Cano v. Davis

211 F. Supp. 2d 1208 (2002)

Maria CANO, et al., Plaintiffs,
v.
Gray DAVIS, et al., Defendants.

No. CV 01-08477 MMM (RCx).

United States District Court, C.D. California.

June 12, 2002.

Vibiana Andrade, Thomas A. Saenz, Antonia Hernandez, Steven J. Reyes, Mexican American Legal Defense & Education Fund, Los Angeles, CA, Denise M. Hulett, Mexican American Legal Defense and Educational Fund, San Francisco, CA, Maria Blanco, Mexican American Legal Defense and Educational Fund, Sacramento, CA, for Plaintiffs.

Otto I. Pena, Otto I. Pena Law Offices, Los Angeles, CA, for Salvadoran Ass'n of Los Angeles and Luis R. Reyes.

Gail H. Morse, Jenner & Block, Chicago, IL, for Intervenors.

Louis R. Mauro, Jennifer Kathryn Rockwell, CAAG—Office of Attorney General of California Correctional Law Section, Sacramento, CA, for Gary Davis, Cruz Bustamante, and John Burton.

Bruce A. Wessel, Jonathan H. Steinberg, Elliot N. Brown, Laura W. Brill, Irell & Manella, Los Angeles, CA, Thomas A. Gauthier, Lance Herman Olson, George Waters, Olson, Hagel, Waters & Fishburn, Sacramento, CA, for John Burton.

Robin B. Johansen, Miguel Marquez, Kathleen J. Purcell, Joseph Remcho, Remcho, Johansen & Purcell, San Leandro, CA, for Herb Wesson, Jr.

Holger G. Besch, Jones, Day, Reavis & Pogue, Los Angeles, CA, Andrew McBride, Wiley, Rein & Fielding, Washington, DC, Michael A. Carvin, Jones, Day, Reavis & Pogue, Washington, DC, Charles, H. Bell, Bell McAndrews, Sacramento, CA, for Intervenor-Defendants.

Eric R. Wiesel, Donald H. Heller, Donald H. Heller Law Offices, Sacramento, CA, for James L. Brulte, Dick Anderson, Bob Margett, and California State Senate Republican Caucus.

Bienclaro M. Llaneta, Ben M. Llaneta Law Offices, San Diego, CA, for Filipino World War II Veterans Federation of San Diego County.

Wayne M. Barsky, Dohoang Thien Duong, Gibson, Dunn & Crutcher, Los Angeles, CA, for Guatemalan Unity Information Agency, El Rescate, Alianza Hondureno De Los Angeles, and Public Immigrants Policy Institute of Los Angeles.

Before REINHARDT, Circuit Judge, SNYDER and MORROW, District Judges.

OPINION AND ORDER GRANTING DEFENDANTS' MOTIONS FOR SUMMARY JUDGMENT

PER CURIAM.

Plaintiffs, who are Latino voters and advocacy groups, challenge two congressional districts and one state legislative district adopted by the State of California as part of its most recent redistricting process. Plaintiffs contend that the three districts violate the Voting Rights Act of 1965, as amended, 42 U.S.C. § 1973, and the Equal Protection Clause of the Fourteenth Amendment, under various legal theories. These theories raise challenging questions regarding the applicability of voting rights doctrines developed in a fundamentally different context than the rapidly-changing multi-racial and multi-ethnic community that is present-day Southern California. We conclude, with respect to each of the three districts at issue in this case, that the legislature permissibly exercised its broad discretion to draw district lines, and that, in doing so, it violated neither the Voting Rights Act nor the Constitution. Accordingly, we grant in full defendants' and defendants-intervenors' motions for summary judgment.

BACKGROUND

Factual and Legal Background

In the summer of 2001, California engaged in the decennial task of redrawing its state and federal legislative districts to account for population changes documented in the 2000 census. Reynolds v. Sims, 377 U.S. 533, 583-84, 84 S. Ct. 1362, 12 L. Ed. 2d 506, (1964) (requiring periodic redistricting); CAL. CONST. Art. XXI, § 1 (requiring legislative redistricting in the year following the national census). As a result of California's population growth relative to that of other states, the census disclosed that the state was entitled to one additional congressional seat. Accordingly, the legislature was charged with drawing 53 congressional districts, as well as the unchanged number of State Senate and Assembly districts prescribed by state law, 40 and 80 respectively. California requires that the districts in each of these categories be compact and contiguous, that they respect as much as possible the integrity of the boundary lines of political subdivisions such as cities and counties, and that they preserve other local communities of interest. CAL. CONST. Art. XXI, § 1; Wilson v. Eu, 1 Cal. 4th 707, 4 Cal. Rptr. 2d 379, 823 P.2d 545, 552-53 (1992).

Over the course of the summer of 2001, the legislature conducted public hearings throughout the state at which various public interest groups and others articulated general concerns regarding the redistricting process, and proposed specific changes to the district lines that had been drawn after the 1990 census. Among those commenting was the highly-influential Mexican-American Legal Defense and Educational Fund ("MALDEF"), which serves as counsel to plaintiffs in this case.^[1] At least two of the organizational plaintiffs in this case, as well as other Latino advocacy groups, also testified at the various hearings on redistricting. On August 31, 2001, the Senate Elections and Reapportionment Committee promulgated a proposed redistricting plan for State Senate and Congressional Districts. The principal map-drawing consultant retained by the Senate was Michael Berman, a fixture in California redistricting processes and the brother of Congressman Howard Berman, an incumbent in one of the congressional districts at issue in this case. It is alleged that Michael Berman was also retained by members of the California congressional delegation, many of whom paid his organization $20,000 a person to represent their interests in the redistricting process.^[2]

A two-day legislative hearing was held on September 4-5 in Sacramento, with video-conferencing in other sites around the state, at which the Assembly and Senate redistricting committees heard public comment on the draft plan. Latino groups, among others, testified regarding the proposed Congressional and State Senate Districts. Some groups, including the Latino advocacy groups mentioned above, suggested changes; certain of these were adopted. Both houses of the California Legislature voted to enact the second version of the plan: the Senate voted 38-2 in favor of the bill on September 12, 2001; the next day the Assembly approved it by a margin of 70-10.^[3] Twenty-three of the twenty-six Latino state legislators voted for the measure, including every Latino State Senator.

Governor Davis signed the redistricting plan into law on September 27, 2001. Plaintiffs filed this action several days thereafter, and this three-judge district court was convened pursuant to 28 U.S.C. § 2284(a). Plaintiffs are a number of Latino voters who reside in and adjacent to the districts that are challenged in the action, as well as several Latino advocacy and human service organizations. Defendants include Governor Gray Davis, Lieutenant Governor Cruz Bustamante, Secretary of State Bill Jones, Senate President John Burton, and Assembly Speaker Herb Wesson. Additionally, several parties, including the California Republican Party and the California State Senate, have intervened as defendants.^[4]

Plaintiffs challenge the legality of three districts: Congressional District 28 ("CD 28"), which is located in the San Fernando Valley of Los Angeles County; Congressional District 51 ("CD 51"), the southern-most congressional district in the state, which includes portions of San Diego County and the entirety of Imperial County; and State Senate District 27 ("SD 27"), which is comprised of several communities in Southeast Los Angeles County, including parts of the City of Long Beach.

In their first amended complaint, plaintiffs advance three legal theories.^[5] They allege a Shaw claim: a claim that CD28 and CD 51 are racially gerrymandered districts in violation of the Equal Protection Clause of the Fourteenth Amendment, under the cause of action first set forth by the Supreme Court in Shaw v. Reno, 509 U.S. 630, 113 S. Ct. 2816, 125 L. Ed. 2d 511 (1993) ("Shaw I"). They next allege that the legislature intentionally diluted the vote of Latinos in the San Fernando Valley and in San Diego County (the areas in which the two challenged congressional districts are located), in violation of both the Equal Protection Clause and § 2 of the Voting Rights Act. Finally, they allege that the new boundary lines of CD 28, as well as of SD 27 (located in Southeast Los Angeles County) have the effect of diluting the Latino vote in contravention of § 2 of the Voting Rights Act.

Procedural History

Soon after the complaint was filed, plaintiffs applied for a temporary restraining order ("TRO"), asking that this court enjoin the use of the new district lines and postpone the March, 2002, primary election until another districting plan could be adopted. On November 1, 2001, the court heard oral argument on plaintiffs' motion, and subsequently denied the TRO application. Cano v. Davis, 191 F. Supp. 2d 1135 (C.D.Cal.2001) (per curiam).

Two challenges to the state redistricting statute were filed in the California state courts. See Andal v. Davis, Sacramento Superior Court No. 01-CS-01397; Kennedy v. Davis, Santa Clara Superior Court No. CV-803679. Citing these two actions, defendants moved to defer adjudication of the case pursuant to the doctrine established in Railroad Comm'n of Texas v. Pullman Co., 312 U.S. 496, 61 S. Ct. 643, 85 L. Ed. 971 (1941). Because it was altogether speculative that the state court actions, which involved districts in Northern California, would lead to changes in the redistricting plan that would affect the districts challenged in this case, we denied the deferral motion. Cano v. Davis, 191 F. Supp. 2d 1140 (C.D.Cal.2002) (per curiam). This court also decided motions asserting the legislative privilege of state legislators and their aides, filed in response to discovery demands by plaintiffs, Cano v. Davis, 193 F. Supp. 2d 1177 (C.D.Cal.2002), as well as a motion to quash the taking of depositions from certain members of Congress.

In light of the important statutory and constitutional claims asserted by plaintiffs, we established an expedited discovery and trial schedule; discovery was scheduled to conclude on May 31, 2002, and has been extended to June 14, 2002, by agreement of the parties. Defendants moved for summary judgment on all claims, and we set a hearing for May 13, 2002. Plaintiffs did not seek to defer the hearing, or our ruling on the motion, pending the completion of discovery. We conducted the hearing as scheduled and heard extensive oral argument.

DISCUSSION

I. Plaintiffs' Racial Gerrymandering Claims

Plaintiffs assert that both CD 28 and CD 51 violate the Equal Protection Clause under the "racial gerrymander" doctrine set forth in Shaw I. The districts in question, however, cannot reasonably be understood to segregate voters into racial enclaves, and thus fail to state a Shaw claim as a matter of law. We reject plaintiffs' expansive theory of the racial gerrymander doctrine, which would throw into doubt the validity of countless legislative districts nationwide. Moreover, even were such an expansive interpretation of the doctrine warranted, plaintiffs have failed to offer evidence showing that the California legislature abandoned traditional districting principles in the configuration of these two districts, or subordinated those principles to racial considerations.

A. Background

In Shaw I, plaintiffs challenged on equal protection grounds North Carolina's 12th Congressional District, which was a majority-black district drawn primarily for the purpose of meeting the preclearance demands of the United States Department of Justice during the 1990 round of redistricting.^[6] The shape of the 12th district was, to put it mildly, bizarre; it was a narrow, serpentine band that stretched for 160 miles along Interstate 85 and featured a series of tentacles reaching out from the interstate corridor to draw scattered and disparate black neighborhoods into the district. Id. at 635-36, 113 S. Ct. 2816. It was the least geographically compact congressional district in the nation. Shaw v. Hunt, 517 U.S. 899, 906, 116 S. Ct. 1894, 135 L. Ed. 2d 207 (1996) ("Shaw II").

The Shaw I Court reversed the three-judge district court's dismissal of the action, and held that plaintiffs had stated a racial gerrymander claim under the Equal Protection Clause. Id. at 658, 113 S. Ct. 2816. "In some exceptional cases," Shaw I held, "a reapportionment plan may be so highly irregular that, on its face, it rationally cannot be understood as anything other than an effort to segregate voters on the basis of race." Id. at 646-47, 113 S. Ct. 2816 (internal citations and quotations omitted). As the Court subsequently explained, in order to prevail on what has come to be known as a Shaw claim, "[r]ace must not simply have been a motivation for the drawing of a majority minority district, but the predominant factor motivating the legislature's decision. Plaintiffs must show that a facially neutral law is unexplainable on grounds other than race." Easley v. Cromartie, 532 U.S. 234, 241-42, 121 S. Ct. 1452, 149 L. Ed. 2d 430 (2001) ("Cromartie II") (emphases in original) (internal citations and quotations omitted).

Although Shaw I relied primarily on the bizarre shape of the district at issue in concluding that it was an impermissible racial gerrymander, 509 U.S. at 647, 113 S. Ct. 2816 ("[R]eapportionment is one area in which appearances do matter"), the Court held in Miller v. Johnson that a district need not "be bizarre on its face before there is a constitutional violation." 515 U.S. 900, 912, 115 S. Ct. 2475, 132 L. Ed. 2d 762 (1995). Rather, Miller held that a district's shape is relevant only as "persuasive circumstantial evidence that race for its own sake, and not other districting principles, was the legislature's dominant and controlling rationale for drawing its district lines." Id. at 913, 115 S. Ct. 2475.

All of the racial gerrymandering cases emphasize that a plaintiff bringing such a claim faces an extraordinarily high burden. See, e.g., Cromartie II, 532 U.S. at 241, 121 S. Ct. 1452 ("[T]he burden of proof on the plaintiffs (who attack the district) is a demanding one.") (internal citations and quotations omitted); Miller, 515 U.S. at 929, 115 S. Ct. 2475 (O'Connor, J., concurring) ("Shaw's basic objective" is to "make[] extreme instances of racial gerrymandering subject to meaningful judicial review"); Shaw I, 509 U.S. at 646, 113 S. Ct. 2816 (noting that a violation will only occur in "exceptional cases"). The Supreme Court has also recognized the undeniable reality that the interests of racial groups are often part of a legislature's rationale for drawing a particular set of district lines. Miller, 515 U.S. at 916, 115 S. Ct. 2475 ("Redistricting legislatures will ... almost always be aware of racial demographics ...."); Shaw I, 509 U.S. at 646, 113 S. Ct. 2816 ("[R]edistricting differs from other kinds of state decisionmaking in that the legislature always is aware of race when it draws district lines ....") (emphasis in original). Indeed, as racial minorities continue to struggle to overcome the legacy of official discrimination in election practices, awareness of race is essential to fairness in districting. For this reason, the provisions of the Voting Rights Act oftentimes require a redistricting legislature to consider the effect of its decisions on the political power of racial groups, and to take race into account when drawing lines. 42 U.S.C. §§ 1973, 1973c. The necessity of considering race, at least to some degree, when redistricting, creates a tension between a legislature's constitutional obligations as set forth in Shaw and its obligations under the Voting Rights Act. See Bush v. Vera, 517 U.S. 952, 994, 116 S. Ct. 1941, 135 L. Ed. 2d 248 (1996) (O'Connor, J., concurring) ("The [Voting Rights Act] requires the States and the courts to take action to remedy the reality of racial inequality in our political system, sometimes necessitating race-based action, while the Fourteenth Amendment requires us to look with suspicion on the excessive use of racial considerations by the government."). It is for this reason that a Shaw violation will result only in the exceptional case in which a district is "unexplainable on grounds other than race" and the particular harms described in Shaw occur. Cromartie II, 532 U.S. at 242, 121 S. Ct. 1452 (internal quotations and citations omitted).

The Shaw doctrine is unusual in that, unlike most constitutional doctrines, it requires no concrete injury. Rather, one of the two principal harms identified in Shaw I is an "expressive" harm that results solely from the classification of voters on the basis of race, a practice the Shaw I Court held "bears an uncomfortable resemblance to political apartheid." 509 U.S. at 647, 113 S. Ct. 2816; see Richard H. Pildes & Richard G. Niemi, Expressive Harms, "Bizarre Districts," and Voting Rights: Evaluating Election-District Appearances After Shaw v. Reno, 92 MICH. L. REV. 483, 485 (1993). This "expressive" harm "reinforces the perception that members of the same racial group ... think alike, share the same political interests, and will prefer the same candidates at the polls." Shaw I, 509 U.S. at 647, 113 S. Ct. 2816. The other harm discussed in Shaw I is a "representational" harm. That is, when a district "obviously is created solely to effectuate the perceived common interests of one racial group," a representative elected from that district—who is supposed to represent all of his constituents, regardless of whether they supported him—will feel bound only to advance the interests of the racial group that dominates the district. Id. at 648, 113 S. Ct. 2816.

B. Plaintiffs' Evidence of Purported Racial Gerrymandering Does Not Establish A Shaw Violation.

Plaintiffs allege that both CD 28 and CD 51 constitute impermissible racial gerrymanders. With respect to CD 51, plaintiffs assert that the exclusion from the district of certain Latino neighborhoods in the City of San Diego, including Barrio Logan, Logan Heights, Golden Hill, Sherman Heights and others, violated the Equal Protection Clause under the doctrine established in Shaw I. All but one of these neighborhoods had previously been represented by Congressman Robert Filner, the incumbent in CD 51 (which had been numbered CD 53 following the enactment of the 1990 Special Masters' Plan);^[7] plaintiffs allege that the failure to include the areas in the reconfigured district violates Shaw because the legislature's decision was based solely on racial considerations. According to plaintiffs, the legislature excluded the neighborhoods in question in order to maintain a specific racial balance in the district. Plaintiffs make similar arguments with respect to CD 28. There, the legislature's redistricting plan placed Latino neighborhoods in Van Nuys, Sylmar, and Burbank that were previously represented by the incumbent congressman, Rep. Howard Berman, in a neighboring congressional district (although a small portion of the transferred territory was assigned to a third district that also adjoined the district represented by Berman). Plaintiffs contend that the legislature removed these areas from CD 28 (which had previously been numbered 26 following enactment of the Special Masters' Plan), solely to reduce the number of Latinos in that district.

Even assuming plaintiffs' allegations of racial intent to be true,^[8] as we do for purposes of this summary judgment motion, the rationale underlying Shaw is simply inapplicable to the districts at issue here.^[9] Plaintiffs' two Shaw claims are not addressed to the types of districts ordinarily at issue in the Supreme Court's racial gerrymandering cases. For one, these are not race-based districts that "balkanize us into competing racial factions" or deliberately segregate voters into separate racial enclaves. Shaw I, 509 U.S. at 657, 113 S. Ct. 2816. They cannot, under any fair reading, be characterized as "white districts" or "Caucasian districts." Nor are they districts that can only be reasonably understood to "belong" to one ethnic or racial group.

To the contrary, the districts at issue here are diverse and multi-ethnic: each contains a variety of racial and ethnic groups; none unites any single group of individuals within its boundaries for the purpose of permitting that group to exercise hegemony. In fact, Latinos are the largest number of persons in any single racial or ethnic group in each district, and the number of whites in each case is substantially lower. In CD 28, for example, the voting-age population is 49.2% Latino, 37.2% white, 6.9% Asian, and 4.3% black. Congressional District 51 is even more diverse: in that district the voting-age population is 49% Latino, 25.1% white, 13.8% Asian, and 9.7% black. There is no evidence that the legislature sought to ensure that voters of any particular race would dominate either district.^[10] Perhaps for this reason, one searches the record in vain for evidence of the type of harms that Shaw claims are designed to remedy—the "representational" and "expressive" harms described in Shaw I.^[11]

First, it is impossible to conclude that either district gives rise to the type of "representational" harm described in Shaw I. In neither district does any single racial or ethnic group comprise more than 39.8% of the registered electorate. It would therefore be political folly, if not suicide, for a representative in either district to consider it his "primary obligation" to act as the representative of a single racial or ethnic group. Shaw I, 509 U.S. at 648, 113 S. Ct. 2816. The message sent to future representatives by the creation of these two districts is not a "pernicious" one: that "they represent a particular racial group rather than their constituency as a whole." Id. To the contrary, the message communicated by the redistricting plan is that in order to succeed in either of these districts a representative must build and maintain multi-racial coalitions. Cf. Johnson v. De Grandy, 512 U.S. 997, 1020, 114 S. Ct. 2647, 129 L. Ed. 2d 775 (1994) ("[M]inority voters are not immune from the obligation to pull, haul and trade to find common political ground ....").

Nor is there evidence of the "expressive" harm that the Court described in Shaw I. These districts are not assemblages of "... individuals who belong to the same race, but who are otherwise widely separated by geographical and political boundaries, and ... have little in common with one another but the color of their skin." Shaw I, 509 U.S. at 647, 113 S. Ct. 2816. Rather, each of the districts is ethnically diverse, and no one group constitutes a majority in either. Their demographic make-up thus conveys no message that constitutional principles have been disregarded, or that the legislature sanctions racial classifications.

This is particularly true when one considers the shape of the districts in question. See Miller, 515 U.S. at 912, 115 S. Ct. 2475 (noting that the harm in Shaw I derived from the "appearance [of the district] in combination with certain demographic evidence."). Where a district's shape signals that it was created for the purpose of aggregating voters of a particular race, without regard to whether they have distinct interests besides their shared racial heritage, a Shaw violation may result. Shaw I, 509 U.S. at 647, 113 S. Ct. 2816. Where such graphic indicia of racial districting are absent, the likelihood of "expressive harm" is reduced, if not eliminated. See Pildes & Niemi, supra, at 614 ("Race-as-a-factor districts of a more compact nature are less likely to send these messages. Rather, they are likely to be viewed as primarily dependent on neutral, nonracial facts, such as geographical and political boundaries. To be sure, a racial group may be disproportionately present in compact districts, but this can be seen as the result of residential housing patterns, not an intent to draw a line in order to reaffirm racial differences."). While, as noted earlier, the Court has instructed that a district need not "be bizarre on its face before there is a constitutional violation," Miller, 515 U.S. at 912, 115 S. Ct. 2475, it is nonetheless true that the shape of the districts has been a significant factor in the outcome of the Shaw cases. See, e.g., id. at 917, 115 S. Ct. 2475. See also Vera, 517 U.S. at 965, 116 S. Ct. 1941; Id. at 972, 116 S. Ct. 1941; Id. at 973-74, 116 S. Ct. 1941; Cromartie II, 532 U.S. at 240, 243, 121 S. Ct. 1452.^[12]

The districts drawn by the legislature in the San Fernando Valley and in San Diego/Imperial Counties do not signal that voters were placed in one district or another because of their race. Indeed, both their shape and the balance of the evidence in the record suggests that the legislature did not violate the principle that race is a permissible criterion to use in the redistricting so long as it is not the predominant criterion. In the case of each district, the population shifts that are the subject of plaintiffs' Shaw claims represent a relatively minor portion of the districts at issue. Moreover, the removals of Latino voters were made at the margins or perimeters of the districts, and the districts as a whole remained largely unchanged (except for the added territories).^[13] The most recent of the Supreme Court's racial gerrymandering cases, Cromartie II, illustrates that such discrete population shifts cannot constitute a Shaw violation. In that case, there was evidence presented that the state legislature's map-drawing consultant had moved approximately 60,000 black voters into a district, presumably on the basis of their race. 532 U.S. at 254, 121 S. Ct. 1452. Nevertheless, the Cromartie II Court reversed the three-judge district court's finding of a Shaw violation as clearly erroneous, because there was insufficient evidence to show that the district was unexplainable on a basis other than race. Id. at 257, 121 S. Ct. 1452. The same holds true here; even if the perimeter Latino neighborhoods in question were removed from the districts intentionally because of their ethnic demographics, there is no evidence in the record that either district as a whole is a "white" district or that it is explainable only in terms of race. To adopt plaintiffs' sweeping construction of the racial gerrymandering doctrine would render suspect virtually any legislative decision regarding the districts in which to place areas or groups of precincts that are predominantly populated by one racial or ethnic group. Such an approach would render meaningless the guarantee that the Shaw doctrine "does not throw into doubt the vast majority of the Nation's 435 congressional districts ... even though race may have been considered in the redistricting process." Miller, 515 U.S. at 928-29, 115 S. Ct. 2475 (O'Connor, J., concurring).

Plaintiffs contend that a group of Latino residents was excluded from each of CD 28 and CD 51 in order to ensure that the number of Latino residents in those districts did not rise above a certain percentage. To the extent that this is true, it implicates the Equal Protection Clause as a claim of constitutional vote dilution, not as an "analytically distinct" claim of racial gerrymandering. Miller, 515 U.S. at 915, 115 S. Ct. 2475. What plaintiffs seek as a Shaw remedy is a redistricting plan that contains the maximum number of majority-Latino districts possible. This is not a requirement of constitutional districting. See Shaw II, 517 U.S. at 913, 116 S. Ct. 1894 ("We have recognized that a ``State's policy of adhering to other districting principles instead of creating as many majority-minority districts as possible does not support an inference that the plan ``so discriminates on the basis of race or color as to violate the Constitution'....'") (quoting Miller, 515 U.S. at 924, 115 S. Ct. 2475). Cf. De Grandy, 512 U.S. at 1017, 114 S. Ct. 2647 ("Failure to maximize cannot be the measure of § 2"). To hold otherwise would subvert the most basic principle of Shaw and the subsequent racial gerrymandering cases.

C. Plaintiffs Put Forth No Evidence That Traditional Districting Principles Were Not Followed and Were Subordinated to Racial Considerations.

Defendants are also entitled to summary judgment for another reason made clear in both Shaw I and Miller: plaintiffs have failed to raise a question of material fact as to whether, in drawing CD 28 and CD 51, the state abandoned its traditional districting principles, or subordinated them to racial considerations. This is so because no reasonable trier of fact could conclude from the evidence in the record that in either district the legislature "subordinated traditional race-neutral districting principles, including but not limited to compactness, contiguity, and respect for political subdivisions or communities defined by actual shared interests, to racial considerations." Miller, 515 U.S. at 916, 115 S. Ct. 2475.

In order to prevail on a Shaw claim, a plaintiff must demonstrate both that the legislature was predominantly motivated by racial intent in constructing the challenged district and that it ignored traditional districting principles in its zeal to pursue a racial objective. Cromartie II, 532 U.S. at 242-43, 121 S. Ct. 1452; Shaw II, 517 U.S. at 902, 116 S. Ct. 1894; DeWitt v. Wilson, 856 F. Supp. 1409, 1412-13 (E.D.Cal.1994) (three-judge court), aff'd, 515 U.S. 1170, 115 S. Ct. 2637, 132 L. Ed. 2d 876 (1995).

Perhaps the paramount race-neutral districting principle under California law is the constitutionally-mandated and court-ordered principle that legislative districts follow local political subdivision boundaries to the extent possible consistent with other districting requirements. CAL. CONST., Art. XXI; Legislature v. Reinecke, 10 Cal. 3d 396, 110 Cal. Rptr. 718, 516 P.2d 6, 10 (1973).^[14] Courts examining racial gerrymandering claims have found adherence to local political subdivisions to be highly probative evidence that a district does not violate the Equal Protection Clause. See, e.g., Harvell v. Blytheville Sch. Dist., 126 F.3d 1038, 1041 (8th Cir.1997) (rejecting a Shaw challenge to a district court's remedial plan in a § 2 vote dilution case in part because the remedial districts, although majority-minority, did not split local political units).

The districts challenged here under Shaw are reasonably compact and contiguous. They are no more irregular in shape than any other district created by the legislature, and certainly do not constitute a threshold showing of bizarreness that would support an inference that the districts are racially gerrymandered.^[15] There is undisputed evidence in the record that the redistricting statute both achieved the precise mathematical equality in congressional district populations required by Karcher v. Daggett, 462 U.S. 725, 730, 103 S. Ct. 2653, 77 L. Ed. 2d 133 (1983), and decreased the number of city-splits in congressional districts statewide from 144 to 32. More important, though, there is also specific, uncontradicted evidence that the traditional districting principles applied statewide were not abandoned, but rather were rigorously applied, in the configuration of both CD 51 and CD 28.

1. Congressional District 51 (San Diego and Imperial Counties)

In the case of CD 51, the entirety of Imperial County was added to the district, which had previously been comprised only of portions of San Diego County. Although this addition resulted in the district becoming less compact, the inclusion of Imperial County occurred between the first and second drafts of the plan at the urging of Latino advocacy organizations which contended that the Latino residents of Imperial County shared a community of interest with the Latino residents of San Diego County, because both counties are situated adjacent to the Mexican border. In light of the addition of the 142,361 residents of Imperial County to CD 51, there is no question that it was necessary to remove other areas from the district in order to maintain the required population level. There is also no question that the areas that were removed were largely Latino neighborhoods. It is the removal of these neighborhoods that plaintiffs challenge.

The objective and undisputed evidence demonstrates that in determining what changes should be made with respect to the geography and demographic composition of the district, the legislature respected local political subdivision lines and, for the most part, kept existing communities of interest united in the district. The district includes Imperial County in its entirety, for instance, rather than solely Latino or non-Latino enclaves from within that county. The district also includes the entire cities of Chula Vista and National City, and does not subdivide either of those localities. Although CD 51 contains just a portion of the City of San Diego, that city is too large to be contained in any single congressional district; thus, that city-split was unavoidable. Moreover, the district maintains intact black and Filipino communities of interest; the record contains undisputed evidence that leaders of both those communities sought to keep their neighborhoods in CD 51 because of an affinity for Rep. Filner and approval of the leadership role he has assumed on issues of importance to them.^[16] Finally, the portion of CD 51 that the legislature removed during redistricting in order to compensate for the addition of Imperial County is a compact area on the northwestern perimeter of the district; in making this choice, the legislature preserved the essential core of the previous district.

As in Cromartie II, defendants also offer the justification that politics, not race, was the reason that the legislature chose to remove particular areas from the district. Here, the proffered political explanation is the desire to strengthen the dominant political party's electoral base in a politically-marginal district that abuts CD 51.^[17] The adjacent district, CD 53, to which the neighborhoods were transferred, was represented by Rep. Susan Davis, a first-term Democrat who defeated a Republican incumbent in 2000. Before the redistricting, CD 53—in contrast to CD 51—was hardly a safe Democratic seat: it contained 39.1% Democrats and 36% Republicans. By the end of the decade, because of population shifts, it also contained 52,000 persons fewer than the ideal congressional district. As a result of the redistricting plan, CD 53 is now a much safer Democratic seat: 42.3% of its voters are registered Democrats, and 32% are enrolled as Republicans. It now also meets precisely the equipopulous requirement. See n. 14, supra.

Plaintiffs do not dispute that the transfer of Latino neighborhoods into CD 53 raised the proportion of Democratic voters in that district; they merely respond that defendants might have achieved the same goal without transferring Latino voters from CD 51 to CD 53. Whether this assertion is true—and we assume in connection with this motion that it is—is irrelevant for purposes of Shaw liability. Defendants have come forward with a substantial, uncontroverted non-racial reason for the drawing of the lines in CD 51: the protection of an incumbent in an adjoining district, a well-established legitimate districting criterion. White v. Weiser, 412 U.S. 783, 793-97, 93 S. Ct. 2348, 37 L. Ed. 2d 335 (1973).

Most important, plaintiffs have put forth no specific evidence demonstrating that defendants abandoned traditional districting principles when drawing CD 51. The district constitutes a typical Congressional district that comports with traditional districting principles.

Plaintiffs dispute this, arguing that the evidence they have adduced that Latino voters were excluded from CD 51 based on their race mandates that this claim proceed to trial.^[18] Citing the Supreme Court's reversal of the district court's grant of summary judgment in favor of the plaintiffs in Hunt v. Cromartie, plaintiffs argue that "[t]he legislature's motivation is itself a factual question," 526 U.S. 541, 549, 119 S. Ct. 1545, 143 L. Ed. 2d 731 (1999) ("Cromartie I"), and thus that summary judgment is not proper. Specifically, they contend that, having produced some evidence that some voters were placed in CD 51 or CD 53 based on their race, there is necessarily a triable issue of fact as to whether race was the predominant factor in the districting decision. In essence, plaintiffs contend that summary judgment is never available to a defendant on a Shaw claim where any evidence of intent is present. This is not the case.

It is true that the Cromartie I Court stated that "[r]easonable inferences from the undisputed facts [could] be drawn in favor of a racial motivation finding or in favor of a political motivation finding," and thus that "it was error for the District Court to resolve the disputed fact of motivation at the summary judgment stage." Id. at 552, 119 S. Ct. 1545. We note, however, Justice Stevens' observation in his concurring opinion that the Court did "not have before [it] the question whether the District Court erred in denying the State's motion for summary judgment...." Id. at 558, 119 S. Ct. 1545 (Stevens, J., concurring). Subsequently, the Court reversed the district court's entry of judgment in plaintiffs' favor following trial, holding that their evidence of intent was not adequate to prove a Shaw violation. Cromartie II, 532 U.S. at 257-58, 121 S. Ct. 1452. The history of Cromartie II convinces us that it is permissible to determine on summary judgment that the evidence of intent proffered by a Shaw plaintiff is not sufficient, under the "demanding" burden such a plaintiff admittedly has, see id. at 241, 121 S. Ct. 1452, to raise an inference that race was the predominant motivation for a districting decision.

Our view of the evidence in the present case, moreover, is informed by the Supreme Court's statement in Vera, 517 U.S. at 962, 116 S. Ct. 1941, that "neglect of traditional districting criteria is ... [a] necessary" element of a Shaw claim. See also Cromartie II, 532 U.S. at 241, 121 S. Ct. 1452 ("... those who claim that a legislature has improperly used race as a criterion ... must show at a minimum that the ``legislature subordinated traditional race-neutral districting principles ... to racial considerations.'").

Plaintiffs contend that, having put forward some evidence of intent, they have no burden other than to meet any non-racial explanations for the district proffered by defendants. They assert they have drawn maps that show that the legislature could have kept the Latino communities at the northwestern boundary of CD 51 within the district, and that this demonstrates that traditional districting criteria do not explain the decision that was made. The fact that the district could have been drawn differently, however, does not demonstrate that traditional districting principles were subordinated to race. Here, as detailed above, the undisputed evidence shows that the district in question is relatively compact and contiguous, that it does not split cities, and that it meets the state's obligation to draw equipopulous districts. Also as noted, the legislature has proffered a non-racial justification for placing the Latino communities in question in CD 53— to increase the Democratic registration of that district, which was the lowest of any district represented by a Democratic member of Congress in the state, and which remains the lowest following redistricting. The fact that this could have been achieved by placing the Latino communities in CD 51 and moving other Democratic voters to CD 53 is not evidence that traditional districting principles were subordinated to race. See id. at 249, 121 S. Ct. 1452 ("we recognize that some ... other ways [to create a safely Democratic district] may exist. But, unless the evidence also shows that these hypothetical alternative districts would have better satisfied the legislature's other nonracial political goals as well as traditional nonracial districting principles, this fact alone cannot show an improper legislative motive.") (emphasis supplied). This is particularly true since the court must accord a presumption of good faith to the legislative decision. See Miller, 515 U.S. at 916, 115 S. Ct. 2475.

"[A] legislature's compliance with ``traditional districting principles such as compactness, contiguity, and respect for political subdivisions' may well suffice to refute a claim of racial gerrymandering." Id. at 919, 115 S. Ct. 2475. Here, given the configuration of the district, and the nature of the changes made during the districting process, the record shows, at most, that the legislature took race into account along with traditional districting principles. Yet a Shaw claim exists "[o]nly if traditional districting criteria are neglected ... predominantly due to the misuse of race...." Vera, 517 U.S. at 993, 116 S. Ct. 1941 (O'Connor, J., concurring) (emphasis supplied). As there is no evidence that such principles were "neglected" in drawing CD 51, plaintiffs' Shaw claim must fail.

A word must be said about the evidentiary significance of the opinion offered by Dr. J. Morgan Kousser, plaintiffs' expert. Relying on the intent evidence described above, Dr. Kousser offers the opinion that race was the predominant factor in the legislature's decision to draw CD 51 as it did. As the Court stated in Cromartie II regarding the similar opinion of plaintiffs' expert in that case, "this statement of the conclusion is no stronger than the evidence that underlies it." 532 U.S. at 249, 121 S. Ct. 1452. Having concluded that plaintiffs' evidence of racial intent—the very evidence on which Dr. Kousser relies— does not raise a triable issue of fact regarding the subordination of traditional districting principles to race in constructing CD 51, Dr. Kousser's statement to the contrary likewise cannot suffice. While he focuses heavily on evidence of intent, Dr. Kousser does not squarely address whether the legislature neglected traditional districting principles. Indeed, he acknowledges that "MALDEF drew districts in the same general areas [as the legislature] that appear just as compact." As respects CD 51 in particular, he asserts only that the legislature's district is "no more compact than MALDEF's suggestion," and "cannot be said to preserve communities of interest more adequately than its alternatives."^[19] These statements suffer the same fate as the argument advanced in plaintiffs' brief that CD 53 could have been strengthened for the Democratic incumbent, Davis, by moving voters other than those in the Latino communities to the north and west of CD 51 into her district.^[20] While there may be "other ways" to achieve the legislature's objectives, the evidence "must show that the[] hypothetical alternative districts would have better satisfied the legislature's other nonracial political goals as well as traditional nonracial districting principles" before a Shaw claim can be stated. Cromartie II, 532 U.S. at 249, 121 S. Ct. 1452 (emphasis supplied).

In short, while we accept for purposes of this motion plaintiffs' allegation that race impacted the drawing of CD 51, plaintiffs have failed to raise a triable issue regarding the "neglect" of traditional districting principles that is a "necessary" element of their Shaw claim. Vera, 517 U.S. at 962, 116 S. Ct. 1941. They have thus failed to raise a genuine issue of material fact as to whether the district is unexplainable on grounds other than race.

2. Congressional District 28 (San Fernando Valley)

Congressional District 28 is also a compact and contiguous district that does not deviate from the ideal size. As with CD 51, plaintiffs have put forward no evidence that the reconfigured CD 28 departs from traditional districting principles. While plaintiffs have offered evidence that CD 28 was drawn with some consciousness of race, that alone is insufficient to support a Shaw claim. DeWitt, 856 F.Supp. at 1415 (concluding that the redistricting plan developed by Special Masters and approved by the California Supreme Court "is not racial gerrymandering, but rather a thoughtful and fair example of applying traditional redistricting principles, while being conscious of race.").

Plaintiffs allege that map-drawer Michael Berman moved Latino voters out of CD 28 so as to reduce the efforts his brother, the incumbent, would be required to make in order to defeat a potential primary challenge by a future Latino candidate. In particular, Michael Berman allegedly sought to limit the number of Latino voters in the district so that his brother would not have to leave his legislative activities in Washington, D.C., and travel to California frequently in order to ensure his re-election. We assume, for the purpose of this motion, that such was Michael Berman's intent, and that the legislature as a whole adopted it.^[21] Additional evidence of a race-conscious motive in the drawing of district lines is found in a statement made by the Senate Redistricting Committee Chair, Sen. Don Perata, at a public hearing. In that statement, Sen. Perata offered his opinion that the representation of Latinos would be enhanced by the movement of some Latinos from the district of Rep. Howard Berman, an "ardent supporter of Latino issues," to the neighboring district of Rep. Brad Sherman, to "encourage Congressman Sherman ... to take leadership on issues of concern to the Latino community as Howard Berman has done for the past 30 years."^[22]

Plaintiffs argue that Sen. Perata's statement, "by itself," necessitates denying defendants' summary judgment motion, as it "directly announces a singular, race-based motivation for drawing the map to split the Latino community between two congressional districts." Read in context, Sen. Perata's statement was a response to the concerns that had been raised by Latino advocacy groups about CD 28 during the course of the redistricting process. It is essentially equivalent to the statement the Supreme Court found insufficient to support a Shaw claim in Cromartie II. See 532 U.S. at 253, 121 S. Ct. 1452 ("We agree that one can read the statement about ``racial ... balance' ... to refer to the current congressional delegation's racial balance. But even as so read, the phrase shows that the legislature considered race, along with other partisan and geographic considerations; and as so read it says little or nothing about whether race played a predominant role comparatively speaking") (emphasis original).

Moreover, even coupled with plaintiffs' other evidence regarding intent, the statement is not sufficient to raise a triable issue as to whether traditional districting principles were subordinated to race. Plaintiffs assert that the horseshoe/stake shape of the district supports an inference that race predominated, citing the fact that old CD 24 and 26, which CD 27 and 28 replaced, were essentially circular or oblong in configuration. The fact that CD 28 is surrounded on two sides by parts of CD 27, however, does not suggest out-of-the-ordinary districting. The district is relatively compact and contiguous; it does not have tentacles, appendages, land bridges, corridors, or wings. See Vera, 517 U.S. at 965, 972, 973-74, 116 S. Ct. 1941; Miller, 515 U.S. at 917, 115 S. Ct. 2475. It does not look like a snake, bird or other animal. See Cromartie II, 532 U.S. at 240, 243, 121 S. Ct. 1452; Vera, 517 U.S. at 973-74, 116 S. Ct. 1941. Indeed, the legislature drew another congressional district with approximately the same shape that plaintiffs do not challenge. Additionally, CD 28 satisfied the legislature's obligation to draw equipopulous districts.

As discussed earlier, the mere use of race as a reason for a redistricting decision cannot lead to a Shaw violation where that reason is not the dominant and controlling rationale, so that the legislature's redistricting is unexplainable on grounds other than race. See Cromartie II, 532 U.S. at 258, 121 S. Ct. 1452. Here, despite some evidence of race-consciousness, there is abundant and uncontested evidence in the record that traditional districting principles were followed in establishing CD 28, and that such principles were not subordinated to racial concerns. In this connection, the evidence in the record shows that one important reason for the new district lines was to respond to requests by residents of the southern San Fernando Valley ("the Valley") to be included in the district of Rep. Berman. Congressman Berman had represented the southern (and largely white) parts of the Valley for decades as a member of the California Assembly and in Congress. The Special Masters in 1991 adopted a plan that abandoned the historical district configurations in the Valley, and substantially altered the district that Rep. Berman had long represented, by removing liberal white neighborhoods in the southern Valley and replacing those areas with predominantly Latino neighborhoods in the northeastern part of the Valley. The 2001 redistricting restored the southern Valley communities of Sherman Oaks, North Hollywood, and Studio City to Rep. Berman's district, and in exchange removed some of the Latino neighborhoods along its perimeter that had been added ten years earlier. The current redistricting plan thus modified the Special Masters' Plan by reestablishing, in part, Rep. Berman's traditional district. It did so, moreover, in conformity with traditional districting principles; the legislature kept intact the bulk of the Special Masters' district, removed perimeter precincts in the north and west, and restored the southern tier of the district. In fact, CD 28 appears to conform to traditional districting principles to at least the same degree as the district created by the Special Masters and the alternative district proposed by plaintiffs.^[23]

Plaintiffs challenge the proffered explanation for the lines of CD 28, noting that the original proposal for the district did not include the neighborhoods in the southern Valley that had been in Rep. Berman's district prior to the adoption of the Special Masters' Plan in 1991. It is undisputed, however, that legislative committees heard testimony from two homeowners' associations in these areas that advocated placing Sherman Oaks within Rep. Berman's district, and uniting the communities of Sherman Oaks and North Hollywood in a single district. Just as the map-drawers responded to MALDEF's initial complaints about the district, by moving the core of the Latino communities back within its boundaries, so too it appears the legislature responded to the concerns expressed by residents in areas formerly represented by Rep. Berman.

Plaintiffs contend that these communities could have been added to CD 28 without splitting Latino communities in the northern part of the district. As noted earlier, however, the fact that the legislature could have achieved its goal in a different way does not give rise to an inference that race predominated or that traditional districting criteria were subordinated to race. Cromartie II, 532 U.S. at 249, 121 S. Ct. 1452; see also id. at 256-57, 121 S. Ct. 1452 ("a showing that the legislature might have ``swapped' a handful of precincts out of a total of 154 precincts, involving a population of a few hundred out of a total population of about half a million, cannot significantly strengthen appellees' case."). Defendants' proffered explanation, in short, demonstrates that the district is explainable on a basis other than race.

The crux of plaintiffs' argument, and the crux of the debate surrounding CD 28 during the redistricting process, is the fact that the legislature's plan reduced the percentage of Latinos in the district. Latino advocacy groups wanted another majority-Latino district created, while the legislature originally proposed creating two Latino influence districts. The legislature, as previously noted, had no obligation to maximize the number of majority-Latino districts drawn. See Shaw II, 517 U.S. at 913, 116 S. Ct. 1894; Miller, 515 U.S. at 924, 115 S. Ct. 2475. Cf. De Grandy, 512 U.S. at 1017. Similarly, it had no obligation to maximize the number of Latinos placed in CD 28. As drawn, a majority of the district's population is Latino and very close to 50% of its voting age population is Latino.

Accepting plaintiffs' evidence as true, it cannot be said that it is more than the "modicum of evidence" the Court found insufficient in Cromartie II, 532 U.S. at 257, 121 S. Ct. 1452. Most importantly, it does not raise a genuine issue of material fact regarding whether the legislature abandoned or subordinated traditional districting principles to racial considerations when it established CD 28. The legislature's adherence to such principles is fatal to plaintiffs' Shaw claim.

D. Conclusion

The two districts that plaintiffs challenge under Shaw are not districts that foster "political apartheid;" to the contrary, they are remarkably diverse multiracial and multi-ethnic districts. Thus, they are not districts that give rise to expressive or representational harms. Moreover, plaintiffs have failed to raise a material question of fact as to whether traditional districting principles were abandoned or subordinated to race, a necessary element of a Shaw claim. Accordingly, summary judgment is properly entered for defendants on plaintiffs' Shaw claims.

II. Plaintiffs' Voting Rights Act Section 2 Claims (Non-Intentional)

Plaintiffs also bring traditional statutory vote dilution claims regarding CD 28 and SD 27 under Section 2 of the Voting Rights Act. To establish a traditional claim under section 2, plaintiffs need prove only that the creation of the challenged district has the "effect" of diluting the electoral power of minority voters. Intent is not an element. Plaintiffs argue that in CD 28 and SD 27 the new boundary lines have the effect of preventing Latinos from electing candidates of choice. In both of the districts that plaintiffs challenge under this theory, Latinos constitute a plurality of the voting-age population. In both districts, substantial numbers of non-Latino voters have voted for Latino candidates—and Latino candidates have won elections in the areas in question with the assistance of white cross-over votes. Ultimately, we conclude that there is no material question of fact regarding the traditional § 2 claims: the record before us demonstrates beyond dispute that the districts as drawn by the legislature do not have the effect of denying Latino voters the ability to elect representatives of choice.

A. Background

Section 2 prohibits the adoption of any election practice, including the drawing of district lines, that "results in a denial or abridgement of the right of any citizen of the United States to vote on account of race or color."^[24] It has been interpreted to bar the adoption of election districts that minimize or cancel out the voting power of minorities so that the minority group cannot elect representatives of choice. Allen v. Bd. of Elections, 393 U.S. 544, 555, 89 S. Ct. 817, 22 L. Ed. 2d 1 (1969); De Grandy, 512 U.S. at 1012-13, 114 S. Ct. 2647. This section of the Voting Rights Act is particularly unusual among civil rights laws because it requires no showing of discriminatory intent; it was amended in 1982 so as to provide for an "effects test" only. Prior to 1980, vote dilution cases were generally adjudicated using an effects standard first adopted in White v. Regester, 412 U.S. 755, 93 S. Ct. 2332, 37 L. Ed. 2d 314 (1973), as refined and further developed by the lower courts, particularly in Zimmer v. McKeithen, 485 F.2d 1297 (1973) (en banc), aff'd. on other grounds sub. nom. East Carroll Parish School Bd. v. Marshall, 424 U.S. 636, 96 S. Ct. 1083, 47 L. Ed. 2d 296 (1976) (per curiam). In 1980, a splintered Supreme Court decided City of Mobile v. Bolden, 446 U.S. 55, 100 S. Ct. 1490, 64 L. Ed. 2d 47 (1980). Although no single opinion garnered five votes, a majority of the Court agreed that plaintiffs in both constitutional and statutory vote dilution cases must prove discriminatory intent. Because civil rights organizations and others argued that it would be exceedingly difficult to prove minority vote dilution claims under an intent test, Congress amended the Voting Rights Act to require only proof of discriminatory effects in order to establish liability under § 2. See Thomas M. Boyd and Steven J. Markman, The 1982 Amendments to the Voting Rights Act: A Legislative History, 40 WASH. & LEE L. REV. 1347 (1983).

The language of § 2 as amended does not give clear guidance to courts as to how to determine whether a voting practice results in the denial of a minority group's ability to elect its representative of choice; the statute merely directs that courts examine the totality of the circumstances. In light of the vague statutory language, the Supreme Court gave structure to the § 2 inquiry in Thornburg v. Gingles, 478 U.S. 30, 106 S. Ct. 2752, 92 L. Ed. 2d 25 (1986). In order to prevail on a vote dilution claim, the Gingles Court held, three "pre-conditions" must be met: First, the "minority group must be able to demonstrate that it is sufficiently large and geographically compact to constitute a majority in a single-member district." Id. at 50, 106 S. Ct. 2752. That is, unless the minority group can establish that an effective majority-minority district can be created, a vote dilution claim is not cognizable because there is no minority voting power to dilute.^[25] Second, "the minority group must show that it is politically cohesive." Id. at 51, 106 S. Ct. 2752. As the Court explained in Gingles, unless members of a minority group have shared political views, "it cannot be said that the selection of a[n] ... electoral structure thwarts distinctive minority group interests." Id. Third, "the minority must be able to demonstrate that the white majority votes sufficiently as a bloc to enable it ... to defeat the minority's preferred candidate." Id. Of course, if minority-preferred candidates actually win elections, then the minority is able to elect representatives of its choice.

Gingles, like most of the voting rights cases to arise in the first two decades after the enactment of the Voting Rights Act, challenged the use of multi-member districts that effectively subordinated the electoral power of the minority voting population to that of the majority through the conduct of at-large elections in which it was impossible for a minority-preferred candidate to be elected.^[26] Although the three Gingles factors were devised in cases challenging multi-member districting, the Supreme Court held in Growe v. Emison, 507 U.S. 25, 40-41, 113 S. Ct. 1075, 122 L. Ed. 2d 388 (1993), that the Gingles analysis applies equally to vote dilution claims involving single-member districts. We therefore apply the Gingles test in this case.

In Johnson v. De Grandy, 512 U.S. 997, 114 S. Ct. 2647, 129 L. Ed. 2d 775 (1994), the Supreme Court further explained the method of analyzing cases that allege vote dilution through the gerrymandering of single-member districts. In De Grandy, the Court held that vote dilution under § 2 is not automatically established even if all three Gingles pre-conditions are established. Because the statute requires a court to examine the "totality of the circumstances," the De Grandy Court held that although the Gingles factors are necessary to prove a § 2 violation, they are not sufficient: when the three Gingles preconditions are met, a court evaluating a § 2 claim must then assess the totality of the circumstances to determine if the "effects" test is met—that is, if minority voters' political power is truly diluted. Id. at 1013, 114 S. Ct. 2647 ("[F]actfinders cannot rest uncritically on assumptions about the force of the Gingles factors in pointing to dilution."); see Little Rock Sch. Dist. v. Pulaski Cty. Special Sch. Dist., 56 F.3d 904, 911-12 (8th Cir.1995) (assuming satisfaction of the Gingles factors, but affirming a finding of no minority vote dilution based on the totality of the circumstances).^[27]

With these legal principles in mind, we turn to plaintiffs' traditional § 2 claims respecting CD 28 and SD 27. We resolve both on the basis of the absence of a necessary Gingles pre-condition, pre-condition three, and thus do not reach the totality of the circumstances test.

B. Senate District 27

Senate District 27 is located in southeast Los Angeles County, and includes the communities of Downey, Paramount, Cerritos, Artesia, Hawaiian Gardens, Signal Hill, Lakewood, and Lynwood, as well as parts of Long Beach and South Gate. The seat is currently held by Sen. Betty Karnette, who was first elected in 1990. Plaintiffs do not dispute that Sen. Karnette was the Latino candidate of choice; although herself white, she received the majority of Latino support within the district during the 1990 primary election, in which there was no incumbent Senator. It is also undisputed that Sen. Karnette voted in accordance with the Senate Latino caucus 95% of the time in the 1997-98 session, and 85% in the 1998-99 session. Due to the limits on state legislative terms in California, Sen. Karnette will be unable to run for re-election in 2004.

The essence of plaintiffs' traditional § 2 claim regarding Senate District 27 is that the redistricting statute packs Latinos into neighboring Senate District 30, a majority-Latino district, and that the challenged district could and should have been drawn as an additional majority-Latino seat. Latinos constitute a plurality of the votingage population ("VAP") in SD 27: 40% of the VAP is Latino, 36% is white, 12% is Asian and 9% is black. In SD 30, Latinos constitute the overwhelming majority of the district: 76% of the total population and 71% of the VAP is Latino. Although it is undoubtedly true that the legislature could have moved some Latino voters from the Latino-dominated SD 30 to SD 27 in order to construct an additional majority-minority district, we conclude that the Voting Rights Act did not require it to do so. Because the undisputed facts reflect that the non-Latino voters in SD 27 have demonstrated a willingness to vote for Latino-preferred candidates, and because plaintiffs have failed to present evidence that Latino-preferred candidates are regularly defeated as a consequence of the voting patterns of a non-Latino bloc of voters, we conclude that the manner in which the legislature configured SD 27 does not prevent Latino voters from electing a representative of choice. In short, plaintiffs have failed to create a question of material fact regarding the existence of Gingles pre-condition three.

1. Gingles Pre-Condition One: Majority-Minority District

The analysis of the first Gingles precondition in this case is complicated by the unavailability of citizen voting age population ("CVAP") data. The Ninth Circuit, along with every other circuit to consider the issue, has held that CVAP is the appropriate measure to use in determining whether an additional effective majority-minority district can be created. Romero v. City of Pomona, 883 F.2d 1418, 1426 (9th Cir.1989), abrogated on other grounds, Townsend v. Holman Consulting Corp., 914 F.2d 1136, 1141 (9th Cir.1990) (en banc); Negron v. City of Miami Beach, 113 F.3d 1563, 1569 (11th Cir.1997); Campos v. City of Houston, 113 F.3d 544, 548 (5th Cir.1997). The Romero court explained that the purpose of the first Gingles pre-condition is to determine whether minority voters have the potential to form an effective majority district; because noncitizens cannot register to vote, Romero concluded that they should not be included in the figures used to determine whether an additional district can be drawn in which the minority group would have sufficient voting power to elect an additional representative of choice. 883 F.2d at 1426.

Here, the CVAP data at the district and precinct level has not yet been released by the Census Bureau; the only CVAP data available is statewide. That data discloses only that approximately 17% of the statewide CVAP is Latino. Plaintiffs' expert witness, Dr. Kousser, has testified that extrapolating from this number to the CVAP percentage at the local precinct or district level is unreliable.

The Supreme Court has to date declined to endorse any one method of proving that a district is one in which a minority group has the "ability to elect" a representative of choice. Gingles, 478 U.S. at 46, n. 12, 106 S. Ct. 2752. The question, therefore, is whether, because CVAP figures are not available, plaintiffs may rely on another measure, such as VAP, to establish Gingles pre-condition one. Courts have, in some circumstances, employed such alternate measures. See, e.g., Old Person v. Cooney, 230 F.3d 1113, 1121 (9th Cir.2000) (noting that parties did not dispute Gingles pre-condition one figures based on Native American voting age population); see also De Grandy, 512 U.S. at 1014, 1021, n. 18, 114 S. Ct. 2647 (noting that citizen voting age population figures were not available at the time of trial, and using voting age population figures to assess proportionality).

Plaintiffs have submitted a plan for a district containing a Latino population of 70.92%, and a Latino VAP of 66.95%, and defendants do not dispute that such a district could be drawn. We conclude that, in light of the unavailability of the CVAP data, the ability to construct a district that is so substantially Latino both in overall population and in VAP is sufficient to raise a genuine issue of material fact as to the first Gingles pre-condition. In constructing remedial districts, courts frequently conclude that a super-majority of the relevant minority group is required to ensure that the group will constitute an effective voting majority within the district. See Ketchum v. Byrne, 740 F.2d 1398, 1413, 1415-16 (7th Cir.1984) (in remanding for reconsideration of the appropriate remedial districting plan, the court noted that "[a] guideline of 65% of total population has been adopted and maintained for years by the Department of Justice and by reapportionment experts and has been specifically approved by the Supreme Court in circumstances comparable to those before us as representing the proposition of minority population reasonably required to ensure minorities a fair opportunity to elect a candidate of their choice."); see also African American Voting Rights Legal Defense Fund v. Villa, 54 F.3d 1345, 1348 n. 4 (8th Cir.1995) ("We conclude that either 60% of the voting age population or 65% of the total population is reasonably sufficient to provide black voters with an effective majority."). We need not determine whether plaintiffs' figures, if established at trial, would be sufficient to meet the requirements of Gingles pre-condition one. It is sufficient for present purposes to find that plaintiffs have raised a triable issue of fact regarding the issue.^[28]

2. Gingles Pre-Condition Two: Racial Cohesion

There is little dispute that Latinos in SD 27 vote cohesively. Indeed, defendants essentially concede the point for summary judgment purposes. The evidence in the record is overwhelming that Latinos in Los Angeles County vote monolithically for Latino candidates, regardless of the ideology of that candidate. For instance, in 2001, 97% Latinos citywide voted for Los Angeles Mayoral candidate Antonio Villaraigosa, a liberal Latino Democrat; an almost identical percentage voted for Los Angeles City Attorney candidate Rocky Delgadillo, a far more conservative Latino who was running against a liberal white candidate. Plaintiffs' data further shows that in the two Assembly Districts that constitute a substantial portion of SD 27, more than 90% of Latino voters supported Latino candidates in recent elections for the state assembly. At the very least, therefore, a question of material fact exists with respect to Gingles pre-condition two.

3. Gingles Pre-condition Three: Majority Bloc Voting

It is with respect to the third Gingles pre-condition that plaintiffs' evidence fails to establish the existence of a question of material fact. Put simply, SD 27 is a district in which Latino candidates and other candidates preferred by Latino voters can win. Consequently, it is not a district in which "minority and majority voters prefer different candidates [and] the majority, by virtue of its numerical superiority, will regularly defeat the choices of minority voters." Gingles, 478 U.S. at 48, 106 S. Ct. 2752. Such an exercise of negative voting power by the white majority is what the Gingles Court called the "theoretical basis" for a vote dilution injury, and it is wholly absent here. It is insufficient for purposes of a vote dilution claim merely to show that the minority group votes cohesively. Plaintiffs must also show that the cohesive minority preference is regularly defeated by a majority bloc that refuses to vote for the minority preferred candidate. Id. at 51, 106 S. Ct. 2752. Plaintiffs have proffered no proof on the latter point with respect to SD 27.

No election has yet been conducted in the precise set of precincts that now constitutes SD 27; accordingly, to understand the voting behavior of the electorate in that district, we must resort to the examination of exogenous elections, including the results in parts of overlapping districts. Citizens for a Better Gretna v. Gretna, 834 F.2d 496, 502-03 (5th Cir. 1987) (approving the use of exogenous elections in a Gingles pre-condition three analysis, and noting that Gingles "suggests flexibility in the face of sparse data."); Johnson v. Hamrick, 155 F. Supp. 2d 1355, 1371-73 (N.D.Ga.2001) (examining both endogenous and exogenous elections as part of Gingles pre-condition three analysis).^[29]

There is much data in the record concerning past voting patterns. We believe that by far the most probative evidence is that relating to the non-Latino voters' willingness to support Latino candidates for office. Evidence that, when given a choice between two white candidates, Latinos and non-Latinos frequently make the same choice is less probative of Gingles precondition three. This is so for two reasons. Most important, any racial hostility on the part of non-Latino voters towards Latinos is unlikely to have a significant manifestation in elections in which none of the major candidates is Latino. Also, the record reflects that in California, nearly two-thirds of Latinos enroll in the Democratic Party, which is the majority party in the state. Accordingly, when Latinos are placed in a district containing a majority of Democrats, as in SD 27 and CD 28, the general election results are ordinarily not at all probative of whether there are meaningful differences between the preferences of Latino and white voters. We therefore agree with the Ninth Circuit's observation that if Gingles pre-condition three could be defeated on a showing that the preferred white candidates of minority groups regularly won elections, the protections of the Voting Rights Act would be substantially weakened. Such a rule would "doom any Section 2 claim in which white candidates, acceptable to minorities and non-minorities alike, are regularly elected, but in which minority candidates, preferred by minorities but unfavored by whites, are consistently defeated." Ruiz v. City of Santa Maria, 160 F.3d 543, 544 (9th Cir.1998). This is the view not only of the Ninth Circuit but of several others as well.^[30]

Every set of election returns presented by either party demonstrates that Latinos win elections, aided by substantial white cross-over voting, in the areas comprising SD 27. For instance, the evidence submitted by plaintiffs' expert shows that in the two Assembly districts that constitute a substantial portion of SD 27, Latino candidates received large numbers of white votes. In the 50th Assembly District, according to Dr. Kousser's report, 57% and 59% of whites voted for the Latino Assembly candidate in the 2000 primary and general elections respectively. In AD 56, "less than 50% of the Anglo vote"—but evidently a still-substantial percentage— went to the Latino candidate for that post. Moreover, Latino candidates won both elections.^[31] In addition, defendants have put forward uncontradicted evidence that Latino candidates have in fact won the sets of precincts in the two Assembly districts that overlap with SD 27. For instance, in the 2000 general election for AD 56, Latina Assembly Member Sally Havice won 59% of the vote in the portion of her district that is contained within SD 27. The overlapping area has a lesser percentage of Latino voters than does SD 27 as a whole: SD 27 has a 45.1% Latino population and 39.8% VAP, whereas the area contained in both SD 27 and AD 56 is 37.2% Latino overall, and includes a Latino VAP of 33.2%. Defendants also cite evidence that Latino candidates for statewide office, such as Lieutenant Governor Cruz Bustamante, and Insurance Commissioner candidate Diane Martinez win the heavily Democratic area that is SD 27. We do not find such data particularly helpful in assessing the racial attitudes of the district's voters, in part for the same reasons we find the success of Latino-preferred white candidates only marginally probative.^[32] Additionally, we prefer to analyze the local elections more closely because the campaigns are necessarily more personalized and racial attitudes are therefore more relevant. See Gingles, 478 U.S. at 59, 106 S. Ct. 2752.

What is most striking about the evidence with respect to SD 27 is that plaintiffs have not identified a single election in any territory that forms a part of the district in which a Latino candidate or a Latino-preferred candidate has ever been defeated, by a non-Latino bloc vote or otherwise. Indeed, plaintiffs' expert does not even address the results of actual candidate elections either statewide or in the district. Thus, the only evidence in the record regarding elections in SD 27 concerns elections in which the Latino-preferred candidates—many of whom were in fact Latinos—were victorious.

This case contrasts starkly with cases in which courts have concluded that plaintiffs satisfied the third pre-condition of Gingles. There need not be a "virtually complete absence of [minority] elected officials" in order to create a material question of fact regarding Gingles pre-condition three. Clark v. Calhoun Cty., 88 F.3d 1393, 1398 (5th Cir.1996). Thus, the election of some minority-preferred candidates will not necessarily defeat a claim on summary judgment. Id. at 1397. See also National Association for the Advancement of Colored People, Inc. v. City of Niagara Falls, New York, 65 F.3d 1002, 1009 (2d Cir.1995) ("the election of one black candidate does not preclude the plaintiffs from establishing vote dilution."). To create a triable issue with respect to Gingles precondition three, a plaintiff must show that, at least under some circumstances, minority-preferred candidates are defeated by the votes of a bloc of non-minority voters. There is no such showing here.^[33]

Plaintiffs argue that Gingles preconditions two and three "merge," and that they may therefore prevail so long as they demonstrate that the electorate is "racially polarized." This contention finds no support in the law, and is directly contrary to the "theoretical basis" of § 2 as explained in Gingles, 478 U.S. at 48, 106 S. Ct. 2752.^[34] Plaintiffs' evidence does show that Latinos prefer Latino candidates in far higher proportions than do non-Latinos; but it also shows that substantial numbers of non-Latinos support Latino candidates, with the result that the Latino candidates actually win elections. Thus, the evidence of racial polarization is insufficient as a matter of law to establish the "effects" required in a vote dilution case.

Plaintiffs offer two arguments with respect to Gingles pre-condition three in lieu of direct evidence that minority-preferred candidates are regularly defeated by a hostile voting bloc. Neither creates a question of material fact with respect to the third pre-condition. First, plaintiffs argue that the findings of fact in Garza v. County of Los Angeles, 756 F. Supp. 1298 (C.D.Cal.1990), virtually assure a trial regarding Gingles pre-condition three. In Garza, the court found that racially polarized voting was prevalent in Los Angles County, and that the Los Angeles County Board of Supervisors intentionally discriminated against Latinos by drawing the boundary lines of the five county supervisorial districts so that Latinos could not win any of them. Plaintiffs ask us to rely on the Garza findings of fact to support an inference of discrimination here. That case, which is twelve years old and relates to events that occurred even earlier, cannot, in view of the evidence submitted by the parties, be the basis for a finding that at the present time, the bloc voting of non-Latinos regularly results in the defeat of Latino candidates, in Los Angeles County generally, or in SD 27 in particular. Although the existence of historical discrimination is a factor in the "totality of the circumstances" analysis conducted after the three Gingles pre-conditions are met, given the evidence actually before us, the Garza findings shed little light on whether the voters of SD 27 presently vote as a bloc against Latino candidates. The only evidence in the record that is probative of this issue is the outcome of recent elections in the area that comprise SD 27. That data reveals a non-Latino electorate that is willing to and does vote for Latino candidates in sufficient numbers to elect Latinos to office.

We also note that this is a very different case from Garza for other reasons. For instance, in Garza the primary allegation was that non-Latino public officials manipulated the political process to shut Latinos out of elective county office entirely, a situation that was perpetuated by the all-white Board of Supervisors' control of the redistricting process. Here, the allegations relate to the districting decisions of the state legislature. In this regard, the objective evidence is quite different: Latinos constitute 22% of that legislature, even though they represent just 17% of the state's CVAP; Latinos were members of the Senate and Assembly redistricting committees that sponsored the redistricting legislation at issue here; and 23 of 26 Latino legislators supported the redistricting statute, including every Latino State Senator.^[35] In contrast to the situation in Garza, it is impossible to conclude here that Latinos were shut out of the redistricting process that followed the 2000 census, or that an objective of that process was to exclude Latinos from elective state office.

It is certainly not our view that racial discrimination no longer affects our political institutions or motivates any portion of the electorate of Los Angeles County. Still, the election returns offered by both sets of litigants reveal that in Los Angeles County, whites and other non-Latinos are currently far more willing to support Latino candidates for office than in the past. In short, at the outset of the 21st century, the data in the record before us paints a far more encouraging picture of racial voting attitudes than did the data in Garza.

Second, plaintiffs argue that their expert's analysis creates a question of material fact regarding Gingles pre-condition three. Dr. Kousser, a highly-experienced and respected political scientist, who is a nationally recognized expert in voting behavior, has offered a somewhat unconventional measure of a district in which Latino candidates can win elections. After canvassing the results of state legislative elections throughout California, he opines that a district in which Latinos can elect a candidate of choice is one in which they constitute more than 40% of the Democratic Party's registered voters. In a paper submitted to the Legislature as part of MALDEF's lobbying efforts during the 2001 redistricting process,^[36] Dr. Kousser concluded that Latinos have a realistic chance of electing representatives of choice if they constitute at least 30% of the Democratic registration, and an excellent chance once they constitute 40% of the Democratic registration. Under Dr. Kousser's theory, a Latino candidate is unlikely to win a district with the demographics of SD 27.

Plaintiffs argue that it is possible to extrapolate from Dr. Kousser's statewide study of voter behavior the conclusion that a non-Latino bloc in SD 27 will not vote for Latino candidates of choice, and those candidates of choice will therefore be defeated. Dr. Kousser's report, at least as it applies to the Gingles pre-condition three inquiry, does not raise a triable issue of fact. Gingles states that the "inquiry into the existence of vote dilution caused by submergence in a multimember district is district specific." 478 U.S. at 59, 106 S. Ct. 2752. The same holds true for vote dilution claims that allege cracking or packing in single-member districts. Magnolia Bar Ass'n v. Lee, 994 F.2d 1143, 1151 (5th Cir.1993) (affirming district court grant of summary judgment for defendants on Gingles pre-condition three because plaintiffs could not prove white bloc voting in a particular district by the use of statewide statistics). Unlike the local data, Dr. Kousser's statewide data does not directly bear on the attitude of non-Latino voters in the disputed area.

In a state that is as diverse as California in every way, including geographically, a statewide analysis is particularly unpersuasive. California's diversity is substantial and significant. The voting behavior of residents in the technological centers near the San Francisco Bay area, the agricultural region of the Central Valley, the border communities adjacent to Mexico, or the desert regions in the eastern part of the state is far less helpful in predicting the voting patterns of urban, cosmopolitan Angelenos than data from Los Angeles County itself. Dr. Kousser's report does not explain how his analysis regarding what may constitute an effective Latino district in Berkeley, Barstow or Bakers-field reliably predicts voting behavior in Los Angeles County. Given the results of elections that were actually conducted in the territory that makes up the challenged districts—which results show that non-Latinos do not vote as a bloc to defeat Latino candidates—Dr. Kousser's contrary extrapolations from statewide data do not create a genuine issue of fact regarding Gingles factor three.

Nor does Dr. Kousser's analysis of the voting patterns in SD 27 with respect to Propositions 187 (denying state services to undocumented immigrants), 209 (prohibiting public agencies in the State of California from adopting affirmative action policies) and 227 (barring the use of bilingual education in California public schools) raise a triable issue of fact.^[37] All three of the measures were opposed by a substantial majority of the Latino voting population. While Dr. Kousser opines that the propositions polarized the electorate in SD 27, the data he provides does not support his conclusion. First, the data Dr. Kousser provides concerns MALDEF's proposed SD 27, not SD 27 as it was drawn by the legislature. Second, it demonstrates that the measures did not produce a consistent non-Latino voting bloc in the areas now included within the district. Dr. Kousser's charts regarding Proposition 187 show that the average overall percentage of individuals voting for the measure in the areas he studied was approximately 55%. In these same areas, however, Proposition 209 was defeated, with the average overall percentage voting in favor of the measure being 35.2%. Proposition 227 also appears to have lost in the districts Dr. Kousser examined, with the overall percentage of voters who favored the measure being 47.7%. As respects the latter two propositions, therefore, it appears that the position favored by a majority of Latino voters prevailed. Since Latinos represent no greater than 30% of any of the districts studied, the fact that the measures were defeated reflects substantial cross-over voting by non-Latino constituencies.

Putting aside the fact that the data does not relate to SD 27 as drawn by the legislature, it shows, at best, that Latinos' preferred position was defeated in one election —that regarding Proposition 187. As noted earlier, voting rights plaintiffs cannot prevail by demonstrating that the minority group's candidate or issue of choice was defeated in a single election. Gingles, 478 U.S. at 57, 106 S. Ct. 2752; City of Niagara Falls, 65 F.3d at 1009. Consequently, Dr. Kousser's data regarding state ballot measures also fails to raise a triable issue of fact regarding Gingles precondition three.

In light of the fact that the record before us reflects that Latino candidates in the area of SD 27 are not ever—much less regularly—defeated by an anti-Latino bloc vote, we conclude that the district as configured by the legislature does not violate § 2 of the Voting Rights Act.^[38] Latinos enjoy plurality status in SD 27. Whether it would have been preferable for the legislature to draw another district such as SD 30, in which Latinos are assured of electing a Latino candidate, or to construct a district such as SD 27, in which Latinos will undoubtedly exert strong influence on whoever the office-holder may be, is an interesting question of politics and political theory. However, because plaintiffs' challenge to SD 27 as presently drawn fails to meet Gingles pre-condition three, the legislature's choice must be upheld.

C. Congressional District 28

Plaintiffs also bring a traditional § 2 vote dilution claim regarding CD 28. As discussed supra, in the context of plaintiffs' racial gerrymandering claims, CD 28 is a diverse San Fernando Valley district in which Latinos constitute a plurality of the total population and the VAP. It, like SD 27, is located in Los Angeles County, and thus much of our discussion with respect to SD 27 is pertinent here as well.

1. Gingles Pre-Conditions 1 and 2

We need not address the first two Gingles pre-conditions at length. For the same reasons that plaintiffs have created a question of material fact with respect to the first two-preconditions in the case of SD 27, they have unquestionably done so with respect to CD 28. First, the fact that the district plaintiffs propose is 66.99% Latino with a 61.77% Latino VAP raises a triable issue of fact as to whether it would be an effective Latino majority district. Second, there can be no doubt that as to Gingles pre-condition two, the cohesion of the Latino vote, plaintiffs have similarly raised a triable question of fact. As noted above, county-wide Latinos vote for Latino candidates at rates well exceeding 90%. Moreover, in the two Los Angeles City Council Districts that comprise a substantial part of CD 28, Dr. Kousser reports that in 2001, nearly 98% of Latinos voted for the liberal Latino candidate for mayor, and 94% of Latinos voted for the more conservative Latinos candidate for city attorney.

2. Gingles Pre-Condition 3

As with SD 27, however, plaintiffs fail to offer evidence that raises a question of material fact regarding the third Gingles pre-condition. Like SD 27, the undisputed factual record shows that Latinos candidates can and do win elections in the territory that constitutes CD 28. Here, unlike in SD 27, plaintiffs do point to one endogenous election in which a Latino-preferred candidate was defeated: in a primary election in 1998, the Latino mayor of the City of San Fernando, Raul Godinez, challenged long-term incumbent Rep. Howard Berman in a Democratic primary election. Berman won approximately two-thirds of the vote, and the vote was undoubtedly polarized: three of four Latinos voted for Godinez, while nine of ten whites voted for Berman.^[39] The election, however, does not raise a triable issue of fact as to whether Latino-preferred candidates are regularly defeated by a non-Latino voting bloc. Even taking every reasonable inference in plaintiffs' favor, we cannot say that this single election—in which a better-funded, popular, well-known and highly effective incumbent defeated a novice primary challenger whose political base was the tiny City of San Fernando—creates a question of fact regarding Gingles precondition three in light of the uncontradicted evidence regarding other elections set forth below.^[40]See Gingles, 478 U.S. at 57, 106 S. Ct. 2752 ("Because loss of political power through vote dilution is distinct from the mere inability to win a particular election, ... a pattern of racial bloc voting that extends over a period of time is more probative of a claim that a district experiences legally significant polarization than are the results of a single election.").

With the exception of the Berman-Godinez primary, all of the record evidence demonstrates that a substantial number of white voters in the San Fernando Valley are willing to support Latino and Latino-preferred candidates, and that white-bloc voting does not result in the regular defeat of Latino or Latino-preferred candidates. Indeed, such evidence is more plentiful than that related to SD 27.^[41] Most telling are the results of the 2001 City of Los Angeles mayoral and city attorney elections.^[42] In the general election for city attorney, the Latino candidate, Rocky Delgadillo received 44.7% of the non-Latino vote and 93.5% of the Latino vote in the two Los Angeles City Council Districts that constitute a substantial portion of CD 28. With the assistance of white cross-over votes, Delgadillo won the precise area that now comprises CD 28, 52% to 48%, even though the other candidate, who was white, was the incumbent Los Angeles City Council member for portions of the district.^[43] In the general election for mayor, the Latino candidate, Antonio Villaraigosa, won 30.5% of the non-Latino vote in addition to 97.5% of the Latino vote in the two overlapping city council districts. More important, Villaraigosa handily defeated his opponent, also a white candidate, and an incumbent city-wide office-holder, in the precincts that now constitute CD 28: 54.7% to 45.3%.^[44]

Even more recently, a special election was held in Los Angeles City Council District 2, which substantially overlaps CD 28. It pitted a white candidate, Wendy Gruel, against a Latino, Tony Cardenas. Although Gruel defeated Cardenas by less than 300 votes out of a district-wide total of 29,351 votes cast, Cardenas won the precincts that are within CD 28 in both the primary and general elections. CD 28 has a slightly higher Latino voting-age population as a whole than do its precincts that are in City Council District 2. Thus, the results of this election also support the conclusion that a non-Latino bloc does not regularly defeat Latino, or Latino-preferred, candidates, but that, to the contrary, Latino candidates prevail within the district with some regularity.^[45]

Plaintiffs contend that the results of certain statewide ballot initiatives, most notably Proposition 187, demonstrate that CD 28 is a district in which the non-Latino voters are hostile to Latino interests. Proposition 187, which passed in 1994, proposed denying public services to undocumented aliens. Plaintiffs cite figures from all the Assembly, Senate, and Congressional districts that overlap to any extent the version of the district that MALDEF proposed in July and/or September 2001 during the course of the redistricting process. The results of this study demonstrate that Latinos voted "no" in margins far higher than non-Latinos. As defendants point out, this data cannot raise a triable issue of fact because it includes vast swaths of territory outside of CD 28. The State Senate, Assembly and Congressional districts that Dr. Kousser studied include such communities as Sunland/Tujunga, Montrose/La Crescenta, Glendale, San Marino, and Temple City. All of these areas are far outside the boundaries of CD 28 as drawn by the legislature. The inclusion of these communities in Dr. Kousser's study altered the demographics of the population being studied significantly from that found within the legislature's CD 28.^[46]

The results of the vote on Proposition 187 in the specific precincts that the legislature transferred in or out of CD 28 also support the conclusion that plaintiffs have failed to raise a triable issue of fact regarding Gingles pre-condition three. Those results are striking: In the area that the legislature removed from CD 28 during the 2001 redistricting, and that plaintiffs challenge as a dilution of the Latino vote, 39% of the population voted no on Proposition 187. However, in the area added to CD 28 in the south part of the Valley, which plaintiffs seek to have removed from the district, 51% of voters voted no. Thus, if opposition to Proposition 187 is indicative of a favorable attitude towards Latinos and an absence of discriminatory animus, then it would appear that the changes to CD 28 rendered it a more favorable district for Latino candidates.

In CD 28, as in SD 27, plaintiffs rely on Dr. Kousser's theory of "Latino electoral success" to argue that Latino candidates will regularly be defeated by non-Latino bloc voting. However, for the reasons we have stated, in this case the third Gingles pre-condition cannot be satisfied by a theoretical projection of a minority candidate's chances of winning or losing that is based on statewide figures; Gingles demands a "searching practical evaluation of past and present reality" in the particular area in which the challenged districts are located. 478 U.S. at 48, 106 S. Ct. 2752. The evidence related to CD 28 gives rise to only one reasonable conclusion: that the district, as presently constituted, is a district in which neither Latinos nor Latino-preferred candidates are regularly defeated by a non-Latino bloc vote. To the contrary, Latino candidates who run in the area covered by the district generally prevail. Accordingly, the 2001 redistricting plan does not have the effect of diluting the minority vote in CD 28.

D. Conclusion

Because no question of triable fact exists regarding Gingles pre-condition three with respect to plaintiffs' traditional § 2 challenges to SD 27 and CD 28, defendants' motion for summary judgment is granted as to those claims.

III. Plaintiffs' Intentional Vote Dilution Claims

Plaintiffs' final set of claims alleges that in amending the boundary lines of CD 28 and CD 51, the legislature intentionally diluted the vote of Latinos in violation of both the Equal Protection Clause and § 2 of the Voting Rights Act. In order to prove intentional vote dilution, plaintiffs must both demonstrate the requisite type of discriminatory intent, Bolden, 446 U.S. at 66, 100 S. Ct. 1490, and establish dilutive effects on "the voting potential of racial or ethnic minorities." Id.; Garza v. County of Los Angeles, 918 F.2d at 771 ("Even where there has been a showing of intentional discrimination, plaintiffs must show that they have been injured as a result.").

A. Discriminatory Intent

We note at the outset that the type of racial intent alleged by plaintiffs appears to be quite different from the type of intent that Bolden requires that plaintiffs prove in intentional vote dilution cases. As discussed earlier, no opinion in Bolden gained the votes of a majority of the justices. Nevertheless, a majority did agree that in order to prove an intentional vote dilution claim, a plaintiff must show invidious discriminatory intent. Bolden characterizes the intent required in such cases as a purpose "invidiously to minimize or cancel out the voting potential of racial or ethnic minorities." 446 U.S. at 66, 100 S. Ct. 1490 (citations omitted). In order to prove that purpose, Bolden continues, a plaintiff must show that "the disputed plan was conceived or operated as a purposeful device to further racial discrimination." Id. (internal alterations and citations omitted).

Assuming plaintiffs' proffered evidence to be true, we seriously question whether it raises a genuine issue of fact with respect to the type of invidious intent required under Bolden. Certainly, the redistricting plan as a whole and the distribution of congressional seats within Los Angeles County does not reflect such an intent. The legislature had the opportunity to create only one additional seat as a result of the most recent decennial census. The seat it created is in a majority-Latino district located in the southeastern portion of Los Angeles County. The district is concededly safely Democratic; Democrats constitute 56% of the electorate. A Latina is the Democratic nominee in this district in the upcoming general election. Her Republican opponent is a Latino. Accordingly, it is fair to conclude that the 2001 redistricting plan was designed to enhance the representation of Los Angeles County Latinos in Congress. As a result of redistricting, five of fifteen (33 1/3%) congressional seats with some territory in Los Angeles County will almost certainly be held by Latinos, although Latinos constitute only 24% of the county's registered voters, 44% of the total county population, probably somewhat less than a third of the votingage population, and an even lower percentage of the citizen voting-age population.^[47]Cf. De Grandy, 512 U.S at 1017-18, 114 S. Ct. 2647 (holding, in an effects case, that rough proportionality in minority representation, although not a "safe harbor," will often be a defense to vote dilution claims).

The evidence in the record is also uncontroverted that Latino legislators and interest groups played a significant role in the 2001 redistricting process with respect to both CD 28 and CD 51. This case is therefore unlike Rogers v. Lodge, in which plaintiffs established invidious intent largely through the exclusion of the minority group from the procedures that established district lines and set the rules for the political process. 458 U.S. at 624, 102 S. Ct. 3272; see also Garza, discussed supra, at pp. 1239-40. To the contrary, here the evidence is undisputed that the legislature heard extensive testimony from Latino organizations during the redistricting process, and that the redistricting plans were significantly revised to address Latino concerns. For instance, the inclusion of Imperial County in CD 51 was a result of lobbying by Latino advocacy organizations, as was the increase in the number of Latino voters in CD 28 in the second draft of the redistricting plan.

Moreover, California's political system is far from closed to Latinos. The Legislature as a whole, as noted earlier, is 22.5% Latino. It has an extremely active Latino caucus; both the Assembly and Senate redistricting committees included Latino members. At the time the redistricting bill was enacted, Latinos holding leadership positions in the legislature included the Senate Majority Leader, Richard Polanco; the Senate Majority Whip, Richard Alarcon; the Senate Judiciary Committee Chair, Martha Escutia; the Assembly Floor Leader, Marco Firebaugh; the Assembly Assistant Majority Leader, Juan Vargas; and the Assembly Budget Committee Chair, Tony Cardenas; as well as the Lieutenant Governor, Cruz Bustamante, who serves as President of the California Senate. In addition, at the time the redistricting bill was being considered, two of the three most recent Speakers of the Assembly were Latinos. Moreover, the record reflects that Latinos are not excluded from the internal hierarchy of the majority political party in California; long-time California Democratic Party Chairman Art Torres is also Latino, and the party has nominated Latino candidates for statewide office.

Plaintiffs' intent allegations regarding the two districts are similar; they contend, essentially, that the legislature deliberately limited the number of Latinos in each district, not as a consequence of any hostility towards Latinos or their role in the political process, but in order to render the incumbents in these districts less susceptible to primary challenges from Latino candidates. With respect to CD 28, plaintiffs allege that map-drawer Michael Berman intentionally removed certain Latino communities in order to reduce the chances that the incumbent congressman, his brother, would face a challenge from a Latino primary opponent. Assembly Member Juan Vargas, a Latino representative from San Diego who was a member of the legislature's redistricting committee, has offered his opinion that by accepting the plan proposed in part by Michael Berman, the legislature operated with a discriminatory intent. Additionally, plaintiffs put forward the testimony of a Latino community activist, Xavier Flores, regarding conversations with Rep. Brad Sherman, who represents CD 27, the district adjacent to CD 28. According to Flores, Sherman was aware that Michael Berman sought to minimize the number of Latinos in CD 28, and Berman explained to Sherman that the first draft of the congressional districting plan artificially lowered the number of Latinos in that district so that there would be an opportunity to negotiate over the final figure. As a result of the 2001 redistricting, the Latino population of CD 28 was reduced from 65% to 56%, and the Latino VAP from 60% to 49%.

The allegations with respect to CD 51 are similar. Plaintiffs contend that the legislature removed San Diego County neighborhoods from the district so as to cap the district's Latino voting strength at 40% of the VAP. The foundation for the claim is primarily the testimony of Assembly Member Vargas, who offers the opinion that the legislature's aim in drawing the revised CD 51 was to insulate the incumbent, Congressman Filner, against a successful primary challenge by a Latino. Vargas himself challenged Filner in what was, by both parties' accounts, an expensive, nasty primary race in 1996, which Filner won 54%-46%. Vargas asserts that many of the incumbent members of the California congressional delegation, including Filner, paid $20,000 per person to obtain the services of the legislature's chief map-drawing consultant, Michael Berman. In exchange, he contends, Berman, took steps to render their districts less susceptible to potential challengers, both partisan and ethnic.

Plaintiffs admit that they do not allege that defendants were motivated by racial hostility. Nor do they suggest that there was any desire to effectuate invidious racial discrimination generally. Although we assume, for summary judgment purposes, the truth of plaintiffs' intent evidence, and of their charge that the legislature sought to limit the number of Latino voters in the two districts at issue, given the background and record of California's 2001 redistricting, the evidence does not support an inference that the legislature intended to marginalize a racial group politically through invidious discrimination, or invidiously to maintain a system that perpetuates racial discrimination. Thus the intent appears not to be of the type that the Supreme Court held necessary for an intentional vote dilution claim in Bolden, Rogers, and Bandemer. Leaving aside plaintiffs' proffers, the other evidence as to intent reflects a complex set of legislative motivations that comprehended several goals, including protecting incumbents, ensuring adequate representation for Latinos and other minority groups through the establishment of majority-minority districts such as the new CD 39, and advancing partisan interests.^[48] Given the facts and circumstances in the record before us, including (1) the use of traditional districting principles to establish the districts in question, (2) the absence of any legal necessity to create another new majority-minority Congressional district in addition to the one being newly-created in the redistricting statute, and (3) the high degree of Latino representation and participation in the redistricting process, we strongly doubt that the 2001 redistricting statute was a "purposeful device to further racial discrimination," in whole or in part. Bolden, 446 U.S. at 66, 100 S. Ct. 1490 (internal alterations and citations omitted).

B. Effects of Intentional Vote Dilution

We do not, however, rest our grant of summary judgment on the lack of invidious intent, for even if the plaintiffs could establish such intent, they have failed to offer proof of the necessary dilutive effects. Plaintiffs in an intentional vote dilution case still bear the burden of proving a dilutive effect. Rogers, 458 U.S. at 621, 102 S. Ct. 3272 (holding that a minority group cannot "prove purposeful discrimination" absent sufficient evidence of dilutive effect); Garza, 918 F.2d at 771.

The necessary showing of effects for a traditional § 2 claim of vote dilution is well-established; as discussed above, the Gingles factors and the De Grandy totality-of-the-circumstances inquiries set forth settled principles to be applied in determining whether a plaintiff can establish the effects that are necessary for a statutory violation. In contrast, the effects standard for an intentional vote dilution claim is uncertain, largely because of a dearth of precedent. The cases provide little direct authority as to the requisite degree of dilutive effect for an intentional discrimination claim under either the constitution or the statute. This is so in part because direct evidence of discriminatory intent is relatively rare, Price Waterhouse v. Hopkins, 490 U.S. 228, 271, 109 S. Ct. 1775, 104 L. Ed. 2d 268 (1989) (O'Connor, J., concurring), and thus plaintiffs generally prefer to bring traditional § 2 vote dilution cases where intent need not be shown, rather than constitutional or statutory intentional vote dilution claims. See Samuel Issacharoff, Pamela S. Karlan & Richard H. Pildes, THE LAW OF DEMOCRACY 406-07 (1998). In fact, since the 1982 amendments to the Voting Rights Act, which eliminated the intent requirement for statutory vote dilution claims, there has been a virtual absence of intentional vote dilution litigation.^[49]

Plaintiffs urge us to adopt a standard of effects in intentional vote dilution cases, both constitutional and statutory, that is considerably less demanding than that required in traditional § 2 cases. In support of this approach, plaintiffs cite Garza v. County of Los Angeles, 918 F.2d 763 (9th Cir.1990), a case that, like this one, included both constitutional and statutory intentional vote dilution claims. Garza, however, does not provide the clear direction that the plaintiffs assert it does. In that case, the court stated that a standard "less rigorous" than the § 2 effects standard applies to intentional vote dilution claims; nevertheless, the court then went on to cite White v. Regester—the case that established the very effects test adopted in 1982—as well as the verbatim language of § 2 itself, as the appropriate measure for determining effects in intentional vote dilution cases. Id. at 771. More important, Garza relaxed only the first Gingles precondition. Garza held that in light of plaintiffs' contentions that the Los Angeles County Board of Supervisors intentionally fragmented the Latino population so as to prevent Latinos from constituting an effective majority district, the plaintiffs did not need to show that the district they proposed would immediately be such a district.

We agree that, where invidious intent exists in a vote dilution case, it may be appropriate to relax the first or even second of the Gingles pre-conditions, as well as to consider intent in connection with the "totality of the circumstances" inquiry.^[50] We do not accept defendants' contention that in a case in which plaintiffs allege an intentional violation of the Fourteenth Amendment, proof of discrimination is wholly irrelevant. Nevertheless, plaintiffs in vote dilution cases must still show a practical effect on the minority group's ability to elect representatives of choice, whether or not intent is shown. In Gingles, the Court recognized this requirement of a practical effect primarily through the inclusion of pre-condition three—that in order to establish a traditional § 2 claim, a plaintiff must establish that a minority group's preferences are regularly defeated by a non-minority bloc of voters. In our view, the irreducible minimum in intentional vote dilution cases is similar.

The essence of any successful vote dilution claim must be that the ability of a minority community to elect representatives of choice is adversely affected. If the legislature intentionally renders impotent the will of the minority community by dividing it among two or more districts in which the other voters are hostile to minority candidates, or by submerging it in a multi-member district in which a non-minority bloc refuses to support minority candidates, then a vote dilution claim will likely prevail. However, if a district is drawn in which Latinos constitute a plurality, non-Latino voters have shown that they are willing to vote for Latino candidates, and Latino candidates receive a majority of the overall vote with some regularity, then the most basic and necessary dilutive effect is lacking. That is precisely the case here. As respects both CD 28 and CD 51, the right of Latinos to elect candidates of their choice has not been abridged.

1. Congressional District 28

Plaintiffs' intentional dilution claims regarding CD 28 fail to raise a question of material fact because of the uncontested evidence described earlier. Latino candidates win elections in the territory that constitutes CD 28, and they do so with the support of non-Latino voters. It would undoubtedly be easier for Latinos to elect candidates of choice in the version of the district proffered by plaintiffs. However, in order for minority group members to establish a redressable vote dilution injury, they must at the very minimum demonstrate that they are unlikely to be able to elect minority candidates due to the hostility of non-minority voters. Because plaintiffs have not raised a question of material fact with respect to this essential part of the vote dilution inquiry, we hold that the intentional vote dilution claims are not supported by sufficient evidence of effects, and summary judgment in favor of the defendants must be granted.^[51]

2. Congressional District 51

We need not deliberate as to whether a genuine question of fact exists with respect to Latino voters' ability to elect candidates of choice in CD 51. Plaintiffs admit that the legislature established a district that is a fair opportunity district for Latinos—one in which they can elect candidates of their choice. Indeed, plaintiffs' expert witness, Dr. Kousser, conceded in his report that "there was no discriminatory effect in the drawing of congressional district 51." The population in the district is 53% Latino, the VAP is 49% Latino, and the registered voters are 39.84% Latino. This is a modest increase from the district as it was previously constituted. Plaintiffs' admission regarding the effectiveness of CD 51 was thus compelled by the facts, and forecloses any claim of vote dilution, constitutional or statutory. Garza, 918 F.2d at 771.

In light of the admission that the district is a fair one, plaintiffs advance another, wholly unprecedented argument in support of their intentional dilution claim. Plaintiffs implicitly acknowledge that Latino voters who reside in CD 51 do not suffer a vote dilution injury, as they vote in a district that provides Latinos a fair opportunity to elect candidates of choice. Accordingly, they assert only that the voters who were not placed in CD 51 suffered a vote dilution injury because they were denied the privilege of residing in a majority-Latino district. No court has ever found such an injury, and plaintiffs cite no cases in support of their novel proposition. More important, to hold that such an injury is cognizable would throw redistricting law into disarray and render almost any remedial § 2 district subject to challenge. As Justice Powell observed, redistricting necessarily requires legislatures to consider individual voters in the context of their membership in various groups—especially racial and ethnic groups, but also political parties and other entities. Davis v. Bandemer, 478 U.S. 109, 167, 106 S. Ct. 2797, 92 L. Ed. 2d 85 (1986) (Powell, J., concurring and dissenting) ("The concept of ``representation' necessarily applies to groups: groups of voters elect representatives, individual voters do not."). It is thus the group and not the individual whose voting power must be diluted if a vote dilution claim is to be upheld. See Heather K. Gerken, Understanding the Right to An Undiluted Vote, 114 HARV. L. REV. 1663, 1678-79 (2001).

Sections 2 and 5 of the Voting Rights Act compel states and localities to draw majority-minority districts under certain circumstances; if each individual minority group member not included in such a district had standing to challenge the creation of the district because he was not placed within it, every § 2 remedial district would be vulnerable to constitutional challenge. When we analyze a claim of vote dilution, we necessarily examine the voting power of the group: individual voters who are members of the group, and are assigned to an otherwise-lawful neighboring district, cannot bring a vote dilution challenge where the group is capable of electing representatives of choice in the district from which the voters claim to have been improperly excluded. There is simply no dilution of the group's voting power as a result of that assignment. Accordingly, summary judgment is appropriate on plaintiffs' intentional vote dilution claims on behalf of voters not placed in CD 51.

CONCLUSION

The evidence in the record before us permits only one conclusion: drawing the necessary inferences for the purposes of summary judgment, it is apparent that the legislature deliberately chose not to create majority-Latino districts in CD 28 and CD 51, and that the effect of its decisions with respect to SD 27 was not to create such a district. These legislative actions, however, violated neither the Voting Rights Act nor the Constitution. The legislature lawfully chose to exercise its considerable discretion in redistricting to create three districts in which Latinos are the largest single ethnic or racial group, in which Latinos will inevitably play a significant role in the election of representatives, and in which Latino candidates may well be elected to office when the current incumbents leave their posts if not before. Moreover, Latinos participated fully in the redistricting process and elected Latino representatives, many of whom hold important positions in the state legislature, overwhelmingly approved the redistricting plan. Accordingly, defendants' motions for summary judgment are GRANTED.

It is so ordered.

NOTES

[1] MALDEF, founded in 1968, is the most prominent national organization devoted to securing the civil rights of the more than 35 million Latinos living in the United States. Its national headquarters is in Los Angeles County.

[2] Defendants have challenged the admissibility of some of plaintiffs' evidence, principally on hearsay and privilege grounds. The admissibility of some of the testimony relating to Michael Berman's role in the redistricting process is disputed, as is much of plaintiffs' other evidence related to legislative intent. We need not rule on these objections. Rather, we assume the challenged evidence is admissible, because it does not affect the results we reach on the various claims.

[3] The legislature was also required to re-draw the boundary lines for the Assembly and State Board of Equalization districts. Those new district lines were enacted in a separate piece of legislation. That bill passed the legislature simultaneously with the Congressional and Senate redistricting measure. It is not at issue in this litigation.

[4] For convenience, in this opinion we will refer to the defendants and defendants-intervenors collectively as "defendants." Several Los Angeles-area Latino community organizations, which argue that the redistricting plan adequately protects Latinos' ability to elect representatives of choice applied to intervene as defendants, as did black and Filipino groups from San Diego. Those applications were untimely, and we therefore permitted the groups to participate instead as amici curiae.

[5] Plaintiffs' complaint alleges only two claims for relief—one under the Voting Rights Act, 42 U.S.C. §§ 1973 et seq., and one under the Fourteenth Amendment. Plaintiffs' Fourteenth Amendment claim alleges, inter alia, that the Congressional Districts adopted have "no rational explanation except as an effort to separate voters on the basis of racial/ethnic origin." It further alleges that each district constitutes "an intentional gerrymander of the Latino population in violation of Plaintiffs' Equal Protection Rights guaranteed by the Fourteenth Amendment to the United States Constitution." From the beginning of the litigation, the parties and the court have understood this single claim to allege a constitutional violation under the theory articulated in Shaw v. Reno, 509 U.S. 630, 113 S. Ct. 2816, 125 L. Ed. 2d 511 (1993) (", and its progeny, as well as a traditional Equal Protection vote dilution claim. Plaintiffs' Voting Rights Act claim is similarly hybrid." It alleges that CD 28 and SD 27 were adopted "with the discriminatory purpose of fragmenting the Latino population in order to dilute the effect of the Latino vote in future elections.") It further alleges that the districts "do not afford Plaintiffs an equal opportunity to elect representatives of their choice."

[6] Forty of North Carolina's 100 counties are covered jurisdictions under § 5 of the Voting Rights Act, 42 U.S.C. § 1973c, which means that North Carolina must receive approval from either the Justice Department or a three-judge court of the District of Columbia before implementing any change to voting procedures, including district lines. Shaw I, 509 U.S. at 634, 113 S. Ct. 2816. As a result of the 1990 North Carolina redistricting, that state, which has a voting age population that is 20% black, elected its first two black congressmen since Reconstruction. Id. at 659, 113 S. Ct. 2816 (White, J., dissenting).

[7] The legislative redistricting process in California that followed the 1990 census was completed by special masters appointed by the California Supreme Court after the legislature and governor reached an impasse and could not adopt a redistricting plan in time for the 1992 primary elections. Wilson v. Eu, 1 Cal. 4th 707, 4 Cal. Rptr. 2d 379, 823 P.2d 545, 548 (1992). We refer to that redistricting plan in this opinion as the "Special Masters' Plan."

[8] We have doubts as to whether the type of direct evidence put forward by plaintiffs regarding the intent of the legislature as a whole is sufficient to raise a triable question of fact regarding that subject. City of Las Vegas v. Foley, 747 F.2d 1294, 1299 (9th Cir. 1984) (holding that the subjective intent of individual legislators is irrelevant to establishing the motivation of the entire legislature). Plaintiffs have adduced, for example, the testimony of a member of the legislature, who offers an opinion regarding the intent of the body. For the purpose of deciding this motion, however, we assume that such testimony correctly describes the legislature's intent.

[9] This is so not because plaintiffs here are members of a minority group. Plaintiffs characterize defendants' arguments against the Shaw claims in this case as asserting that non-white plaintiffs cannot bring a racial gerrymandering claim. To the extent that this characterization is accurate, we reject the argument. The language of the racial gerrymandering cases makes clear that the claim may be brought by any person regardless of race. Shaw I, 509 U.S. at 650-51, 113 S. Ct. 2816; see Diaz v. Silver 978 F. Supp. 96, 97-98 (E.D.N.Y.1997) (three-judge court) (black and Latino plaintiffs have standing to challenge an alleged racial gerrymander that created a majority-Puerto Rican congressional district).

[10] In Bush v. Vera, the Supreme Court struck down Texas's 30th Congressional District as a racial gerrymander. That district featured demographics that in some respects appear to be similar to the demographics of CD 28 and CD 51; its population was 50.9% black, 17.1% Hispanic, 31.4% white, and 2.4% other. 517 U.S. at 965, 116 S. Ct. 1941; Javier Aguilar, Congressional Redistricting in Texas: Time for a Change, 27 STETSON L. REV. 781, 802 n. 154 (1998). However, there is a fundamental difference between that district and those at issue here. The Texas district was found to be a racial gerrymander that constituted a race-based black district, and in fact blacks comprised more than half of the population. In contrast, in CD 28 and CD 51—which plaintiffs contend, in essence, were drawn as "white" districts—the largest single group is Latinos, and whites constitute a far smaller percentage of the populations in each.

Moreover, on the basis of its unusual geographic configuration, which included an overwhelmingly black core from which tentacles extended to grab far-flung black neighborhoods, the Texas district could only be reasonably understood by any objective observer as a "black" district. See discussion n. 12, infra. That district's configuration every respect is starkly different from those of CD 28 and 51.

[11] Indeed, plaintiffs' expert, Dr. J. Morgan Kousser, stated, in a paper prepared and submitted to the California legislature during its redistricting efforts, that the "concerns that the Supreme Court mentioned in Shaw v. Reno and subsequent related opinions cannot apply to districts in which there is no ethnic majority of voters." While, on such a matter of law, Dr. Kousser's opinion is not binding on this court, it is telling that plaintiffs' expert, who served as a consultant to plaintiffs' counsel, MALDEF, during the redistricting process, holds such a view.

[12] The Court's discussion in each of these passages is instructive. See Miller, 515 U.S. at 917, 115 S. Ct. 2475 ("In our view, the District Court applied the correct analysis, and its finding that race was the predominant factor motivating the drawing of the Eleventh District was not clearly erroneous. The court found it was ``exceedingly obvious' from the shape of the Eleventh District, together with the relevant racial demographics, that the drawing of narrow land bridges to incorporate within the district outlying appendages containing nearly 80% of the district's total black population was a deliberate attempt to bring black populations into the district.... Although by comparison with other districts the geometric shape of the Eleventh District may not seem bizarre on its face, when its shape is considered in conjunction with its racial and population densities, the story of racial gerrymandering seen by the District Court becomes much clearer."). See also Vera, 517 U.S. at 965, 116 S. Ct. 1941 ("Fifty percent of the [District 30's] population is located in a compact, albeit irregularly shaped, core in south Dallas, which is 69% African-American. But the remainder of the district consists of narrow and bizarrely shaped tentacles—the State identifies seven ``segments'—extending primarily to the north and west.... Over 98% of the district's population is within Dallas County, ... but it crosses two country lines at its western and northern extremities...."); Id. at 972, 116 S. Ct. 1941 ("District 30's combination of a bizarre, noncompact shape and overwhelming evidence that that shape was essentially dictated by racial considerations of one form or another is exceptional; Texas Congressional District 6, for example, which Justice Stevens discusses in detail, ... has only the former characteristic. That combination of characteristics leads us to conclude that District 30 is subject to strict scrutiny."); Id. at 973-74, 116 S. Ct. 1941 ("District 18[] ... ``has some of the most irregular boundaries of any congressional district in the country[,] ... boundaries that squiggle north toward Intercontinental Airport and northwest out radial highways, then spurt south on one side toward the port and on the other toward the Astrodome.'... Its ``many narrow corridors, wings, or fingers ... reach out to enclose black voters, while excluding nearby Hispanic residents.'... District 29 ... resembles ``a sacred Mayan bird, with its body running eastward along the Ship Channel from downtown Houston until the tail terminates in Baytown. Spindly legs reach south to Hobby Airport, while the plumed head rises northward almost to Intercontinental. In the western extremity of the district, an open beak appears to be searching for worms in Spring Branch. Here and there, ruffled feathers jut out at odd angles.' ... Not only are the shapes of the districts bizarre; they also exhibit utter disregard of city limits, local election precincts, and voter tabulation district lines.") (citations omitted); Cromartie II, 532 U.S. at 240, 243, 121 S. Ct. 1452 (noting that the Court had considered "the district's snakelike shape, the way in which it split cities and towns and its heavily African-American (47%) voting population," and "found [such factors] insufficient to support summary judgment").

[13] This is best illustrated by the district maps provided by plaintiffs, which demonstrate that the core of the districts remained the same, although different territory was added to or removed from the perimeters. Plaintiffs fail to provide the precise number of voters whose removal from CD 28 gives rise to the alleged Shaw violation. An approximation can be calculated, however, using plaintiffs' statistical evidence, which reveals that the district now contains 14,438 fewer Latino registered voters than it did before the boundary lines were re-drawn. The district as a whole contains 234,017 registered voters. In the case of CD 51, the Latino population increased as a result of the change, and the number of Latino registered voters grew by 14,833. The district as a whole contains 251,256 registered voters.

[14] Chief among those other districting principles is the federal constitutional rule that congressional districts must be precisely equipopulous. Karcher v. Daggett, 462 U.S. 725, 730, 103 S. Ct. 2653, 77 L. Ed. 2d 133 (1983); Gray v. Sanders, 372 U.S. 368, 381, 83 S. Ct. 801, 9 L. Ed. 2d 821 (1963). In California, the ideal congressional district size for redistricting purposes in 2001 was 639,088. The size of the two districts, as reconfigured in 2001, was identical: the near-ideal 639,087.

[15] We recognize that a showing of bizarre shape is not a necessary condition for a successful Shaw claim. Miller, 515 U.S. at 913, 115 S. Ct. 2475. Nevertheless, shape remains a highly probative method of proving that a district is primarily race-based, and the vast majority of successful racial gerrymandering claims have challenged bizarrely shaped districts. See Chen v. City of Houston, 206 F.3d 502, 509-13 (5th Cir.2000); Diaz, 978 F.Supp. at 118; Moon v. Meadows, 952 F. Supp. 1141, 1147 (E.D.Va.) (three-judge court), aff'd 521 U.S. 1113, 117 S. Ct. 2501, 138 L. Ed. 2d 1006 (1997); but see Smith v. Beasley, 946 F. Supp. 1174, 1185 (D.S.C.1996) (three-judge court).

[16] The alternative district proposed by the Latino advocacy groups during the redistricting process eliminated these communities from CD 51. The district originally proposed by plaintiffs in connection with their application for temporary restraining order similarly excised these communities. As part of their submission in opposition to the motions for summary judgment, plaintiffs have proposed a district that includes these communities in CD 51.

[17] This version of incumbency protection differs from plaintiffs' allegations that the state intentionally limited the Latino population in certain districts so as to protect incumbents from a primary challenge. Here, the state justifies its shift of Latinos on the ground that the addition of the heavily Democratic Latino precincts to a marginal Democratic incumbent's district will protect the incumbent against a Republican general election opponent.

[18] The evidence plaintiffs proffer regarding intent is the following: (1) that Latino communities that had been placed in CD 51 by the Special Masters' Plan ten years earlier were placed within another district; (2) that Assembly Member Juan Vargas understands this was done because the principal map drawer, whose firm had been paid $20,000 by the incumbent, wished to ensure that the percentage of registered Latino voters in the district did not exceed 40%; (3) that Latino advocacy groups warned the legislature that their proposed districting plan violated Latinos' civil rights and nonetheless rejected alternative plans that would have placed the Latino communities at issue within CD 51; (4) that, as contrasted with the Assembly, the State Senate utilized a secretive process, and made its intended district lines available to the public late in the process by way of data that was not easily accessible; (5) that the intended Senate-Assembly conference committee was scrapped and the districting legislation was placed in an entirely different bill than that originally proposed to move the legislation to the floor; and (6) that old CD 50, which became CD 51, saw a dramatic increase in the number of Latinos who were registered Democratic voters between 1990 and 2000, such that a Latino candidate was able to mount a serious challenge to the white incumbent. Plaintiffs also offer the expert opinion of Dr. Kousser based on these facts that race was the predominant factor in the drawing of district lines.

For purposes of this motion, we accept all of these facts as indicia of intent. We question, however, whether a legislature can be said to have acted with racially discriminatory intent simply because it did not adopt the districting preferences of advocates who asserted that its intended decision would violate the law. We further question whether, simply because a district has seen a large increase in a particular ethnic population, and there has been a difficult race between a member of that ethnic group and an white incumbent, any exclusion of members of the group from the district is evidence of racial motivation. This is particularly true where the percentage of voters included within that group was actually increased in the districting plan over what it had been at the time the election occurred. This is the case in CD 51, in which Latino registration increased from 35.04% to 39.84%, and Latino Democratic registration increased from 40.31% to 47.49%.

[19] To the extent that plaintiffs contend that rejection of the MALDEF redistricting plan constitutes evidence of discriminatory intent, we note that MALDEF's plan adhered to traditional districting principles less rigorously than did the legislature's final plan. Dr. Kousser asserts that cities and neighborhoods were split in both the legislature's and MALDEF's plan. Yet the facts are undisputed that the legislature's plan respected city boundaries more than the alternative proposed by MALDEF, and more than the remedial district proposed by plaintiffs for purposes of this litigation. The plans proposed by plaintiffs and MALDEF, moreover, deviate from the equipopulous standard that the legislature had to meet more than the plan ultimately adopted.

[20] The fact that a plan proposed by the Assembly map drawers would have strengthened Democratic registration in Davis' district even more than the plan ultimately adopted (or the alternative proposed by MALDEF during the redistricting process) does not alter the fact that Democratic registration in CD 53 was increased. While Dr. Kousser contends that the Democratic registration in Davis' district was reduced, this is belied by other evidence plaintiffs have submitted, which shows that it was in fact increased by more than 10%. It is also belied by plaintiffs' admission that "AB 632 increased the percentage of Congressional District 53 that is comprised of registered Democrats."

[21] Gonzalez's declaration consists of the hearsay repetition of conversations he allegedly had with Rep. Berman. Plaintiffs sought to depose three sitting members of Congress: Howard Berman, Brad Sherman, who represents the district adjacent to CD 28, and Robert Filner. Plaintiffs contend that these depositions would have revealed that the congressmen instructed Michael Berman to limit the number of Latinos in each district. The congressmen sought to quash the subpoenas on the ground that the exceptional circumstances required to justify the deposition of a high-ranking government official are absent here. We deferred ruling on the motion to quash in light of the pending summary judgment motions. We will, for purposes of this motion, assume that the depositions would have produced the information plaintiffs allege.

[22] In addition to their allegations regarding Michael Berman's intent and the statement of Sen. Perata, plaintiffs cite the following evidence that they contend demonstrates racial considerations were involved in constructing CD 28:(1) the horseshoe/stake configuration of CD 28 and neighboring CD 27; (2) the sequence of the legislature's two districting proposals, each of which split Latino communities in the northern part of the proposed district that had been joined under the Special Masters' Plan; (3) the fact that the map decreased the percentage of Latinos in CD 28; (4) the fact that, despite warnings by Latino advocacy groups that the legislature's proposed districting plan violated Latinos civil rights, the legislature rejected alternative plans that would have placed the Latino communities at issue within CD 28; (5) the fact that, as contrasted with the Assembly, the State Senate utilized a secretive process, and made its intended district lines available to the public late in the process by way of data that was not easily accessible; and (6) the fact that the district saw a dramatic increase in the number of Latinos who were registered Democratic voters between 1990 and 2000. As they did with CD 51, plaintiffs also offer the expert opinion of Dr. Kousser based on these facts that race was the predominant factor in the drawing of district lines. Consistent with our consideration of plaintiffs' evidence regarding CD 51, we accept these facts as indicia of intent for purposes of this motion. As noted, however, we question whether some of plaintiffs' evidence is probative of racially discriminatory intent. See n. 18, supra.

[23] The Special Masters appointed by the Supreme Court admittedly made districting decisions without regard to one of the most traditional of districting principles: politics. Wilson, 4 Cal. Rptr. 2d 379, 823 P.2d at 549 (noting, in adopting the Special Masters' plan, that the masters "excluded such political factors as the potential effects on incumbents or the major political parties."). The 2001 redistricting process, in contrast, was completed wholly by the so-called political branches, and the interests of partisan advantage and incumbency protection therefore assumed a major role. The final redistricting plan was passed by the votes of an overwhelming majority of the members of both political parties.

[24] Section 2 provides in full as follows:

(a) No voting qualification or prerequisite to voting or standard, practice, or procedure shall be imposed or applied by any State or political subdivision in a manner which results in a denial or abridgement of the right of any citizen of the United States to vote on account of race or color, or in contravention of the guarantees set forth in section 1973b(f)(2) of this title, as provided in subsection (b) of this section.

(b) A violation of subsection (a) of this section is established if, based on the totality of circumstances, it is shown that the political processes leading to nomination or election in the State or political subdivision are not equally open to participation by members of a class of citizens protected by subsection (a) of this section in that its members have less opportunity than other members of the electorate to participate in the political process and to elect representatives of their choice. The extent to which members of a protected class have been elected to office in the State or political subdivision is one circumstance which may be considered: Provided, That nothing in this section establishes a right to have members of a protected class elected in numbers equal to their proportion in the population

42 U.S.C § 1973.

[25] This first Gingles precondition has given rise to the question of whether a § 2 claim is cognizable when a legislature fails to create a feasible minority "influence district"—one in which the minority group constitutes a sufficiently large proportion of the district that it exerts a significant amount of influence on the elected representative. The Supreme Court has repeatedly declined to decide this question. De Grandy, 512 U.S. at 1009, 114 S. Ct. 2647; Voinovich v. Quilter, 507 U.S. 146, 152, 113 S. Ct. 1149, 122 L. Ed. 2d 500 (1993). Most, but not all, lower courts have generally rejected influence district claims under § 2. See Latino Political Action Committee v. Boston, 784 F.2d 409 (1st Cir.1986); Hastert v. State Bd. of Elections, 777 F. Supp. 634 (N.D.Ill.1991) (three-judge court); but see Armour v. Ohio, 895 F.2d 1078 (6th Cir.1990), vacated on other grounds en banc, 925 F.2d 987 (6th Cir.1991). No such claim is asserted here, and we therefore do not consider the question. Specifically, we do not consider whether the intentional dilution of a minority group's influence in a district can, without more, give rise to a cognizable claim. See infra, Section III, Intentional Dilution Claims.

[26] In the jargon of redistricting law, such claims are typically referred to as "submergence" claims. De Grandy, 512 U.S. at 1012-13, 114 S. Ct. 2647.

[27] Chief among the circumstances that the De Grandy Court held to be relevant to the § 2 inquiry are those listed in the Senate Judiciary Committee Report regarding the 1982 Amendments to § 2. S. REP. No. 97-417, at 28-29 (1982), reprinted in 1982 U.S.C.C.A.N. 177, 206-07. Those factors, frequently called the "Senate factors" or the "Zimmer factors," owing to their origin in Zimmer v. McKeithen, 485 F.2d 1297 (1973) (en banc), are set forth below:

(1) the extent of any history of official discrimination in the state or political subdivision that touched the right of the members of the minority group to register, to vote, or otherwise to participate in the democratic process;

(2) the extent to which voting in the elections of the state or political subdivision is racially polarized;

(3) the extent to which the state or political subdivision has used unusually large election districts, majority vote requirements, anti-single shot provisions, or other voting practices or procedures that may enhance the opportunity for discrimination against the minority group;

(4) if there is a candidate slating process, whether the members of the minority group have been denied access to that process;

(5) the extent to which members of the minority group in the state or political subdivision bear the effects of discrimination in such areas as education, employment and health, which hinder their ability to participate effectively in the political process;

(6) whether political campaigns have been characterized by overt or subtle racial appeals;

(7) the extent to which members of the minority group have been elected to public office in the jurisdiction;

(8) whether there is a significant lack of responsiveness on the part of elected officials to the particularized needs of the members of the minority group; and

(9) whether the policy underlying the state or political subdivision's use of such voting qualification, prerequisite to voting, or standard, practice or procedure is tenuous.

1982 U.S.C.C.A.N. at 206-07.

[28] Although we conclude that a question of material fact exists with respect to Gingles pre-condition one, we do not rest that conclusion on Dr. Kousser's analysis. We note that Dr. Kousser's opinion is based on Democratic Party registration figures, a measure that is unprecedented in redistricting law and that raises difficult analytical questions, particularly in view of the existence of a number of Latino Republican office-holders in California, and of recent efforts by the California Republican Party to increase its share of the Latino vote and registration. We need not determine the validity of Dr. Kousser's approach here because a question of material fact exists for other, unrelated reasons.

[29] In the abstruse language of redistricting, the term "endogenous elections" refers to elections for the particular office and district that is at issue. "Exogenous elections" are those held for other offices conducted in the same approximate geographic area. Clay v. City of St. Louis, 90 F.3d 1357, 1360 n. 6 (8th Cir.1996).

[30] See Nipper v. Smith, 39 F.3d 1494, 1540 (11th Cir.1994) (while not foreclosing consideration of races involving only white candidates, the court stated that "the most probative evidence of whether minority voters have an equal opportunity to elect candidates of their choice is derived from elections involving black candidates. Certainly, where only white candidates are competing for office in a particular jurisdiction, it is ``virtually unavoidable that certain white candidates would be supported by a large percentage of ... black voters.'... Such elections, however, may reveal little about the issue to be determined: the capacity for white bloc voting usually to defeat black candidates of choice. Particularly where voting is extremely polarized by race in elections in which black candidates participate, white-on-white elections in which a small majority (or a plurality) of black voters prefer the winning candidate seem comparatively less important.") (quoting Citizens for a Better Gretna, 834 F.2d at 502); Jenkins v. Red Clay Consolidated School District Board of Education, 4 F.3d 1103, 1128-29 (3d Cir. 1993) (noting that "white versus white elections tend to be less probative," the court stated that "[i]n deciding which, if any, of the white candidates were minority-preferred, the court must engage in a detailed, practical evaluation of the extent to which any particular white candidate was, as a realistic matter, the minority voters' representative of choice."); League of United Latin American Citizens, Council No. 4434 v. Clements, 999 F.2d 831, 864 (5th Cir.1993) (en banc) ("This court has consistently held that elections between white candidates are generally less probative in examining the success of minority-preferred candidates, generally on grounds that such elections do not provide minority voters with the choice of a minority candidate."), cert. denied, 510 U.S. 1071, 114 S. Ct. 878, 127 L.Ed.2d (1994); Westwego Citizens for Better Government v. Westwego, 872 F.2d 1201, 1208, n. 7 (5th Cir.1989) ("the evidence most probative of racially polarized voting must be drawn from elections including both black and white candidates.").

[31] Dr. Kousser's figures totally ignore the black and Asian vote. It is reasonable to assume that at least some of these voters supported the Latino candidates as well. However, because we must draw all reasonable inferences in plaintiffs' favor, we will assume for purposes of this opinion that the black and Asian communities voted for Latino candidates in no greater proportion than white voters.

[32] Defendants offer similar evidence that we also find of little probative value regarding statewide races in which only white candidates ran, i.e., past races won by Senators Barbara Boxer and Diane Feinstein, as well as Vice President Gore's 2000 presidential run.

[33] We need not consider the question of how to apply Gingles pre-condition three when no minority candidates have ever sought office in the area that is in dispute. In the areas that comprise SD 27 and CD 28 respectively, there is a sufficient record of how voters behave when Latino candidates are on the ballot.

[34] Some courts have noted that Gingles preconditions two and three are logically related in that they both may be proven by a showing of racially polarized voting. Nevertheless, those courts have not suggested that a showing of racial polarization in voting excuses a plaintiff from satisfying each of the three preconditions. "Evidence of racially polarized voting is the linchpin of a section 2 vote dilution claim, and is relevant to establishing two of the three elements set forth in the Gingles decision—the political cohesiveness of the minority group and the ability of the white majority usually to defeat the minority's preferred candidate." Westwego Citizens for Better Government v. Westwego, 946 F.2d 1109, 1118 (5th Cir.1991). Whether such polarization is sufficient to satisfy pre-condition three will depend, however, on the percentages of voters in the majority and minority groups, and the degree of polarization. Plaintiffs rely on the statement in Ruiz, 160 F.3d at 551, that "[r]acially polarized voting, one of the most important Senate factors, is the combined effect of the second and third Gingles preconditions (minority political cohesion and majority bloc voting"). Nothing in this observation, however, suggests that the second and third Gingles pre-conditions need not be proved separately. Indeed, a careful reading of Gingles reveals that the two are distinct inquiries. See Gingles, 478 U.S. at 56, 106 S. Ct. 2752 ("The purpose of inquiring into the existence of racially polarized voting is two-fold: to ascertain whether minority group members constitute a politically cohesive unit and to determine whether whites vote sufficiently as a bloc usually to defeat the minority's preferred candidates."). To determine whether racially polarized voting is "legally significant," the court must make "discrete inquiries into minority and white voting practices." Id. Dr. Kousser's conclusion that there is racially polarized voting in the area encompassed within SD 27 focuses exclusively on the relative percentage of Latino and white voters who chose the Latino candidate. It does not address whether the percentage of white (or non-Latino) voters who voted against that candidate was sufficient to defeat him or her. Consequently, to the extent Dr. Kousser concludes that there is "racially polarized" voting in the district, it is not the type of "legally significant" polarization about which Gingles speaks.

[35] Another important change since the time of Garza is a demographic one: the number of Latinos in Los Angeles County is growing rapidly. Over the course of the 1990's, the Latino population in Los Angeles County increased by 34%, in contrast to the county population as a whole, which grew by 11.5%. During that time, the white population declined both in percentage and absolute numbers. See Los Angeles County Demographic Profile, available at http://planning.co. la.ca.us/rsrch_LACountyProfile.pdf (last visited May 29, 2002). The only inference that can be drawn from the demographic data is that Latinos are, as a practical matter, a far more formidable political force than they were in the 1980's.

[36] As noted earlier, MALDEF is counsel to plaintiffs in this case. We do not consider pre-litigation positions taken by MALDEF to bind plaintiffs in this matter under an estoppel theory or otherwise. MALDEF was a leading participant in the redistricting period, however, and Dr. Kousser worked with it, applying the theories he now offers as an expert witness in this case. Accordingly, some of MALDEF's submissions to the legislature may be relevant evidence, not because they constitute plaintiffs' position in this litigation, but because they reflect the opinion of MALDEF's expert on the issues presented by this case.

[37] We assume for purposes of this opinion that the results in ballot measure elections may be used to show that Latino voters' desires are regularly rejected as a result of the cohesive vote of the non-Latino majority. We express no view, however, on the underlying question of the relevancy of such data. To the extent that plaintiffs may be suggesting that if an individual voted "yes" on the three propositions, he must necessarily be hostile to Latino interests and unlikely to vote for Latino candidates, we are unwilling to adopt such a generalization. It may be reasonable, however, to assume that if a voter voted "no" on the propositions, then he is probably not unreceptive to the candidacy of a Latino.

[38] Our conclusion is limited to the traditional vote dilution claim discussed in this Section. It is not intended to resolve the intentional vote dilution claim which we consider in Section III, infra.

[39] We note that the percentage of Latinos voting for Godinez is significantly lower than the percentage voting for other Latino candidates, which is regularly well above 90%. We have no doubt that this is so at least in part because Rep. Berman has a well-documented record of support for the interests of Latino communities and has historically championed Latino causes both as a state and a federal legislator. See, e.g., Aff. of Dolores C. Huerta, First Vice-President Emeritus of the United Farm Workers Union ("Congressman Howard Berman has always been one of our most effective political supporters.... [W]hen then-Assemblyman Berman authored and worked to pass the Agricultural Labor Relations Act, which provided farm workers with the right to collectively bargain and organize, there were very few farm workers or Latino voters in his district."). Congressman Berman has received MALDEF's Annual Legal Award for being a "leader in the fight to oppose the creation of a new foreign agricultural guest worker" program, and an Annual Congressional Award from the National Council of La Raza in recognition of his "support of public policies benefitting the Hispanic Community." Moreover, when the district contained the Latino voters whose removal plaintiffs challenge, Berman was so popular that in the last elections, in 2000, he had no opponent in the Democratic primary, and no major-party opponent in the general election.

[40] As defense counsel noted, paraphrasing Aristotle's well-known and much-quoted aphorism, "[O]ne swallow does not make a summer, nor does one day." 1 NICHOMACHEAN ETHICS 1098, in COMPLETE WORKS OF ARISTOTLE, trans. by W.D. Ross, ed. Jonathan Barnes, Princeton University Press (1984). See, e.g., United States v. Virginia, 518 U.S. 515, 561, 116 S. Ct. 2264, 135 L. Ed. 2d 735 (1996) (Rehnquist, C.J., concurring); Commissioner v. Wodehouse, 337 U.S. 369, 415, 69 S. Ct. 1120, 93 L. Ed. 1419 (1949) (Frankfurter, J., dissenting). See also Uno v. City of Holyoke, 72 F.3d 973, 985 (1st Cir.1995) ("the results of a single election are unlikely, without more, to prove the existence or nonexistence of embedded racial cleavages. Thus, race-conscious politics (or its absence, for that matter) can more readily be seen by producing a documentary that spans a series of elections than by taking an isolated snapshot of a single election. After all, to be legally significant, racially polarized voting in a specific community must be such that, over a period of years, whites vote sufficiently as a bloc to defeat minority candidates most of the time.").

[41] In addition to the elections discussed in the text, defendants also present evidence regarding the success of Latino candidates for statewide office within CD 28. Specifically, they cite the fact that Lieutenant Governor Cruz Bustamante won both the 1998 primary and general elections within the district, that unsuccessful Insurance Commissioner candidate Diane Martinez won the district in her 1998 primary race, and that Art Torres won the district in his 1994 general election race for Insurance Commissioner, after losing in the area in the primary. For the reasons stated earlier, we find such evidence only marginally relevant in assessing voter attitudes within CD 28. To the extent relevant, however, it demonstrates that Latino candidates are not routinely defeated in the precincts that comprise CD 28.

[42] Ninety-six and one-third percent of CD 28 is within the City of Los Angeles. The remainder of the district is the City of San Fernando, which is predominantly Latino; 87% of the VAP in San Fernando is Latino. Thus, the only reasonable inference based on plaintiffs' racial cohesion statistics is that a Latino candidate would fare better in the district as a whole than he did in the Los Angeles portions of the district. And, as noted in the text, Latino candidates win elections against white opponents even in the Los Angeles portions of the district.

[43] Delgadillo finished second in the primary election in the precincts comprising CD 28. The white incumbent City Council member received 41.79% of the votes cast, while Delgadillo received 37.58%. A third candidate, a white woman, received 16.32% of the vote. As noted in the text, however, Delgadillo defeated the City Council member in the run-off election where they were the only candidates on the ballot.

[44] Villaraigosa also won the primary election in the precincts within CD 28. He garnered the highest number of votes of the five candidates in the race, securing 36.51% of the vote within the district. The next closest finisher —a white candidate—received 21.66% of the vote.

[45] The essence of plaintiffs' complaint is that as a result of the redistricting legislation, a Latino is less likely to win a Democratic primary in CD 28; given the overwhelmingly Democratic character of the district, plaintiffs correctly represent that the only election that really matters to voters' ability to elect candidates of choice in CD 28 is the Democratic primary. Accordingly, it is important to note that the figures cited above most likely understate the extent to which white voters would cross over to support Latino candidates in a Democratic primary. This is so because white Democrats generally appear more willing to support Latino candidates than do white Republicans. The only reasonable inference to be drawn is that the percentage of white voters who supported Latino candidates in the nonpartisan elections discussed in the text—i.e., the elections for Mayor, City Attorney, and City Councilman—is lower than would be the cross-over percentage in a Democratic primary.

[46] According to United States Census Bureau statistics, of which we take judicial notice, San Marino has a population that is 48.1% white, 46% Asian, 0.2% black and 4.2% Latino. The population of La Canada Flintridge is 74.2% white, 19% Asian, 0.4% black, and 4.3% Latino. Similarly, the population of La Crescenta-Montrose is 69.8% white, 17.5% Asian, 0.5% black and 8.8% Latino. Temple City's population is 41% white, 38.3% Asian, 0.7% black, and 18.1% Latino, while Glendale's population is 56.4% white, 16% Asian, 1.2% black, and 17.8% Latino. These figures reflect substantially higher non-Latino populations than CD 28, which is 55.65% Latino, 31.41% white, 4.29% black, and 6.29% Asian.

[47] We note that Latinos in Los Angeles County have substantial representation in the California State Senate as well; plaintiffs do not dispute that there are five Latino State Senators whose districts are entirely or in part within Los Angeles County. These Latino Senators represent 37% of the county's population. Id. Of the twelve Senate Districts that are predominantly within Los Angeles County (SD 17, 20-30), one-third of the seats are currently held by Latinos (SD 20, 22, 24, 30).

[48] Indeed, the Latino community organizations that sought to intervene as defendants, see n. 4, supra, contend that Latino interests are better advanced by protecting the incumbency of Congressman Howard Berman, who they assert has strongly supported Latino causes, and by the creation of districts in which Latinos maintain influence through coalition-building with other ethnic and racial groups.

[49] Only for a two-year period between the decision in City of Mobile v. Bolden and the 1982 Voting Rights Act Amendments was a plaintiff required to prove intent in order to prosecute a claim of vote dilution. And, "[a]fter the Supreme Court decided Bolden, vote dilution litigation virtually shut down." ISSACHAROFF ET AL., id., at 441. Thus, the principal intentional vote dilution cases decided by the Supreme Court are Rogers and Bolden. Courts and commentators are divided as to whether White v. Regester required proof of intent for a vote dilution claim, or merely proof of effects. In amending § 2, however, Congress adopted the view that Regester required effects only, and we agree. 1982 U.S.C.C.A.N. at 205.

[50] Plaintiffs cite language from De Grandy in support of their contention that the dilutive effect that need be shown in an intentional discrimination case under the constitution and the statute is lower than that which need be shown in a traditional § 2 case. Specifically, they quote a statement found in the Court's discussion of proportionality, which rejects the state's argument that "the most blatant racial gerrymandering in half of a county's single-member districts would be irrelevant under § 2 if offset by political gerry-mandering in the other half, so long as proportionality is the bottom line." De Grandy, 512 U.S. at 1019, 114 S. Ct. 2647. We note that the Court's discussion concerns the "totality-of-the circumstances" test, and agree, as indicated in the text, that proof of intentional discrimination should be considered in this context. The quoted language does not convince us, however, that the required proof of dilutive effects should be relaxed to the full extent plaintiffs appear to suggest.

[51] We note that plaintiffs bring intentional vote dilution claims under both the Equal Protection Clause and § 2 of the Voting Rights Act. It is unclear whether there is a separate claim of intentional dilution under § 2 that differs from the traditional § 2 claim. Barnett v. Daley, 32 F.3d 1196, 1202 (7th Cir.1994) ("It is, indeed, no longer clear that intent plays any role in a suit under section 2 of the Voting Rights Act ..."). Nor is it clear that any such intentional § 2 claim, if it exists, would require any different proof than a constitutional vote dilution claim. We need not consider those questions, however, because regardless of the form of the intentional claim asserted here, plaintiffs have failed to offer proof that satisfies the minimal effects requirement described above.