapes recalled that RD 96 �contained an incumbent Republican House member� (Rep. Adam Brown, a plaintiff in this case), and explained that any changes he made to the proposed boundary lines of RD 96 reflected �the interest []in trying to find a competitive district looking at Democratic performing precincts to see and make sure all of those where possible could be contained in that district.� Mapes Dep., Pl.�s L.R. 56.1 Stmt., Exh. 5, at 66:22-67:10 [DE 88-5]. In pursuit of that interest, Mr. Mapes reviewed a map of RD 96 overlaid with the results of the �Democratic Index� created by staff to evaluate the partisan composition of the district by precincts. DE 73-19 at ¶ 8. Mr. Mapes did not review a map of RD 96 similarly overlaid with the results of census data reflecting the district�s racial demography before finalizing its boundaries. Id. A comparison of the partisan map, Def.�s L.R. 56.1 Stmt., Exh. M [DE 73-20], and the census-based map, Def.�s L.R. 56.1 Stmt., Exh. N [DE 73-21], shows a strong correlation between race and partisan preference. This is consistent with the testimony of defendants� expert Dr. Allan Lichtman, who testified to a joint House-Senate hearing that actual voter behavior reveals a �very substantial correlation� generally between African-American voters and preference for the Democratic candidate in general elections. Tr. of April 24, 2011 hearing, Def.�s L.R. 56.1 Stmt., Exh. B, at 83 [DE 73-5]. A graphic depiction of the boundaries of RD 96 is unremarkable in its shape. Its borders form a curved, vaguely rectangular strip with ends that taper at its northeast (Decatur) and northwest (Springfield) points. The maximum border-to-border distance in the district is 41.89 miles, and it is visibly one of the smallest downstate districts in terms of geographic size. Visual inspection of the 2011 redistricting map as a whole reveals that RD 96 is not nearly the longest, the widest, or the most irregularly shaped of Illinois� 118 representative districts. 5 The narrative discussions in the House and Senate resolutions describe the boundary lines outlining the borders of RD 96 and summarize the rationales behind them. With respect to the northern border of RD 96, HR 385 explains that �[t]o maintain a continuous district from east to west, many township lines and the Christian County border are utilized as the majority of the northern line of proposed RD 96.� HR 385, Def.�s L.R. 56.1 Stmt., Exh. C, at 277 [DE 73-6]. The resolutions further explain that Rochester and Rochester Township are split for equal-population purposes so that the Mid-Illinois Medical District in Springfield remains intact. Id. at 278; SR 249 at 84-86 [DE 73-7]. The House resolution states that the southern border of RD 96: goes east to west along the Pleasant View Township border and then follows Illinois Route 48 diagonally south. Further west, the southern border splits Taylorville Township to allow the vast majority of the non-rural parts of the city of Taylorville to remain in proposed RD 95, as they are in current RD 98, and cuts across southern South Fork Township where it meets the western border of proposed RD 96. In part because the boundaries of proposed RD 96 are predominantly along county and township boundaries, much of the southern boundary of proposed RD 96 is drawn in order to meet equal population. DE 73-6 at 279. The western border of the district �runs along the borders of South Fork and Cotton Hill townships and then moves into the city of Springfield and takes in the low-income areas of the city.� Id. at 280. The boundaries within Springfield follow precinct lines, �keeping neighborhoods together that are socioeconomically similar.� Id. at 278. In this conjunction, HR 385 explains: The majority of the territory in proposed RD includes socioeconomically similar residents, with much of the city of Springfield and city of Decatur in proposed RD 96 having median household incomes of less than $45,000. Rather than creating two representative districts with a significant portion of lower-income residents, proposed RD 96, by joining much of Decatur with the east side of Springfield, creates a representative district in which the needs and concerns of the lower-income residents can be better addressed by one representative. 6 Id. at 278-79. Along similar lines, HR 385 notes that although the boundaries in Decatur �are to a large extent based upon major roadways and Decatur�s municipal borders[,] [p]roposed RD 96 does not contain the more affluent areas of Decatur on the east and south sides of Lake Decatur.� Id. at 279. After describing the district�s boundary lines, HR 385 discusses several communities of interest within its borders. For example, HR 385 explains that: � �[Interstate]-72 provides a significant link between the communities of Springfield and Decatur. Many Decatur residents who work in state government live in Decatur and travel to work via I-72. Several trades and other businesses use the highway to conduct business back and forth between the two communities. I�72 is included in proposed RD 96 as much as possible, while still using existing local government boundaries as the northern border of proposed RD 96.� Id. at 280. � �With the seat of state government in Springfield, many state workers commute from Decatur to Springfield, and some Springfield residents work at the hospitals and manufacturing facilities in Decatur, creating a shared interest on those fronts. Additionally, a state prison is located in Decatur with many employees living in proposed RD 96. That, in addition to the fact that both Springfield and Decatur are also the government seats of Sangamon and Macon Counties, respectively, creates a vocal and active community of interest of government employees that are more effectively represented by one representative.� Id. at 281. � �The significant bodies of water in proposed RD 96 represent another community of interest. Proposed RD 96 includes most of Lake Springfield, Sangchris Lake, and Lake Decatur. As recreational and residential areas, these lakes and the people who enjoy them form a community of interest.� Id. This portion of the narrative concludes: Linking the residents of Decatur and eastern Springfield into one district is beneficial to those residents for a number of reasons. Both communities are central Illinois, urban population centers wtih a high percentage of African Americans. Under the current map, both of these communities are isolated and surrounded by rural farm communities with few minorities and have little in common with their neighbors. Under proposed RD 96, these two urban population centers are enjoined and are able to form a more influential community of interest. When considering economics, both areas have lower median incomes, requiring a different level of social services than their surrounding rural communities. Socially, the communities are linked by a major 7 highway, and many residents who live in Decatur work within state government in Springfield. In both Springfield and Decatur, many African American residents of one community have links to the other either through family, churches, or their employment. Id. at 281-82. HR 385's discussion of RD 96 concludes with a mention of how the district compares with others in terms of partisan advantage;2 notes that RD 96 has an African-American VAP of 24.87%, a Hispanic VAP of 1.73%, and an Asian VAP of 1.01% located primarily in the two urban centers of Springfield and Decatur; and observes that RD 96 �unites the only two significant African Americans (sic) communities of interest in the region.� Id. at 282. Analysis Summary judgment is appropriate when �the pleadings, depositions, answers to interrogatories, and admissions on file, together with the affidavits, if any, show that there is no genuine issue as to any material fact and that the moving party is entitled to a judgment as a matter of law.� Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 247 (1986). We construe all facts in the light most favorable to the plaintiffs and draw all justifiable inferences in the their favor. Id. at 255. In the context of racial gerrymandering claims, however, summary judgment review �must be understood in the context of the courts� traditional reluctance to interfere with the delicate and 2 The penultimate paragraph of the discussion of RD 96 states, �[p]roposed RD 96 contains a mixture of current RDs 87, 98, 99, 100, and 101 and has a lower partisan advantage than current RD 96, as well as current RD 101, which makes up a majority of proposed RD 96.� As best we can interpret the partisan advantage comparisons made here, the House seems to have found that RD 96 is less partisan than either District 96 or District 101 under the 2001 plan. The comparison to District 96 under the 2001 plan does not strike us as relevant to the claims in this case, since that district is located in the Chicago region and encompasses a different population entirely from the district now under Shaw scrutiny. And while HR 385's partisan comparison with District 101 under the previous plan is perhaps relevant to plaintiffs� Shaw claim, the comparison plainly falls short of supporting the only �undisputed� fact for which plaintiffs cite it: that RD 96 purportedly �disregards the partisan composition of the area.� Pl.�s L.R. 56.1 Stmt. ¶ 24 [DE 88]. 8 politically charged area of legislative redistricting.� Chen v. City of Houston, 206 F.3d 502, 505 (5th Cir. 2000). The difficulty in ascertaining the fine line between a legislature�s inevitable awareness of racial considerations and its impermissible motivation by them, �together with the sensitive nature of redistricting and the presumption of good faith that must be accorded to legislative enactments,� requires courts to exercise �extraordinary caution in adjudicating claims that a State has drawn district lines on the basis of race.� Miller v. Johnson, 515 U.S. 900, 916 (1995). Because plaintiffs ultimately must prove that racial considerations were not merely �a motivation for the drawing� of a district, but the �predominant factor,� Easley v. Cromartie, 532 U.S. 234, 241 (2001) (original emphasis) (internal quotations and citations omitted), a legislature�s �presentation of valid evidence of nonracial intent, which transforms the case into one of mixed motives, advances rather than hinders its case for summary judgment.� Chen, 206 F. 3d at 507. We now turn to the case before us. The implicit theory of plaintiffs� Shaw claim�though nowhere do they articulate it in so many words�is that defendants intentionally elevated race above all other considerations when drawing RD 96, �solely to effectuate the perceived common interests of one racial group.� Shaw v. Reno, 509 U.S. 630, 648 (1993). Such a purpose, the Shaw Court explained, is �altogether antithetical to our system of representative democracy� because it �reinforces racial stereotypes� and �signal[s] to elected officials that they represent a particular racial group rather than their constituency as a whole.� Id. at 648, 650. So it is, and so it does. But no �observant and informed analyst� of RD 96 could reasonably conclude that that was defendants� design here. Scott v. U.S. Dept. of Justice, 920 F. Supp. 1248, 1255 (M.D. Fla. 1996) (three-judge court), aff�d sub nom. Laywer, 521 U.S. 567 (1997). The most immediate problem we perceive with plaintiffs� theory is that it fails to see the 9 forest for the trees. Plaintiffs dismiss as an inconsequential detail the undisputed fact that the African-American VAP of RD 96 is a mere 24.87%, while the non-black voting population�over 70% of which is white�represents not just a majority but a super-majority of the district�s total VAP. But these demographics set the stage on which plaintiffs� Shaw claim must be played out, and they cannot be brushed aside as easily as plaintiffs would like. Defendants argue that the statistics belie plaintiffs� Shaw theory, insisting that �a state representative from RD 96 would be chancing political suicide� if he or she catered exclusively to the needs of the African-American voting bloc. Def.�s Reply, at 7 [DE 92]. It is indeed difficult to deny the logic of this argument and to reconcile the racial composition of RD 96 with plaintiffs� theory. The best plaintiffs can muster is the observation that the Supreme Court has not expressly held �the minority�s proportion of the overall VAP to be the dispositive factor� in a Shaw claim. Pl.�s Opp., 14 [DE 90]. True. But it is likewise true, as defendants submit, that the Court has never applied Shaw principles to invalidate a district in which the allegedly favored minority population does not represent a controlling electoral majority. Def.�s Mem., 9-10 [DE 71].3 Moreover, the Court�s analysis in Lawyer v. Department of Justice, 521 U.S. 567 (1997), explains why such districts are unlikely to pose the constitutional threat identified in Shaw. Rejecting a claim similar to the one raised here in that it challenged a non-minority-majority district (Florida House District 3 In what ultimately comes across as an ill-advised scramble to create the illusion of a dispute, plaintiffs quibble with defendants� assertion that �no court has ever invalidated a district under Shaw where the racial group allegedly targeted for success consisted of less than half of the district�s VAP.� Plaintiffs� quarrel points to the lower court decisions in Hunt v. Cromartie, 526 U.S. 541 (1999), and Easley, where less-than-majority minority districts were held to violate Shaw. But of course, the Supreme Court later reversed these decisions in the very cases plaintiffs cite. Accordingly, while defendants� formulation of the point may have been hyperbolic, plaintiffs� bone pick serves only to emphasize the overall weakness of their argument. 10 21) that the plaintiffs claimed �was drawn specifically to encompass members of minority groups with divergent interests residing in several different communities,� Scott, 920 F. Supp. at1250, the Lawyer Court noted, �[t]he fact that District 21 under Plan 386 is not a majority black district, the black voting-age population being 36.2%, supports the District Court's finding that the district is not a �safe� one for black-preferred candidates, but one that �offers to any candidate, without regard to race, the opportunity� to seek and be elected to office.� 521 U.S. at 581. See also Cano v. Davis, 211 F. Supp. 2d 1208, 1218 (C.D. Cal. 2002) (three-judge court), aff�d, 537 U.S. 1100 (2003) (rejecting Shaw challenges to districts in which no single racial group comprised more than 39.8% of the registered electorate, explaining that these demographics did not send the �pernicious� message with which Shaw was concerned). The racial composition of the voting-age population in RD 96 likewise does not suggest to candidates for office that his or her �primary obligation� is to act as the representative of African-Americans to the exclusion of the remainder of the electorate. But even if we set aside the problematic gestalt of this forest and examine the individual trees on which plaintiffs base their opposition to defendants� motion, their Shaw claim fares no better. Defendants cite abundant, uncontroverted evidence that the legislature did not elevate racial considerations above legitimate redistricting principles when drawing RD 96. To begin with, defendants point to evidence that partisan considerations played a central role in the creation of RD 96. It is undisputed, for example, that Timothy Mapes, who was responsible for approving RD 96 in its final form, analyzed the partisan demographics, but not the racial demographics, of the district, and that his objective was to create a district that was �competitive� for Democrats in the region. Plaintiffs� only answer to this is that defendants could have drawn an even more heavily Democratic district. But even if that is true, plaintiffs cite no authority to suggest 11 that the failure to maximize partisan advantage in RD 96 supports the inference that the legislature�s drawing of the district is �unexplainable on grounds other than race.� Easley, 532 U.S. at 242. Moreover, plaintiffs themselves acknowledged, in the Second Amended Complaint, that RD 96 �lowers the partisan advantage of the Republican voters within the district.� DE 65 at ¶ 193. Regardless of whether the legislature maximized Democratic partisan advantage, the evidence is undisputed that the mapmakers sought to (and plaintiffs� own allegations reflect that they did in fact), increase Democratic competitiveness in the district.4 Defendants find further support for the argument that racial considerations did not predominate over legitimate ones in evidence that the mapmakers could have, but did not, increase the African-American VAP in the district by 1) including additional African-American populations adjacent to the borders of RD 96, or 2) deviating from exact population equality by up to 10% without sacrificing the presumption of constitutionality. Defendants argue persuasively that the legislature�s decision to exclude from RD 96 precincts with sizeable African-American populations on the one hand, and to fix RD 96's population at the level required for exact population equality on the other, distinguishes it from districts the Court has subjected to strict scrutiny, and militates against the conclusion that race was not simply �a motivation� but the �predominant factor motivating the legislature�s districting decision.� Easley, 532 U.S. at 241 (2001). See Shaw, 509 U.S. at 635-36 (suspect district found to wind �in snakelike fashion...until it gobbles in enough enclaves of black neighborhoods�); Bush v. Vera, 517 U.S. 952, 973 (1996) (district lines contained �many narrow corridors, wings, or fingers [that] reach out to enclose black voters, while excluding nearby [non-black] residents.�). See also Miller v. Johnson, 515 U.S. 900, 908-09 (1995) 4 See also note 2, supra. 12 (legislature created a �monstrosity� that �swapped� black populations between cities and joined black neighborhoods that were �260 miles apart in distance and worlds apart in culture.�). Plaintiffs do not dispute that each of the districts created by P.A. 97-6, including RD 96, contains either 108,734 or 108,735 residents, or that RD 96 excludes sizeable African-American populations that could have been included in the district. Nor do they take issue with defendants� assertion that the legislature could have deviated from exact population equality by up to 10% without sacrificing prima facie constitutional validity. See Connor v. Finch, 431 U.S. 407, 417-18 (1977). In fact, plaintiffs offer no response at all to defendants� contention that these factors are inconsistent with the conclusion that the legislature considered race above all else when drawing RD 96. Moreover, despite plaintiffs� insistence that RD 96 resembles, in some esoteric fashion advanced by their expert (which we shall address shortly), the North Carolina district that was subject to scrutiny in Shaw, they concede that the mapmakers took affirmative steps to alleviate Shaw concerns that may have been raised by the AALR�s original proposal. Plaintiffs assert that the AALR proposal was similar to the district in Shaw �in that it snaked along the I-72 corridor then swelled on the east and west ends to gobble up the African-American communities in Springfield and Decatur,� but then go on to argue that the legislature modified that proposal by �extend[ing] its reach south of the I-72 corridor further into the rural portions between Springfield and Decatur in an obvious attempt to cover up the AALR proposal�s problematic comparison to the district [in] Shaw.� DE 90 at 10 (emphasis added). This argument enfeebles plaintiffs� claim several times over. To begin with, plaintiffs� failure to cite any evidence to support their ominous-sounding hypothesis about the legislature�s reason for extending the RD�s southern boundary smacks of speculation and of overzealous advocacy. But more importantly, the argument underscores two additional flaws in 13 plaintiffs� Shaw theory. First, it acknowledges that to whatever extent some earlier version of RD 96 may have resembled the district criticized in Shaw, the legislature sought to correct any constitutional flaws suggested by a comparison to that district. Second, it demonstrates that whatever the AALR�s purpose may have been in drafting its proposed RD 96, the legislature brought its own analysis to bear on the version it ultimately enacted. Plaintiffs rely heavily on the testimony of Lawrence Hill to support their claim that the legislature�s �sole motivation� in creating RD 96 was to join two African-American communities. DE 90 at 10. But it is undisputed that Mr. Hill was not a legislator, or a legislative staff member, or a decision maker of any sort with respect to the enactment of P.A. 97-6. Even assuming that his testimony, read in context, could be construed as expressing AALR�s �primary goal� of joining these two communities for race�s sake, as opposed to non-racial common interests (a construction we find difficult to reconcile with the evidence, but which we allow for present purposes), its probative value in ascertaining the primary goal of the legislature is slight, particularly when stacked against the unrefuted testimony of the principal drafter of the 2011 redistricting plan that partisan considerations, not race, were at the fore of decisions regarding RD 96; the legislature�s own, detailed explanation of its intent; and its receipt of testimony regarding the numerous, non-racial shared interests of the communities within the district, and the benefits of joining those communities in one district.5 5 Of course, some members of the public testified against RD 96. Plaintiffs cite Springfield Alderman Sam Cahnman, who testified that �it is possible to fit almost all of the city of Springfield in one House district,� and opined that �the citizens of Springfield should be kept together to the greatest extent possible� because of their �commonality and community of interest.� Tr. of April 25, 2011, hearing, Pl.�s L.R. 56.1 Stmt., Exh 9, at 72 [DE 88-9]. Jerry White, a local real estate agent and member of the Sangamon County Board of Review, also testified in favor of keeping Springfield and Sangamon County together, and stated his view that 14 And there is more in defendants� favor. Plaintiffs do not dispute defendants� evidence of a strong correlation between race and partisan preference, or that African-American voters strongly tend to prefer the Democratic party. This evidence makes plaintiffs� already substantial burden of proving that �race rather than politics predominantly explains� RD 96's boundaries even more difficult to meet. Easley, 532 U.S. at 243 (original emphasis). See also Bush, 517 U.S. at 968 (�If district lines merely correlate with race because they are drawn on the basis of political affiliation, which correlates with race, there is no racial classification to justify.�) Nor does the shape itself of RD 96�which, despite plaintiffs� attempt to shoehorn it into Shaw with descriptors such as �bizarre,� �erratic,� and �meandering,� is wholly unremarkable�suggest the machinations of insidious racial stereotyping. It is not a 160 mile long, snakelike district that winds through demographically disparate regions, gobbling in enclaves of the targeted population while avoiding others, as in Shaw. 509 U.S. at 635-36. Nor does it contain �narrow corridors, wings, or fingers� that �reach out to enclose black voters� as in Bush. 517 U.S. at 973. And although plaintiffs� expert opines, based on his analysis of where the population within RD 96's geographic area is concentrated and how the cities of Springfield and Decatur are �fractured,� that RD 96's �racial compactness� is similar to that of the North Carolina district considered in Shaw, he acknowledges that the Shaw district was �substantially less geographically �the city of Decatur is not at all like Springfield. It is more blue collar workers, a different kind of environment, Archer Daniels Midland.� Id. at 77-78. Defendants respond to certain of these statements (pointing out, for example, that Sangamon County was also split under the previous, 2001 redistricting plan, and that Springfield also �had multiple representatives� under that plan), but most salient is their observation that �unanimity cannot be the standard, or no map could survive a Shaw challenge.� DE 92 at 10. Indeed, plaintiffs� heavy burden presumably requires them to come forward with evidence more compelling than the fact that the legislature established RD 96 over the objection of some members of the public. 15 compact� than RD 96. Moreover, plaintiffs admit that RD 96's compactness varies only moderately from the mean for the redistricting plan as a whole, as measured by the Reock Measure and the Poslby-Popper measure�the two most commonly employed indicators.6 Meanwhile, plaintiffs fail to explain how their expert�s discussion of �racial compactness� (a concept that, from all the record reveals, is of his own vintage) fits into the legal framework governing their claim. Indeed, plaintiffs� only reference to their expert�s analysis is in support of their abstruse argument�buttressed by no legal authority�that �any measure of compactness must take into account how the district combines and divides individuals and communities based on political and demographic data.� DE 90 at 11. Whatever this may mean, it surely does not transform a district whose compactness is close to the plan mean, and whose borders follow many county or township lines, and in areas stretch for miles in entirely straight lines, into the kind of geographic �monstrosity� that is so far from compact and so illogical on its face that it rationally cannot be understood as anything other than an effort to segregate voters on the basis of race. We have not exhausted defendants� evidence and arguments in favor of summary judgment of count 5. We are convinced by the foregoing, however, that plaintiffs� evidence is not such as could persuade a reasonable fact finder that race was not merely �a motivation for the drawing� of RD 96, but the �predominant factor motivating the legislature�s districting decision.� Easley, 532 6 Plaintiffs dispute defendants� assertion that RD 96's compactness is also several orders of magnitude greater than the minimum �cut-off� proposed in a 1993 Michigan Law Review article commonly cited in conjunction with analyses of district compactness on the ground that the proposed �cut-off� is, in the opinion of their expert, �arbitrary.� But plaintiffs do not quarrel with defendants� assertions that RD 96's compactness measures, on a scale of 1.0 (most compact) to 0.0 (least compact), only .03 less (under the Reock Measure) or .05 less (under the PolsbyPopper measure) than the plan mean. These figures, far from raising eyebrows, demonstrate that RD 96's compactness reasonably approximates that of the plan as a whole. 16 U.S. at 241. We arrive at this conclusion untroubled by plaintiffs� insistence that intent is an inherently factual inquiry, or by their citation to Hunt v. Cromartie, 526 U.S. 541 (1999). In Hunt, the Court reversed the district court�s grant of summary judgment for the plaintiffs�the party with the ultimate burden of persuasion. 526 U.S. at 553. As the Fifth Circuit explained in Chen v. City of Houston, 206 F.3d 502 (5th Cir. 2000), the Hunt Court �took care to stress that summary judgment was inappropriate not only because of the factual nature of the intent requirement, but also because the nonmovant did not bear the burden of proof on the issue.� Chen, 206 F.3d at 506. The Chen court further noted Hunt�s reliance �on the traditional presumption that the legislature acted in good faith while districting.� Id. In this case, above and beyond the presumption to which the legislature is entitled, defendants have identified copious evidence that legitimate, non-racial considerations played a significant role in the General Assembly�s adoption of RD 96. Plaintiffs� circumstantial evidence and often misguided arguments raise no significant inference to the contrary. Representative District 23 Representative District 23, located in the greater Chicago area, has a Latino VAP comprising 46.27% of the district�s total voting-age population. Count 1 of the Second Amended Complaint asserts that RD 23 violates § 2 of the Voting Rights Act by diluting the votes of Latinos. DE 65 at ¶ 143. To prevail on count 1, plaintiffs must prove that, through some election procedure, practice, or standard, Latinos have been denied an equal opportunity to participate in the political process and to elect candidates of their choice in that particular district. 42 U.S.C. § 1973(b); Thornburg v. Gingles, 478 U.S. 30, 43 (1986). As a threshold matter, a plaintiff is required to prove three facts�known as the Gingles factors�before reaching the ultimate question of equal opportunity under 17 § 2. The plaintiff must establish that: 1) the minority group in question is sufficiently large and geographically compact to constitute a majority in a district; 2) this minority group is politically cohesive; and 3) whites vote sufficiently as a bloc usually to defeat the minority�s preferred candidate in that district. Gingles, 478 U.S. at 49-51. The failure to establish any of these �necessary preconditions� is fatal to a § 2 challenge. Johnson v. De Grandy, 512 U.S. 997, 1011 (1994). Defendants argue that they are entitled to summary judgment on count 1 because the report of plaintiffs� own expert, Dr. Liu, reveals that plaintiffs cannot meet the third Gingles prong as a matter of law. Defendants focus on Table 13 of Dr. Liu�s report, titled �Results of Elections in District 23 Involving a Hispanic Candidate, based on the Adopted Plan,� and reproduced below, which analyzes four elections within the area comprising RD 23: District 23 % Whites supported a White candidate % Hispanics supported a Hispanic candidate % Blacks supported a Black candidate Racially polarized Whites as the winning racial group 2008 St. Attorney 74* 81* 17 Yes Yes 2010 Lt. Gov 61* 51* 33 Yes Yes 2008 County Judicial 65* 82* 2010 County Judicial 87* 59* Overall 4/4 4/4 White Winner Yes Yes 6 Yes 4/4 18 Yes 3/4 1/4 Note: the * symbol denotes the elections where the majority of a racial group voted for a candidate of the same race There is no dispute that this table reveals, in the column entitled �White Winner,� that the white-preferred candidate actually lost three out of four races to the Latino candidate of choice. Under no plausible reading of Gingles, defendants insist, can these statistics be construed as establishing�or even as raising a triable issue of fact�that white-bloc voting usually defeats the minority�s preferred candidate. Plaintiffs do not dispute defendants� interpretation of Table 13. They insist, however, that the table, together with the unspecified �rest� of Dr. Liu�s report, actually shows that white-bloc voting does usually defeat Latino candidates in RD 23.7 But we conclude that their argument is flawed on nearly every front. The first problem, though not the most serious, is evidentiary: the portion of the record plaintiffs cite for their ultimate conclusion that �[w]hat Table 13 and the rest of Dr. Liu�s report actually show is that white voters vote sufficiently as a bloc to usually defeat Latino candidates in Representative District 23,� DE 90 at 15, is ¶ 15 of Dr. Liu�s Affidavit. DE 88-13. But that paragraph is devoted exclusively to an analysis of how white and Latino voters voted, and the turnout rates of the respective groups, in one electoral race�for �2010 House District 23." DE 88-13 at ¶ 14-15. It does not begin to support the broad proposition for which plaintiffs cite it. Moreover, we find merit to defendants� argument that election data relating to House District 23 as it existed in 2010 (i.e., under the previous map) is irrelevant to plaintiffs� claim against RD 23 under the 2011 7 Plaintiffs� undeveloped argument that defendants� focus on Table 13, rather than on the whole of Dr. Liu�s report, somehow �dooms� their motion rings particularly hollow in view of the fact that plaintiffs do not cite any other specific portion of the report as a basis for controverting defendants� interpretation of the data in Table 13. 19 map, as the two districts do not overlap substantially, if at all.8 But these issues are minor in comparison to the far more serious flaw in plaintiffs� argument, which is legal, not factual. Unable to ignore the actual results of the four elections selected for analysis by their own expert, which they admit reveal that Latino-preferred candidates in fact won three out of four times in RD 23, plaintiffs turn to a novel interpretation of Gingles prong 3. This interpretation would allow them to circumvent the actual election results highlighted by Dr. Liu, and to premise their claim instead on the showing that white candidates, as a group, would have won the elections, if �you aggregate all votes for a particular ethnicity.� But this analysis finds no support in Gingles or its progeny. In answer to defendants� observation that �neither whites nor any other race run as a group,� DE 71 at 21, plaintiffs cite Barnett v. City of Chicago, 969 F. Supp. 1359, 1424 (N.D. Ill. 1997) aff�d in part, vacated in part, 141 F.3d 699 (7th Cir. 1998), for the proposition that �failure to look at combined election data by ethnicity results in a misleading analysis.� But plaintiffs concede that Barnett �was looking at this question related to the issue of [Gingles] Prong 2.� DE 90 at 16. This distinction is not without import. While the second prong of Gingles is directed solely to the issue of whether minority voters vote cohesively in a district, the third prong comprises two distinct elements: first, whether the majority votes cohesively, and second, whether the historical effect of majority-bloc voting has been, �as a practical matter,� the usual defeat of minority-preferred 8 Defendants argue that the two districts share �not one inch of similar geography.� While they do not offer evidence to support this categorical claim, we may take judicial notice of maps of District 23 as it existed in 2010 and of RD 23, a comparison of which shows that if the previous District 23 overlaps at all with RD 23, the overlap is de minimis. Compare http://www.elections.il.gov/Downloads/VotingInformation/PDF/Districts/RepDist23.pdf. with http://www.precinctmaps.com/maps/GA/House2011/23rdHouse.pdf. 20 candidates. Jenkins v. Red Clay Consol. Sch. Dist. Bd. Of Educ., 4 F.3d 1103, 1123 (3th Cir. 1993). There is no dispute that Table 13 of Dr. Liu�s report shows that as a �practical,� rather than a theoretical, matter, Latino candidates have won more often than not in RD 23. Nor does plaintiffs� invocation of �special circumstances� compel the result they seek. Plaintiffs cite several cases to support the argument that elections in which multiple white candidates split the vote, with the result that minority candidates win with only a plurality, are among the kinds of �special circumstances� that the Gingles Court acknowledged could sometimes account for the success of a minority candidate even in a racially polarized contest. Gingles, 478 U.S. at 57. But even if plaintiffs are correct that the elections in Table 13 reflect special circumstances, not one of their cited cases provides any support for the manner in which plaintiffs seek to modify the analysis of whether the third Gingles factor has been established: by counting actual victories for the Latinopreferred candidate as if they were wins for a white candidate. Jenkins v. Manning, 116 F.3d 685, 694 (3d Cir. 1997) (special circumstances meant that victory for black candidate �still should be counted, although it should be accorded less weight�); Little Rock School Dist. v. Polaski County Special School Dist., 56 F.3d 904, 911 (8th Cir. 1995) (assuming, without deciding, that Gingles preconditions had been met, and considering �special circumstances� only in the context of the subsequent, �totality of the circumstances� portion of the § 2 analysis); Neal v. Coleburn, 689 F. Supp. 1426, 1436 (E.D. Va. 1988) (discounting election of black candidate elected by plurality vote where two white candidates split the white vote); U.S. v. Charleston County, S.C., 365 F.3d 341, 352 (4th Cir. 2004) (same). See also Red Clay, 4 F.3d 1103 at 1133 n. 31 (declining to decide �whether a court may discount, rather than entirely disregard, the results of elections accompanied by special circumstances.�). 21 We decline plaintiffs� invitation to be the first court ever to count actual Latino victories as putative white victories, and to conclude, on that basis, that Gingles� third prong has been met, despite uncontroverted evidence that Latino victories in fact outnumbered white victories three-toone in RD 23. We agree with defendants that this would turn the law on its head, and nothing in plaintiffs� analysis persuades us otherwise. Accordingly, even assuming that special circumstances account for all of the Latino victories in the elections Dr. Liu analyzed in Table 13, and disregard those victories entirely (rather than merely discount them), the most plaintiffs are left with is evidence that a white candidate defeated a Latino candidate in one election. This is insufficient to meet the third prong of Gingles. See Gingles, 478 U.S. 30, 57 (�Racial polarization should be seen as an attribute not of a single election, but rather of a polity viewed over time. The concern is necessarily temporal and the analysis historical because the evil to be avoided is the subordination of minority groups in American politics, not the defeat of individuals in particular electoral contests.�). See also Williams v. State Bd. of Elections, 718 F. Supp. 1324, 1328 (N.D. Ill. 1989) (construing �usually� as �more often than not,� and granting summary judgment for defendants where minority candidate had prevailed in fourteen out of twenty-one contested elections, including seven out of eleven �polarized� elections); Valladolid v. City of Nat�l City, 976 F.2d 1293, 1297-98 (9th Cir. 1992) (affirming summary judgment for defendants where �special circumstances� characterizing elections plaintiffs relied upon to support their claim left plaintiffs with no evidence to establish third Gingles requirement). Conclusion For the foregoing reasons, we grant defendants� motion for summary judgment in its entirety. SO ORDERED. 22 ENTERED: December 7, 2011 s/ Elaine E. Bucklo ELAINE E. BUCKLO, JUDGE UNITED STATES DISTRICT COURT NORTHERN DISTRICT OF ILLINOIS s/ Diane S. Sykes DIANE S. SYKES, JUDGE UNITED STATES COURT OF APPEALS SEVENTH CIRCUIT s/ Philip P. Simon PHILIP P. SIMON, CHIEF JUDGE UNITED STATES DISTRICT COURT NORTHERN DISTRICT OF INDIANA 23