State Defendants' Post-Trial Reply Brief

    THEE: ATTORNEY (CN EIR.WL. 

Or TEXAS 

JIM NIATTOX 

ATTORNEY GENERAL October 4 1989 
2 

VIA FEDERAL EXPRESS 

John D. Neil 

Deputy U. S. District Clerk 
200 E. Wall, Room 316 

Federal Building 
Midland, Texas 79701 

  

Re: LULAC Council #4434, et al. v. Mattox, et al., 

No. MO-88-CA-154 

Dear Mr. Neil: 

Enclosed for filing in the above-referenced matter are the 

original and one copy of the State Defendants’ Post-Trial Reply Brief. 

Sincerely, 

Cin rcs 
Renea Hicks 
Special Assistant Attorney General 

P. O. Box 12548, Capitol Station 

Austin, Texas 78711-2548 

(512) 463-2085 

CC: Counsel of record 

512/463-2100 SUPREME COURT BUILDING AUSTIN, TEXAS 78711-2548 

 



  

UNITED STATES DISTRICT COURT 
WESTERN DISTRICT OF TEXAS 
MIDLAND-ODESSA DIVISION 

LULAC COUNCIL #4434, et al., 

Plaintiffs, 

Civil Action No. 

MO-88-CA-154 

VS. 

JIM MATTOX, et al., 

Defendants. LO
N 

LO
R 

LO
R 

LO
R 

CO
N 

LO
R 

O
P
 

STATE DEFENDANTS' POST-TRIAL REPLY BRIEF 

The State Defendants! reply as follows to the post-trial briefs 

of the plaintiffs, the Harris County plaintiff-intervenors, and the 

Dallas County plaintiff-intervenors: 

1. 

PARTISAN BLOC VOTING 
  

The three plaintiff briefs offer at best weak rejoinders to the 

salience of partisan bloc voting in analyzing vote dilution cases under 

Section 2 of the Voting Rights Act. The LULAC plaintiffs dismiss 

Whitcomb v. Chavis in a footnote.2 The Dallas County plaintiff- 

intervenors suggest that the point begs the question. The Harris 

  

1 Paragraphs 12-14 on pp. 4-5 of the LULAC plaintiffs’ post-trial bricf 
accurately lists the State Defendants except for John F. Onion, who has been 
replaced by Michael J. McCormick as Presiding Judge of the Texas Court of 
Criminal Appeals and a member of the Judicial Districts Board. 

4 In footnote 26 on page 32 of their post-trial brief, the LULAC plaintiffs 
mischaracterize Whitcomb's status under the law. As pointed out in the 
extensive discussion in the State Defendants’ opening post-trial brief, Congress 
clearly intended to revivify Whitcomb when it amended Section 2 in 1982. At 
least five of the justices in Gingles still viewed it as good law, and Justice 
Brennan's plurality opinion did not reject it. 

 



County intervenors say that it would "immunize" partisan elections 

from Section 2 attack. All are wrong. 

Recognition of partisan bloc voting as the best description of 

the voting patterns would not immunize partisan elections from 

Section 2's reach. The State Defendants’ opening post-trial brief 

carefully explained this point and delineated how inquiries into 

partisan bloc voting and racial bloc voting can be meshed together to 

lead to precisely the kind of inquiry Gingles requires -- that is, a 

practical, realistic inquiry which takes a functional view of the local 

political processes under challenge. On the other hand, not taking 

partisanship into account (as the plaintiffs would have it) would 

create an immunity of sorts. Given the realities of present-day 

partisan voting patterns, ignoring it essentially immunizes plaintiffs 

from defeat in vote dilution challenges to at-large partisan elections. 

In 1982, Congress obviously did not intend to surreptitiously slip the 

per se invalidation of such elections into Section 2 and guarantee 

victory to plaintiffs.3 Gingles, in fact, holds that at-large elections are 

presumptively valid. 106 S.Ct. at 2764. 

That the applicable legal concepts in vote dilution cases are less 

simplistic and more realistic than the plaintiffs would have it is 

becoming increasingly clear as the courts grapple with Gingles’ 

meaning in a variety of electoral contexts. The Dallas County 

plaintiff-intervenors reluctantly acknowledge the force of this 

observation at page 17 of their brief when they refer to the more 

  

3 The State Defendants’ opening post-trial brief inadvertently 
characterized the President who signed the 1982 Voting Rights Act 
amendments as a Democrat. Of course, he was a Republican.  



  

refined inquiries conducted in recent Section 2 decisions out of the 

Fifth Circuit. One such decision especially points toward the 

importance of taking partisanship into account in these cases. In 

Monroe v. City of Woodville, 881 F.2d 1327 (5th Cir. 1989), the Fifth 

Circuit holds: "That a group's voting behavior is racially polarized 

indicates that the group prefers candidates of a particular race." 881 

F.2d at 1331. Here, the facts are that voters’ behavior is indifferent 

to the candidates’ race; instead, their voting behavior demonstrates 

that they prefer candidates of a particular party. Thus, in the 

counties involved in this suit, the evidence shows that there is not 

racially polarized voting, but partisan polarized voting. That this 

pattern does not violate Section 2 is the new lesson of the Fifth 

Circuit's Monroe decision and the old lesson of the Supreme Court's 

Whitcomb v. Chavis decision.4 

1. 

NON-DISCRIMINATORY POLICY SUPPORTING COUNTYWIDE 

ELECTIONS FOR DISTRICT JUDGESHIPS 
  

  

Chief Justice Phillips testified about the policy underlying 

countywide elections for district judges in Texas. Contrary to the 

representation on page 16 of the Harris County plaintiff-intervenors’ 

post-trial brief, he did not testify that his only concern was the 

system which might be substituted for it. Instead, he (along with Dr. 

Champagne) explained that historically in Texas counties are the 
  

4 That the lesson is the same, old or new, shows that Gingles rcally 
changed nothing in terms of defining of racially polarized voting. Accord 
Romero v. City of Pomona, ___ F.2d __, ___ (5th Cir. Aug. 24, 1989) ("[Gingles], 

moreover, did not alter the statistical methods used to prove racially polarized 
voting." : 

 



basic unit of government. He also testified generally about what 

might best be described as the inherent tension between the role of a 

trial judge and their being subject to popular election. Nothing 

prevents the state from doing what it can to relieve that tension 

without eliminating either of the elements that might create it. 

Countywide election of all the district judges sitting in a particular 

county embodies a policy which performs that function. It can 

hardly be deemed tenuous if the state's choice in the first place to 

elect its trial judges is to be respected. 

The LULAC plaintiffs’ response to such a policy appears to be 

that the county unit is irrelevant to any state concern because 

elected judges always are subject to the potential of undue influence. 

Abstractly, such an argument cannot be refuted, but it also has no 

meaning because it has no logical bounds. In fact, its logical 

conclusion is that there is no threat to judicial integrity -- and 

"integrity" is used in a broad sense -- no matter how small the unit is 

that elects a judge. Only empirical, comparative research can resolve 

the matter, and such research has not yet occurred in Texas because, 

other than for a very brief period many years ago in two Texas 

counties, smaller than countywide elections for district judges have 

not occurred. The state is not required, however, to conduct such 

empirical, comparative experiments before adopting a policy. The 

basis offered suffices, if deference to legislative policy choices means 

anything. 

Facts adduced by the plaintiffs in this case further buttress the 

policy of countywide elections. While Anglos as a group are 

proportionately better off from a socioeconomic perspective than 

oA  



  

blacks or Hispanics as a group, the raw numbers tell a different 

story. That story is that there is significant common ground in Texas 

between blacks, Hispanics, and large numbers of Anglos. In Tarrant 

County, for example, approximately 2,600 more white families than 

black families were below the poverty level in 1979. See Exhibit Ta- 

09, 4th page. In Ector and Midland Counties during the same period, 

there were more white families in poverty than black or Hispanic 

families. See Exhibits E-11 & M-13, 4th page. The further 

fragmentation of these three groups through the creation of smaller 

than countywide judicial elections would be likely to further reduce 

the common ground they share.>_ There is certainly nothing tenuous 

about a state policy that tends to increase the common bonds among 

different groups of its citizens. 

Chief Justice Phillips’ testimony also explained the non- 

discriminatory policy basis for the 1985 ensconcement in the state 

constitution of a preference -- and, as he carefully explained, it is 

nothing more than a preference, because the local voters can change 

it -- for countywide elections for state district judgeships. The 1985 

state constitutional amendments to the judiciary article of the Texas 

Constitution created an important new state institution with 

significant power to affect the configuration of the trial bench across 

Texas. Section 7a(i) of Article 5 was a limitation on the powers of 

  

3 This observation about the common ground among blacks, Hispanics, 
and whites of low socioeconomic status also is pertinent to Dr. Taebel's analysis. 
It buttresses his point that there is no such thing as a white bloc. Instead, 
there at least two white blocs in every county, one a Democratic bloc and onc a 
Republican bloc. Common political science wisdom is that whites who arc 
lower on the socioeconomic rung tend to vote Democratic. 

 



that new state institution -- the Judicial Districts Board -- to change 

longstanding local judicial arrangements. Senator Washington's later 

expressions of vague disquietude hardly suffice to convert this 

readily ascertainable policy into one of intentional, invidious 

discrimination against racial minorities in Texas. Federal courts 

should be "reluctant to attribute unconstitutional motives to the 

state, particularly where a plausible [constitutional] purpose may be 

discovered from the face of the statute." Mueller v. Allen, 463 U.S. 

388, 394-95 (1983). 

111. 

MISCELLANEOUS MISDIRECTIONS 
  

Before turning to replies appropriate to specific counties, other 

aspects of the post-trial briefs require some response: 

A. Hypothetical districts/hypothetical elections -- The plaintiffs 
  

repeatedly employ the device of comparing the actual outcomes of 

the analyzed elections with hypothesized outcomes of contests 

between the same candidates in hypothesized minority-voter 

dominated single-member districts. The problem with this approach 

is its attenuated relation to political reality. Even people with only a 

passing interest in American politics understand that changing the 

rules for elections brings a wholly different set of dynamics to bear 

on their outcome. The plaintiffs’ use of this concept of hypothetical 

elections suggests that proportional representation is the standard 

against which they ask the Court to measure the current Texas 

system. That standard is an invalid one, as the Dole proviso to the 

amended Section 2 establishes.  



Dr. Engstrom's testimony regarding Harris County presents 

another problem with the concept's use. According to page 3 of the 

post-trial brief of the Harris County plaintiff-intervenors, Dr. 

Engstrom testified that if the same election were held in the white 

community and in the black community, the winner of the election 

would be different. This is very similar to the hypothetical district 

device. The problem is that, even assuming that the political 

dynamics were otherwise the same, the question of whether the 

election outcome would be different in the white and black 

communities cannot be answered without specifying the particular 

white community to which reference is being made. If the same 

election were conducted in the white Democratic community, the 

electoral outcome would be precisely the same in it as in the black 

community in Harris County. 

The types of comparisons criticized here are invalid either 

because they are fundamentally misguided flights of fancy or 

because they are too simple to take into account the political reality 

that whites do not vote as a monolithic bloc. 

B. Broad-brushed pictures of racial discrimination -- The LULAC 
  

plaintiffs repeatedly invoke three sets of historical references to try 

to buttress their case against the State Defendants. None of them 

adds anything to their case, and all should be ignored. One device to 

which they regularly retreat is the recitation of what the Department 

of Justice has said in letters of objection under Section 5 of the 

Voting Rights Act. The first problem with the use of these letters is 

that they really are not the product of any adjudicatory factfinding 

process at all. Second, most of them are so dated that, even if the 

5  



facts were true and relevant in the past, they mean nothing about 

current practices. Third, and most critically, a careful reading of the 

overwhelming bulk of those so-called § 5 letters reveal that they 

prevented possible discriminatory action from occurring, instead of 

describing discriminatory action that had occurred. Thus, the letters 

rebuffed the very actions which the plaintiffs seek to invoke to help 

their case. They typically show that incriminatory things did not 

occur in a particular county, not that they did occur. 

Another device repeatedly used by the plaintiffs is the citation 

to federal court decisions finding violations of other federal 

constitutional provisions or statutes protecting individuals from 

racial discrimination in various aspects of governmental conduct. 

Again, most of these decisions are too dated to mean anything in this 

case. Many involve isolated conduct by individuals working for some 

political subdivision somewhere in the county in question. The are 

by and large wholly irrelevant to this case. Their principal use is as 

a makeweight for the plaintiffs’ arguments. 

Finally, the plaintiffs repeatedly cite the Graves v. Barnes 

findings with regard to the counties at issue. These citations add 

nothing other than an historical footnote to the analysis of this case. 

Fifteen years have passed since those findings were announced. The 

state has become a two-party state. The Democratic party, which 

really was the principal focus of Graves v. Barnes, bears no 

resemblance to the party depicted then. In fact, the LULAC plaintiffs 

themselves make this point when they write in § 76 on page 33 of 

their post-trial brief that "voters generally know that if they vote 

Democratic, rather than Republican, they will be voting for a 

2.  



  

minority candidate or for a candidate who is sympathetic to concerns 

of the minority community." This acknowledgment alone means that 

the Graves v. Barnes findings are irrelevant here and should be 

disregarded, except insofar as they contain certain historical 

references. 

The plaintiffs’ approach on these matters is exactly the one the 

Fifth Circuit has warned against: 

[T]he fluidity of the Section 2 analysis, as 
glossed by Thornburg, may encourage 
litigants to aggregate statistics or facts rather 
indiscriminately in the hope that a district 
court will simply count facts rather than 
analyze them carefully. 

Overton v. City of Austin, 871 F.2d 529, 533 (5th Cir. 1989). 

C Majority vote requirement in primaries -- The depth with 
  

which Dr. Brischetto studied the challenged election practices at issue 

here was revealed when he testified incorrectly that a majority vote 

requirement for electing state district judges in Texas enhanced the 

system's discriminatory effect on minority voters. Of course, there is 

no such requirement in the general elections, which require only a 

plurality for victory. In their brief, the LULAC plaintiffs adopt a 

fallback position which criticizes the discriminatory enhancement 

that follows from a majority vote requirement for party nomination. 

The problem with the plaintiffs’ discussion on this point is that it has 

virtually nothing to do with what they are challenging in this case. 

Their challenges, except for those in Travis and Jefferson Counties, 

are in reality to the system of general elections, not of primaries. 

The rather indiscriminate mixing of complaints about the primaries 

 



and general elections runs afoul of yet another warning from the 

Fifth Circuit which recently noted the criticism of "mixing data from 

primary and general elections.” Monroe, supra, 881 F.2d at 1331. 

D. Fixation with R%_ Coefficients -- Plaintiffs employ a three step 
  

process in determining whether racially polarized voting exists in a 

county. First, Dr. Brischetto runs regression analyses on selected 

races. Second, Dr. Brischetto looks at the correlation between the 

percentage of minorities in voting precincts and voting behavior. 

Third, if the correlation coefficient (the R2) exceeds a certain value, 

Dr. Brischetto concludes that racially polarized voting exists. While 

writing for another circuit, Judge John Minor Wisdom, a long-time 

member of the Fifth Circuit, has cautioned against such a simplistic 

methodology for determining whether racially polarized voting exists 

and encouraged the examination of factors "other than race that may 

also correlate highly with election outcomes . . . [such as] party 

identification." Lee County Branch of NAACP v. City of Opelika, 748 

F.2d 1473, 1482 (11th Cir. 1984) ("Opelika"). Citing Judge 

Higginbotham, Judge Wisdom explains why such a methodology 

cannot fully explain the critical factual inquiry of racially polarized 

voting. 

The point is that there will almost always be a raw 

correlation with race in any failing candidacy of a 

minority whose racial or ethnic group is [a] small 

percentage of the total voting population. 

Opelika, 748 F.2d at 1482 n.15 (brackets added by the court). Thus, 

a thorough analysis must include a determination whether the 

seemingly racially polarized voting pattern is best explained by race  



  

or whether it is best explained by party identification, for example. 

Opelika, 748 F.2d at 1482. We have examined this matter; the 

plaintiffs ignored it. 

  

IV. 

SPECIFIC COUNTY RESPONSES 

A. Harris County -- The post-trial brief of the Harris County 
  

plaintiff-intervenors incorrectly says at page 5 that Dr. Taebel, an 

expert of the State Defendants, analyzed some "white-on-white" 

clections in connection with Harris County. He conducted no such 

analysis, as a review of Exhibit D-5 will show. At page 8 of the same 

brief, the plaintiff-intervenors employ a transition which seems to 

ascribe opinions to Dr. Taebel that are the opposite of those to which 

he testified. Among other things, after noting that Dr. Taebel had 

described swing voters in the closely partisan-balanced county as 

largely white, the plaintiff-intervenors assert that the Republican 

vote, the swing vote, and the white Democratic vote clearly do not 

support black candidates running in district judge elections. In case 

it is not clear, Dr. Taebel did not so testify. On the contrary, Dr. 

Taebel testified that his analysis showed that the white Democratic 

voters supported Democratic candidates, irrespective of race, and 

that the swing voters showed no consistent pattern in the kind of 

candidate that they supported. It was that testimony that led Dr. 

Taebel to his conclusion that, after taking partisan bloc voting into 

account, whites in the county -- including the swing voters and the 

white Democratic voting bloc -- did not vote consistently as a bloc so 

as usually to defeat the preferred candidate of minority voters in 

district judge elections. 

i, 55 

 



B. Dallas County -- Although not strictly a reply point, the State 

Defendants do note for the Court that the Eighth Circuit recently held 

that unsuccessful minority candidates do not have standing to raise 

Section 2 claims. Roberts v. Wamser, 883 F.2d 617 (8th Cir. 1989). 

To the extent they bring their claim as defeated candidates, the 

Dallas County plaintiff-intervenors lack standing in this action. 

C Tarrant County -- The plaintiffs’ brief on this county really 
  

requires no response. 

D. Bexar County -- The plaintiffs’ characterization of Bexar County 
  

as an area where Hispanic candidates do not get elected because 

Anglo voters do not vote for Hispanic candidates is simply not 

reflective of the actual outcomes of elections in the county. 

Moreover, the plaintiffs’ assessment of politics in Bexar County does 

not recognize the political reality that the Hispanic community wields 

extensive political clout. 

Anglo voters frequently vote for and elect Hispanic candidates. 

Judge Anthony Ferro testified that the qualified Hispanic can win, 

and frequently does win, in Bexar County. 

We have five Hispanic district court judges here: 

Machado, Chavarria, Mireles, Gonzales, and Chapa. Chapa 

is now a court of appeals judge. We have five Hispanic 
county court-at-law judges: Tony Jimenez, myself, 

Anthony Ferro, Paul Canales, Miguel Rodriguez, and 

Robert Lozano. We had Henry Cisneros who, of course, 

won by 70 to 75 maybe 80 percent of the vote. We had 

Raymundo Villarreal who ran and won county wide for 

16 years as county surveyor. We have Rudy Garza, tax 

assessor collector. We have David Garcia who wins 

county wide as the district clerk; David Garcia defeated 

an Anglo opponent in the general election. We had 

Sheriff Joe Neaves. We had another man named Rudy 

12. 
der  



  

Garza as Sheriff who defeated an Anglo. We had Albert 
Bustamante as county judge who defeated two Anglo 

opponents. We have Chapa and Cadena on the Fourth 

Court of Appeals. Judge Cantu who was on the Fourth 
Court of Appeals defeated himself when right before the 
election he stated to the general public that he never 

wanted to be a lawyer or a judge. Fred Rodriguez who is 
the district attorney defeated the incumbent Sam 
Millsap in 1986. We have a lot of Hispanics who get 
elected in the community college district races, which 
are almost county wide. 

Hispanic candidates that campaign county-wide and do not 

limit themselves to targeting the Hispanic community win in Bexar 

County. They win both as Democrats and as Republicans. Judge Paul 

Canales beat Michael Patterson (an Anglo Republican) in the General 

Election. Judge Canales not only won in the Hispanic West and South 

Sides, he also won many of the Anglo boxes in the near north side 

(inside Loop 410). Judge Canales also ran well (as well as Anglo 

Democrats) in the Republican Anglo boxes outside of Loop 410. Roy 

Barrera Jr. running as a Republican Hispanic defeated, by a large 

margin, Mr. Stolhandski (an Anglo Democrat) in the 1982 general 

election. Mr. Barrera had a broad base campaign. He did not target 

any area of town. He campaigned over the whole county. In fact, 

Judge Anthony Ferro testified that he did not think that Barrera lost 

a single Anglo North side box. 

In general the Hispanic candidates that lose (like Adam 

Serrata) in Bexar County, lose because they target the Hispanic 

community to the exclusion of the Anglo Community. Judge Anthony 

Ferro calls this failed campaign strategy a 1960's campaign; Judge 

Ferro concludes that the Hispanics that lose: 

13. 

 



[h]ave been giving up the north side and northeast side, 

and campaigning strongly in the south side and the west 
side figuring that they can win with that vote alone. 
They figure that since there are more Mexicans than 
Anglos they can win the election simply by campaigning 

in that area. They do not take into consideration that not 
all of them come out and vote. And if they do come out 

and vote that they do not get down to the bottom of the 
ballot where the judges are. If 1,000 people vote for the 
governor in a west side precinct, 400 will vote for the 

county officials, especially the judges. 

The current political reality of the eighties (not the sixties) in 

Bexar County that the Plaintiffs ignore, and the Judge Ferro 

acknowledges, is that "it is coming to the point to where I do not 

know whether an Anglo is going to be able to win a county wide race 

because of the demographics.” 

Plaintiffs focus on the 1988 Republican Primary race where 

Robert Arellano lost to Stephen White to show that Hispanics lose 

because of their race. Judge Paul Canales testified that he had 

spoken with Arellano and that Arellano believes he lost because he 

did not work hard enough, he took things for granted, and because of 

the name identification problem with former DA Bill White and 

former Governor Mark White. Judge Ferro believes that Mr. Arellano 

lost because he campaigned almost exclusively in the Hispanic 

community: "That is a very bad tactic in this county to campaign in 

just a certain area of town, because that is all you are going to get." 

The political reality in Bexar County is that Hispanics who 

campaign county-wide and who seek the votes of all voters win. 

They win at all levels of government. Henry Cisneros, San Antonio's 

former Mayor, frequently received 70 to 80 percent of the vote. 

Rudy Garza, county tax assessor and collector, beat an Anglo 

14.  



  

opponent by campaigning throughout the county -- indeed, he was 

the highest vote getter in the county for that election. 

Another political reality in Bexar County is that elected 

Hispanics who are perceived to do well in office do not draw 

opposition. Judge Ferro has won reelection for county court at law 

twice, both times unopposed. According to Judge Ferro, in Bexar 

County "if the lawyers perceive you to be a good judge, and fair, you 

are not going to get opposition." Judge Canales testified that the 

Mexican American judges in Bexar county run unopposed because 

people perceive them as strong and people do not want to risk losing 

to them. Lack of opposition is not a weakness, as the plaintiffs 

suggest, but a sign of the political strength of the Hispanic 

community -- a community that now dominates politics in Bexar 

County. 

E Travis County -- The third sentence of {| 236 on page 89 of the 
  

LULAC plaintiffs’ post-trial brief is inaccurate. On cross-examination, 

Mr. Coronado acknowledged an awareness that, in addition to the 

1988 Democratic primary race between Gallardo and McCown, there 

had been other contested races for the Democratic party nomination 

to run for district judge in this decade in the county. He recalled the 

race between Brock Jones and Joe Hart (who now is an incumbent) 

and found no reason to quarrel with the assertion that there also had 

been a Democratic party contest between Jon Wisser and Joe Dibrell 

(both of whom now are district judges in Travis County). As several 

other State Defendants’ exhibits show, there have been numerous 

contested Democratic primary races in Travis County over the years, 

with Hispanic candidates showing a consistent ability to garner large 

x5. 

 



Anglo support and, even more importantly, achieve victory in 

electoral jurisdictions dominated by Anglo voters. 

The plaintiffs analysis falls far short of the kind of searching 

local inquiry required by Gingles. In fact, it falls even further short 

than one might gather from a first glance at what they did. In § 42 

on page 18 of their brief, the plaintiffs explain that they only 

reviewed elections where a minority candidate opposed an Anglo, 

but in Travis County they violated their own injunction by analyzing 

the three-way county-court-at-law race among an Hispanic (Castro), 

a black (Kennedy), and an Anglo (Hughes).® Throwing out that 

election, as the plaintiffs argue is really the only appropriate way to 

do, leaves them with having analyzed only two judicial races, both in 

1988. By their own terms, this level of analysis is insufficient to 

permit conclusions to be drawn about whether vote dilution occurs in 

the Democratic primary in Travis County. Mr. Richards’ testimony, 

exactly the kind of intensely local inquiry called for by Gingles, on 

the other hand show a consistent pattern of Anglo support for 

Hispanic candidates and a consistent pattern of victory for the 

preferred candidates of Hispanic voters. 

There are other obvious flaws in the plaintiffs’ analysis. Their 

Exhibit Tr-O1 shows a total voting age population for the county of 

312,392, yet State Defendants’ Exhibit D-15 shows that in 1988 

registered voters alone in the county equalled 319,059, more than 

  

6 Kennedy, the black candidate in this countywide judicial race in the 
1988 Democratic primary, received 52% of the Anglo vote. See Exhibit D-8, p. 
44a. Castro, the Hispanic candidate, received approximately equal levels of 
black and Anglo support. See Exhibit D-§, p. 41. Together, the two minority 
candidates received 70% of the Anglo vote.  



  

than the plaintiffs’ total for people even old enough to vote. The 

reliance in their brief on relatively ancient Graves v. Barnes 

statements about concentrations of Hispanic residents has been 

superseded by the passage of time. For example, Judge Nowlin noted 

the dispersal of Hispanic residents in Austin on page 36 of his 1985 

opinion in Overton v. City of Austin. See Exhibit D-40 ("Majority 

Hispanic census tracts account for only 30 percent of Austin's 

Hispanic population; thus, 70 percent of Hispanics live elsewhere in 

Austin.")7 

The plaintiffs’ post-trial brief makes it clear, if it was not 

before, that their case comes down to one race, the Gallardo/McCown 

race. The Overton decisions at the trial and appellate levels, David 

Richards’ testimony, and the very way in which Gallardo conducted 

his campaign (stressing to all segments of the Travis County voting 

population that he was the only minority candidate) conclusively 

demonstrate that the outcome of that race was, in some sense, an 

aberration in Travis County politics. To permit it to be the sword 

that strikes down a well-functioning system that is solicitous of 

minority voters’ concerns would be to let the exceptional electoral 

contest be treated as a typical Travis County electoral contest. 

Section 2 neither requires nor permits such an approach to the facts. 

F. Jefferson County -- The plaintiffs rely heavily on the testimony 

of Mr. Davis to support their argument that black candidates cannot 

run at large for judgeships and win in the county. Mr. Davis's 

  

7 This circumstance explains why the plaintiffs were unable to draw a 
district mecting the first Gingles factor of geographical compactness. 

17 

 



testimony offers nothing other than conclusory, self-serving 

statements on this point. Other details of his testimony, however, 

refute his conclusory allegations. First, he twice ran for a judgeship 

himself, obviously operating under some hope that a black judicial 

candidate could run and win countywide. In fact, in one of his two 

races, a black judicial candidate did run and win countywide. It just 

happened not to be Mr. Davis. That same candidate, Judge Floyd, 

since has run again, this time unopposed, and still sits as a county 

court at law judge in the county. Second, Mr. Davis testified that 

three black lawyers had sought gubernatorial appointments to 

district judgeships when they came open in the county. Such 

appointments are only interim, and the appointees would of 

necessity have had to run countywide to hold their judgeships. 

Obviously, those black office-seekers were of the view that a black 

judicial candidate could run and win countywide in Jefferson County. 

Otherwise, obtaining the sought-after appointments would have been 

a fairly hollow success for them. 

The plaintiffs analyzed only two judicial races in the county in 

this decade, both for justice of the peace. In the more recent of the 

two races, the plaintiffs’ own expert found no racially polarized 

voting. And in that race, the candidate supported by Mr. Davis was 

the white opponent of the black candidate. 

Only by surmise can a vote dilution case be discerned in 

Jefferson County. The plaintiffs have offered no evidence sufficient 

to find that whites votes sufficiently as a bloc in the Democratic 

primary in the county to usually deny victory to the preferred 

candidate of minority voters. The evidence, in fact, points the other 

18.  



  

direction -- that is, that the preferred candidate of black voters (e.g., 

Representative Price) can and does win. There is no support in the 

law for the plaintiffs’ effort to avoid this result by painting 

Representative Price as a candidate of special circumstance merely 

because of his employment record. There is no evidence that this 

circumstance had any special effect on Jefferson County voters. 

G Lubbock County -- In their post-trial brief, the State 
  

Defendants explained how the plaintiffs have failed to prove political 

cohesion between black and Hispanic voters in the county. In § 294 

on page 109 of their brief, the plaintiffs show the limited perspective 

from which they view the issue of political cohesiveness. They rely 

nearly exclusively on the regression analysis that, as Monroe, supra, 

now makes clear, is addressed to racially polarized voting patterns, 

not political cohesiveness. See also Romero v. City of Pomona, ___ 

F.2d ___ (9th Cir. Aug. 24, 1989). The testimony of Mr. Shelton and 

Ms. Mercado provide the sources for a more localized inquiry which 

shows that blacks and Hispanics in Lubbock County cannot be 

considered politically cohesive to an extent which will permit them 

to be aggregated for purposes of meeting the first Gingles factor. 

The plaintiffs’ post-trial brief in § 310 on page 114 relies on 

Ms. Mercado's deposition testimony to support an argument that 

minorities have not run countywide for judicial positions in Lubbock 

County because they are convinced they could not win. As with Mr. 

Davis's testimony concerning Jefferson County, the details of Ms. 

Mercado's testimony refutes her broad, conclusory generalizations. 

She names at least three Hispanic lawyers in the county who have 

sought the position through the appointive process. One may only 

-19- 

 



  

ask rhetorically why they would undertake such an effort if they 

truly were convinced that they could not run and win in the 

countywide race which would follow close on the heels of their 

sought-after appointment. 

H Ector and Midland Counties -- In Exhibits E-12 and M-14, the 
  

plaintiffs purport to show the exceedingly low rate of registration of 

Hispanic voters in Ector and Midland Counties. When combined with 

other items of evidence, this low registration level shows that the 

plaintiffs have failed to meet the first Gingles factor in these two 

counties -- even if, as the State Defendants dispute in their opening 

post-trial brief, blacks and Hispanics are politically cohesive enough 

to be formed into a single district. The other items of evidence are 

Exhibits E-13 and M-15. These are the plaintiffs’ efforts to factor out 

of their "Gingles I" calculations those Hispanics who are not eligible 

voters because they are not United States citizens. Those exhibits 

lower the margin of safety in terms of creating a "safe" minority 

district to a level closer to fifty percent of the voting age population. 

When considered with this factor, the low registration rates of 

Hispanic voters effectively turn the posited Gingles [ district into 

something less than a "safe" minority district. The plaintiffs 

therefore failed to meet their burden on one of the three threshold 

Gingles issues. 

  

8 Cases such as the Ninth Circuit's City of Pomona decision, when 
combined with Brewer v. Ham, 876 F.2d 449 (5th Cir. 1989), require the Court to 
begin its inquiry into the first Gingles factor by looking at the number of 

eligible minority voters, which excludes non-citizens. 

20 - 

 



CONCLUSION 
  

Incantation of the phrase "racially polarized voting" is 

insufficient for the plaintiffs in this case. It ignores too many of the 

political realities in the counties involved in this lawsuit, chief among 

them being the overwhelming domination of the outcome of state 

district judge races by partisan bloc voting patterns. The plaintiffs’ 

claims should be rejected, and judgment entered for the State 

Defendants. 

Respectfully submitted, 

JIM MATTOX 
Attorney General of Texas 

MARY F. KELLER 
First Assistant Attorney General 

=) “Heal 
2 WE Vn Tot 
  

RENEA HICKS 
Special Assistant Attorney General 

JAMES C. TODD 

Division Chief, General Litigation 

JAVIER P. GUAJARDO 
Assistant Attorney General 

P. O. Box 12548, Capitol Station 

Austin, Texas 78711-2548 

(512) 463-2085 

ATTORNEYS FOR STATE DEFENDANTS 

CERTIFICATE OF SERVICE 

I certify that on this 5th day of October, 1989, I sent a copy of 

the foregoing document by first class United States mail, postage 

oie § I  



  

prepaid, to each of the following: William L. Garrett, Garrett, 
Thompson & Chang, 8300 Douglas, Suite 800, Dallas, Texas 75225; 
Rolando Rios, Southwest Voter Registration & Education Project, 201 
N. St. Mary's, Suite 521, San Antonio, Texas 78205; Sherrilyn A. Ifill, 

NAACP Legal Defense and Educational Fund, Inc., 99 Hudson Street, 

16th Floor, New York, New York 10013; Gabrielle K. McDonald, 301 

Congress Avenue, Suite 2050, Austin, Texas 78701; Edward B. 
Cloutman, III, Mullinax, Wells, Baab & Cloutman, P.C., 3301 Elm 

Street, Dallas, Texas 75226-1637; J. Eugene Clements, Porter & 

Clements, 700 Louisiana, Suite 3500, Houston, Texas 77002-2730; 

and Robert H. Mow, Jr., Hughes & Luce, 2800 Momentum Place, 1717 

Main Street, Dallas, Texas 75201. 

Cre Het.   

Renea Hicks