NO IN THE UNITED STATES COURT OF APPEALS FOR THE FIFTH CIRCUIT

Transcription

1 NO IN THE UNITED STATES COURT OF APPEALS FOR THE FIFTH CIRCUIT TEXAS DEMOCRATIC PARTY AND BOYD L. RICHIE, In his capacity as Chairman of the Texas Democratic Party, Plaintiffs-Appellants, v. ROGER WILLIAMS, in his capacity as Secretary of State for the State of Texas, Defendant-Appellee. On Appeal from the United States District Court for the Western District of Texas APPELLANTS BRIEF Chad W. Dunn Attorney-In-Charge Randall Buck Wood General Counsel, Texas Democratic Party RAY WOOD & BONILLA K. Scott Brazil 2700 Bee Caves Road BRAZIL & DUNN Austin, Texas FM 1960 West, Suite 550 Telephone: (512) Houston, Texas Telephone: (281) Jared G. LeBlanc BUCK, KEENAN, GAGE, LITTLE & LINDLEY, L.L.P. 700 Louisiana, Ste Houston, Texas Telephone: (713) ATTORNEYS FOR PLAINTIFFS-APPELLANTS

2 NO IN THE UNITED STATES COURT OF APPEALS FOR THE FIFTH CIRCUIT TEXAS DEMOCRATIC PARTY AND BOYD L. RICHIE, In his capacity as Chairman of the Texas Democratic Party, v. Plaintiffs-Appellants, ROGER WILLIAMS, in his capacity as Secretary of State for the State of Texas, Defendant-Appellee. On Appeal from the United States District Court for the Western District of Texas CERTIFICATE OF INTERESTED PARTIES The undersigned counsel of record certifies that the following listed persons and entities, as described in the fourth sentence of Fifth Circuit Rule , have an interest in the outcome of this case. These representations are made in order that the Judges of this Court may evaluate possible disqualification or recusal. 1. Texas Democratic Party, Plaintiff-Appellant. 2. Boyd Richie, Plaintiff-Appellant, in his capacity as Chairman of the Texas Democratic Party. 2

3 2. Boyd Richie, Plaintiff-Appellant, in his capacity as Chairman of the Texas Democratic Party. 3. Phil Wilson, in his capacity as Secretary of State for the State of Texas, substituted for Roger Williams, in his capacity as Secretary of State for the State of Texas, Defendant-Appellee. 4. Hart InterCivic, Inc., as the manufacturer of the voting machines at issue in this lawsuit. Respectfully submitted, By: /s/ Jared G. LeBlanc Chad W. Dunn Attorney-In-Charge General Counsel, Texas Democratic Party K. Scott Brazil BRAZIL & DUNN 4201 FM 1960 West, Suite 550 Houston, Texas Telephone: (281) Randall Buck Wood RAY WOOD & BONILLA 2700 Bee Caves Road Austin, Texas Telephone: (512) Jared G. LeBlanc BUCK, KEENAN, GAGE, LITTLE & LINDLEY, L.L.P. 700 Louisiana, Ste Houston, Texas Telephone: (713) ATTORNEYS FOR PLAINTIFFS-APPELLANT 3

4 STATEMENT CONCERNING ORAL ARGUMENT Because of the important constitutional questions and the complexity of the eslate interface, Plaintiffs-Appellants request oral argument. Specifically, the issues in this case concern constitutional election jurisprudence as applied to recently developed electronic voting platforms. The impact of this case will reach thousands of citizens when they exercise their fundamental right to vote, and given the gravity of those implications, oral argument is warranted. 4

5 TABLE OF CONTENTS CERTIFICATE OF INTERESTED PARTIES..2 STATEMENT CONCERNING ORAL ARGUMENT...4 TABLE OF AUTHORITIES...8 JURISDICTIONAL STATEMENT...13 STATEMENT OF ISSUES...14 STATEMENT OF THE CASE...17 I. Introductory Statement...17 II. Procedural History of the Case...17 III. This Case is Not About a Constitutional Right to Emphasis Vote...19 IV. Background on Hart InterCivic s eslate...20 STATEMENT OF FACTS...21 I. The eslate s User Interface & The Emphasis Voter...22 A. The Warning Screen is Ineffective and Ambiguous...23 B. The Ballot Summary screen is patently misleading...25 II. TDP Responds to the Secretary s Motion with Competent Evidence.26 A. The Affidavits of Madison County Voters Establish that Emphasis Voting Occurs on eslate...27 C. The Affidavit of Robert Parten Establishes the Correlation Between Emphasis Voting and eslate...28 C. The Affidavit of William Brannon Corroborates the Existence of Emphasis Voters...30 D. The General Election Data from Nueces County Shows a Disparity Between eslate and Other Systems

6 E. The Affidavit of Daniel J. Francis Confirms that eslate has a Statistically Significant Higher Undervote Rate Than Paper Ballots...32 SUMMARY OF THE ARGUMENT...34 ARGUMENT & AUTHORITY...36 I. The District Court Erred When it Excluded the Affidavit of Robert Parten Under Federal Rule of Evidence II. The District Court Erred When it Dismissed TDP s Claims Arising Under the Equal Protection Clause of the Fourteenth Amendment...40 A. The Standard of Review...42 B. The District Court Erred When It Used the Anderson-Burdick Test Because This is Not a Ballot Access Case This right to vote case is unlike those ballot access cases that apply a rational basis inquiry The district court s reliance on Wexler & Weber is misplaced...48 C. The Reynolds line of right to vote cases require strict scrutiny...52 D. Bush v. Gore Requires Strict Scrutiny of the Secretary s Certification The Secretary s certification of eslate s confusing interface is arbitrary The Secretary s certification causes disparate treatment of similarly situated voters because eslate treats absentee/optical scan voters differently. 59 D. Even If Bush v. Gore Does Not Apply, TDP Raised Genuine Issues of Material Fact Under the Anderson-Burdick Rational Basis Test The impact of eslate on voters constitutional rights is severe There is no specific state interest that eslate s flawed user interface serves

7 III. IV. TDP s Evidence Created Genuine Issues of Material Fact on Its Due Process Claims TDP Submitted Competent Summary Judgment Evidence Creating Genuine Issues of Material Fact on TDP s Claims Under The Texas Election Code V. TDP raised competent summary judgment evidence that created genuine issues of material fact on TDP s Help America Vote Act Claims...74 CONCLUSION...76 CERTIFICATE OF SERVICE...78 CERTIFICATE OF COMPLIANCE

8 TABLE of AUTHORITIES Cases Anderson v. Calabrese, 460 U.S. 780, 784 (1982)... passim Boyd v. State Farm Ins. Cos., 158 F.3d 326, 331 (5th Cir.1998)...32 Burdick v. Takushi, 504 U.S. 428, (1992)... passim Bush v. Gore, 531 U.S. 98 (2000)... 47, 48, 49, 64 Carrington v. Rash, 380 U.S. 89 (1965)...50 Cipriano v. City of Houma, 395 U.S. 701 (1969)...50 City of Phoenix v. Kolodziejski, 399 U.S. 204 (1970)...50 Clements v. Fashing, 457 U.S. 957, 964 (1982)...38 Douglas v. California, 372 U.S. 353 (1963)...46 Duncan v. Poythress, 8

9 657 F.2d 691, 702 (5th Cir. 1981)...60 Dunn v. Blumstein, 405 U.S. 330 (1972)... 50, 59 Evans v. Cornman, 398 U.S. 419 (1970)...50 Gomillion v. Lightfoot, 364 U.S. 339 (1960)...38 Hainze v. Richards, 207 F.3d 795, 798 (5th Cir. 2000)...37 Hendon v. North Carolina State Bd. of Elections, 710 F.2d 177, 182 (4th Cir. 1983)...61 Herndon v. North Carolina Bd. of Elections, 633 F. Supp 454, (W.D.N.C. 1986)...45 Herndon v. North Carolina State Bd. of Elections, 710 F.2d 177 (4th Cir. 1983)... 43, 44 J & B Entertainment, Inc. v. City of Jackson, Miss., 152 F.3d 362, (5th Cir. 1998)...47 James v. Sadler, 909 F.2d 834, 837 (5th Cir.1990)...37 Kramer v. Union Free School District No. 15, 395 U.S. 621 (1969)...50 Logan v. Zimmerman Brush Co., 9

10 455 U.S. 422, 432 (1982)...61 Lubin v. Panish, 415 U.S. 709 (1974)...38 McDonald v. Board of Election Com rs of Chicago, 394 U.S. 802, 807 (1969)...46 McLaughlin v. Florida, 379 U.S. 184 (1964)...46 Norman v. Reed, 502 U.S. 279 (1992)...53 Northeast Ohio Coalition for Homeless v. Blackwell, 467 F.3d 999 (6th Cir. 2006)...61 O'Brien v. Skinner, 414 U.S. 524 (1974)...51 Pennhurst State School & Hosp. v. Halderman, 465 U.S. 89 (1984)...64 Puerto Rican Organization For Political Action v. Kusper, 490 F.2d 575 (7th Cir. 1973)...53 Reynolds v. Sims, 377 U.S. 533 (1964)... passim Roman v. Sincock, 377 U.S. 695 (1964)...50 Shelton v. Tucker, 10

11 364 U.S. 479, 488 (1960)...59 St. Romain v. Industrial Fabrication and Repair Service, Inc., 203 F.3d 376, 381 (5th Cir. 2000)...34 Stewart v. Blackwell, 444 F.3d 843, (6th Cir. 2006)...39 Strayhorn v. Williams, 430 F. Supp. 2d 661, (W.D. Tex. 2006)...39 Texas Democratic Party v. Benkiser, 459 F.3d 582 (5th Cir. 2006)...34 Texas Indep. Party v. Kirk, 84 F.3d 178, 182 (5th Cir. 1996)...39 Torres v. Sachs, 381 F. Supp. 309, 31213(S.D.N.Y.1974)...53 United States v. Classic, 313 U.S. 299 (1941)...38 United States v. Louisiana, 265 F. Supp. 703, (E.D. La. 1966)...53 Viterbo v. Dow Chemical Co., 826 F.2d 420, 422 (5th Cir.1987)...34 Weber v. Shelly, 347 F.3d 1101, (9th Cir. 2003)... 41, 42, 43, 62 Wexler v. Anderson, 11

12 452 F.3d 1226, (11th Cir. 2006)... 41, 42 Williams v. Rhodes, 393 U.S. 23, (1968)...36 WMCA, Inc. v. Lomenzo, 377 U.S. 633, 653 (1964)...50 Word of Faith World Outreach Ctr. Church, Inc. v. Morales, 986 F.2d 962, 966 (5th Cir. 1993)... 64, 65 Statutes 28 U.S.C U.S.C U.S.C (a) (6)... 65, U.S.C TEX. ELEC. CODE ANN (Vernon 2003)...52 TEX. ELEC. CODE ANN (Vernon 2003 & Supp. 2007)...63 Rules 5th Cir. R. 43(c)(2)...14 FED. R. APP. P FED. R. CIV. P Constitutional Provisions U.S. Const. Amend

13 JURISDICTIONAL STATEMENT The district court s subject matter jurisdiction in this case rests on 28 U.S.C because this is a civil action alleging that the challenged provisions of the Texas Election Code violate the U.S. Constitution. This Court has jurisdiction under 28 U.S.C to review the district court s final judgment, issued August 16, 2007, ordering that Plaintiffs-Appellants take nothing. (R. at 819.) Plaintiffs-Appellants, Texas Democratic Party and Boyd L. Richie, now appeal as of right of that order and the judgment. FED. R. APP. P. 4. Plaintiffs-Appellants timely filed their Notice of Appeal on August 23, (R. at 821.) See FED. R. CIV. P. 4(a). 13

14 STATEMENT OF ISSUES I. In response to Defendants Motion for Summary Judgment, Plaintiffs- Appellants submitted the Affidavit of Robert Parten, who has been the Election Administrator of Tarrant County, Texas for 20 years. The Parten Affidavit was based on his review of voting data from Nueces and Madison Counties from the 2006 General Election, and in which Parten opined that the significant undervotes in those counties was caused by the manner in which eslate handles emphasis voting. 1 Did the district court abuse its discretion when it ruled that the Parten Affidavit was inadmissible? II. In response to Defendants Motion for Summary Judgment, Plaintiffs- Appellants submitted evidence that the eslate voting platform certified by the Texas Secretary of State makes it less likely that an emphasis voter using eslate will cast an accurate and effective ballot than voters using a different platform and that the eslate ballot will be counted differently in a manual recount. Did the district court err when it found under a rational basis analysis that the eslate system does not violate the Due Process and Equal Protection Clauses of the Fourteenth Amendment? 1 An emphasis vote occurs when a voter selects a straight-party ballot, but then selects the party s candidates again in each individual race. 14

15 III. Did the district court err when it found under a rational basis analysis that the Secretary of State s certification of the eslate system does not violate the Due Process Clause of the Fourteenth Amendment? IV. In response to Summary Judgment, Plaintiffs-Appellants alleged that the eslate machine certified by the Texas Secretary of State violates the Texas Election Code because it does not record a vote even though the intent of the emphasis voter is reasonably clear regarding emphasis votes. Did the district court err when it granted Defendants Motion for Summary Judgment on Plaintiffs-Appellants claims arising under the Texas Election Code? V. In response to Summary Judgment, Plaintiffs-Appellants submitted evidence that the eslate machine records votes differently than other balloting devices used in the State of Texas, thereby creating disparity within the state for what does and does not constitute a vote. Did the district court err when it granted Defendants Motion for Summary Judgment on Appellants claims arising under the Help America Vote Act? 15

16 I. Introductory Statement STATEMENT OF THE CASE This case involves the Texas Secretary of State s certification of an e-voting machine that does not record a significant percentage of the votes cast in the approximately 100 Texas counties that use them. If the machines, which are made by Hart InterCivic, Inc. ( Hart InterCivic ), are not repaired by the next general election less than a year away thousands of voters will be disenfranchised by eslate. This is particularly true in the case of voters who grew up on the paper ballot system. II. Procedural History of the Case In February 2007, the Texas Democratic Party and Boyd Richie, as chairman of the Texas Democratic Party (collectively TDP ) filed this lawsuit against Roger Williams, in his capacity as the Texas Secretary of State (the Secretary ), 2 in the Austin Division of the Western District of Texas, and the case was assigned to Judge Sam Sparks. (R. at 8.) Seeking declaratory and injunctive relief, TDP alleged that the misleading user interface of the eslate voting machines used by several counties in Texas violated the First and Fourteenth Amendments, the Texas Election Code, and 2 Roger Williams is no longer in office, and he has been replaced by Phil Wilson, in his capacity of Secretary of State for the State of Texas. See 5th Cir. R. 43(c)(2). 16

17 the Help America Vote Act. (R. at ) The specific defects in eslate are that it records a blank ballot even though the summary screen indicates a straight-ticket ballot, it treats votes differently than paper ballots filled out identically, and it does not record a vote even when the intent of the voter is reasonably clear. (R. at 12, 18 & 19.) On March 5, 2007, the Secretary filed his answer and then a motion for summary judgment the following day. (R. at 4.) TDP filed a jury demand shortly thereafter. (R. at ) To engage in the discovery necessary to defend their claims against the Secretary s motion for summary judgment, TDP sought to expand their deadline to respond, the district court granted the request, and it ordered TDP to file its response by August 1, (R. at ) During the Secretary s designated representative s deposition, TDP discovered that the Secretary had done no study or collected any data concerning emphasis voters and the eslate platform. (R. at 377.) It thus became necessary for TDP to conduct third-party discovery from counties in Texas that used the device. (R. at ) TDP subpoenaed voting data from several counties, which were subsequently informed by the Secretary that responding to the subpoena could be illegal. (R. at ) The counties either failed to respond or filed motions to quash the subpoenas. (R. at & 807.) Still needing the voting records, TDP filed a motion to compel, which the district court declined to address before ruling on the summary judgment motion. 17

18 On August 10, 2007, the district court held a hearing on Defendant s Motion for Summary Judgment. (R. at 807.) At the hearing, the district court indicated that it was primarily interested in the constitutional right involved, in whether that right should be evaluated under a rational basis or strict scrutiny analysis. (TR. at 5:16 21.) At that hearing, the Secretary contended that TDP s claims should be evaluated under the rational basis standard. (TR. at 9:18 10:2.) He contented that a rational basis analysis was required under Anderson-Burdick, whereas TDP contended that the Reynolds line of cases notably employed by the Supreme Court in Bush v. Gore required strict scrutiny because this is a right to vote case; it is not a ballot access case. (TR. at 9:18 10:2.) On August 16, 2007, the district court entered an order granting the Secretary s motion, and it dismissed Plaintiffs Motion to Compel as moot. (R. at 819.) That same day, the district court entered a judgment that TDP take nothing, after which it timely filed a notice of appeal. (R. at ). TDP asks this Court to reverse the judgment and remand this case for discovery and a trial on the merits. III. This Case is Not About a Constitutional Right to Emphasis Vote Throughout this case, Defendant-Appellee, the Texas Secretary of State (the Secretary ) has attempted to obfuscate the real issue in this case by casting this action as a time, place, and manner claim of ballot access case. To be clear, the 18

19 interest TDP asserts is not the right to emphasis vote, but the right to have the intent of the voter recorded accurately and uniformly throughout Texas. The real question is whether federal and state law requires the voter to conform to the voting system or whether the voting system should conform to the known actions of voters. Plaintiffs- Appellants claim that it is the latter. The Secretary will likely argue that Texas has an interest in giving counties choices on which systems they use. (See R. at 758.) The State=s interest in providing a variety of voting systems that uniformly record the intent of the voters is not served if the intent of voters who choose to emphasis vote on the eslate is systematically not recorded, while the intent of voters who emphasis vote on other systems, such as paper ballot, is recorded. (See R. at 758.) Moreover, the Secretary cannot hide behind alleged uniform standards when the evidence shows these standards are insufficient to assure that voter intent is recorded. Indeed, a Astandard@ that does not insure a standard result is no standard at all. IV. Background on Hart InterCivic s eslate Hart InterCivic s eslate disenfranchises voters. Based upon the data TDP presented in response to Defendant s Motion for Summary Judgment, an average 7% of voters lose their votes, and in some precincts, the rate is as high as 24%. (R. at & ) Because the eslate is in use in more than 100 counties throughout 19

20 Texas, including many of the most populous counties, including the county seats of Houston, Fort Worth, Austin, Corpus Christi, and El Paso, there is a significant volume of voters affected. (R. at & 405.) The loss of these votes will, and more than likely has, changed the outcome of elections. (R. at 295.) TDP shares the concerns of many others that electronic voting machines raise serious issues regarding the accuracy of our elections. For example, the lack of a verifiable audit trail (an issue that has come up in this case), serious possibilities of manipulation, and unauthorized access to election data are real problems. However, this case does not directly address those important issues. This lawsuit results from a very specific, identifiable and objective problem with the eslate which is shown to cause a certain and identifiable volume of disenfranchised voters. STATEMENT OF FACTS The Secretary certified a direct record entry ( DRE ) electronic voting system known as eslate for use in the State of Texas. (See R. at ) The eslate machines were designed, developed, and are manufactured by Hart InterCivic. (See generally R. at ) The eslate machines record votes electronically, and they have a complicated user interface, into which the voter makes his or her selections. (R. at ) This case arises out of the way the user interface functions when the voter attempts to cast a straight-party ballot for either the Democratic or Republican 20

21 slate of candidates and then emphasis votes for candidates of the same party. (R. at 9 22.) I. The eslate s User Interface & The Emphasis Voter A voter who casts a general election ballot on the eslate begins his or her voting by entering a pin number provided by the election precinct staff. (R. at ) The voter is then presented with the first of several ballot screens. (R. at ) Assuming the voter desired to cast a Republican Party straight-ticket ballot, the voter would go to the straight-party section of the ballot and press the Select button when the Republican Party is highlighted. (R. at ) When that selection is made, each Republican candidate running for office on the ballot would have the oval next to their name automatically filled in by the eslate system. (R. at ) Assuming a voter desired to emphasis vote, as they had for many years in the past, the voter would use the wheel to scroll down to the first race presented on the ballot. (R. at ) The eslate does not prevent the voter from making choices in individual races that have already been decided by the straight-party choice. (R. at ) Because eslate s user interface does not prevent voters from emphasizing straightparty votes, an emphasis voter who selected straight-party Republican and then continued to the first set of candidates in the 2006 General Election would highlight Kay Bailey Hutchison for United States Senator and press the select key. (R. at ) 21

22 A. The Warning Screen is Ineffective and Ambiguous When the voter uses the select wheel to scroll to the first individual race and attempts to emphasis vote, a screen that Hart InterCivic calls a warning screen appears. (R. at 544.) The warning screen states Be aware that you are changing a straight-party choice. (R. at 544.) It does not state Caution: You are no longer voting for Kay Bailey Hutchison, or even, Caution: You are no longer casting a vote for United States Senator, which is the ultimate impact of the emphasis vote. (See R. at 544.) The warning is of questionable impact, as the 2006 General Election Madison County voters who provided affidavits to TDP do not recall such a screen. (R. at 297, 299 & 301.) If the voter interprets the warning screen correctly, and he or she desires to back out of it, there is no way to do so. (R. at ) Instead, the voter must press any key to continue and they are returned to the ballot screen. (R. at ) Then, the voter is left to his or her own technological acumen to figure out how to undo the de-selection of that candidate, which assumes the voter understands the consequence of his or her selections in the first place. (R. at ) Once back at the ballot screen, and assuming the voter understands the Computer Generation concept of clicking and un-clicking boxes on web pages and other computer screens, the voter might notice that no boxes are filled in for any candidate in that race. (R. at ) If the voter were to immediately cast his or her 22

23 vote the ballot would be blank for United States Senator. (R. at ) In order to regain or reselect their vote for that office, the voter must highlight his or her choice with the select wheel a second time and press the ENTER button. (R. at ) If the voter does not press the ENTER button again, but instead continues to the next race, no vote has been secured for United States Senator. (R. at ) In other words, the system is counterintuitive in that an action selecting a candidate s name actually counts as no vote at all. (See R. at ) If the voter again attempted to emphasis vote in the next race, there is no evidence that the warning screen would appear again after each race, even though no candidates were selected. (R. at ) Assuming the voter continued down the ballot in this fashion, pressing the ENTER button on each candidate from the political party of their straight-party choice, the voter would be de-selecting each candidate without warnings until the end of the ballot. At the end of the ballot, the voter is taken to the summary screen. (R. at ) B. The Ballot Summary screen is patently misleading At the conclusion of the ballot, the voter is presented with a summary screen, which, assuming a voter has made the choices described above, namely a Republican party straight ticket, and then goes down the ballot and presses the Select button on 23

24 each candidate from that same political party office, the Summary screen would appear as below: Contest Selected Straight-Party United States Senator United States Representative Governor Lt. Governor Attorney General * * * Republican No Selection No Selection No Selection No Selection No Selection *** (R. at ) A voter who sees this screen could reasonably believe that he is voting straight Republican, when actually he is casting no vote at all. (R. at ) The Ballot Summary page of the eslate would appear virtually identical to a paper ballot where the voter had selected a straight-party choice but had made no other marks in the individual races. However, when that situation occurs on a paper ballot, a vote is recorded for everyone from the political party selected. When this situation occurs on the eslate, no votes are recorded for any candidates and a blank ballot is cast. (R. at 443.) When the voter presses the Cast Ballot button, there 24

25 is no warning screen that alerts the voter in this example that he or she is casting a blank ballot. (R. at ) II. TDP Responds to the Secretary s Motion with Competent Evidence The day after filing his answer, the Secretary filed a lengthy motion for summary judgment. (R. at ) In response, TDP filed voluminous voting records, the affidavits of three Madison County emphasis voters, as well as the affidavits of the former Tarrant County election administrator, a statistical expert, TDP s Regional Field Director, and a Nueces County party official. (R. at ) The affidavits establish that emphasis voting occurs on eslate and that eslate does not effectively record votes because the system shows a significantly higher percentage of undervotes 3 (potential failed emphasis votes) than the paper ballots in the same counties. (See R. at ) For example, the Nueces County (Corpus Christi) 2006 voting records exhibit almost no undervotes for early voting (paper ballot) whereas Election Day returns (eslate) exhibit undervotes as high as 24%. (R. at ) A. The Affidavits of Madison County Voters Establish that Emphasis Voting Occurs on eslate 3 Undervotes occur when a voter casts a ballot but it is not recorded as placing a vote in one or more races. 25

26 The competent evidence TDP submitted in opposition to Defendants Motion for Summary Judgment regarding the 2006 General Election in Madison County indicate that emphasis voting occurs on eslate. TDP submitted the affidavits of Lois Brown, Frederick Randle, and Kent Kennard, all of whom are Madison County voters ( Madison County Voters ). (R. at ) All three affidavits state that the affiant participated in the 2006 General Election, using the electronic machine used in Madison County, which is eslate. (R. at ) All three state that he or she does not recall seeing any kind of warning screen. (R. at ) All three testify that each person voted for the straight ticket, but then each person also emphasis voted by selecting the individual candidates in addition to the straight ticket option. (R. at ) B. TDP Submits Ballots from Madison County Establishing Emphasis Voters Existence and eslate Disenfranchisement In addition to the ballots from Madison County, TDP produced certified copies of 36 paper ballots from Madison County that reflect emphasis voting taking place. (R. at ) In that same county during the same election, TDP produced 21 Cast Vote Records from votes placed on eslate that TDP believes reflect lost emphasis votes. (R. at ) For example, the Record at page 572 shows a voter selected straight party Republican but had no individual race selections. This ballot was counted as blank by eslate, even though it would have 26

27 counted as a vote for Republican candidates if it had been manually counted, or if it had been slipped into a ballot box. C. The Affidavit of Robert Parten Establishes the Correlation Between Emphasis Voting and eslate TDP submitted the Affidavit of Robert Parten, who was the election administrator in Tarrant County (Fort Worth) for 20 years, and has worked in the administration of elections for 42 years. (R. at 291.) In fact, Hart InterCivic s own corporate representative, Scott Flom ( Flom ), testified under oath that Parten is one of the most recognized people in [the election] field. (R. at 507.) After dealing with the eslate machine during the 2000 Presidential Election, Parten conducted an analysis of the undervotes recorded by eslate in comparison with the undervotes recorded via paper balloting devices, and he noticed a significantly higher undervote in the eslate than in the paper ballots. (R. at ) Parten opines that the reason for the difference was the voter learning curve, and that the eslate machines were not recording voter intent. (R. at 292.) Flom testified that Hart InterCivic listened to what Parten had to say. (R. at 507.) The Parten Affidavit then states that he received and collected data in the form of Cast Vote Records from the eslate voting machines used in Madison County during the 2006 General Election. (R. at 292.) Parten notes that some of the Cast Vote Records indicate a voter made a straight party choice, but cast no selection in 27

28 any of the down-ballot races, which Parten notes is suspicious. (R. at 293.) Parten further notes that the summary screen for such a voter would look similar to that of a paper ballot by showing a straight-party choice and then indicating no other marks on the ballot. (R. at ) Parten further notes that had the voter cast a paper ballot, a vote would have recorded for every candidate that the voter selected. (R. at 294.) Based on the operation of the eslate machine in reviewing the Cast Vote Records for Madison County, as well as his forty-plus years in the election administration, Parten opines that voters could be confused by the review screen, and believe they are casting a vote for all of the candidates in a particular party when instead they are casting a vote for no one. (R. at 294.) Parten also opines that the same applies to those ballots which were partially undervoted. (R. at 294.) Ultimately Parten opines that, the eslate s operation concerning straight party and emphasis voters is likely to explain some or all of the undervote observed when comparing paper and eslate voting systems. (R. at 294.) Moreover, the Parten Affidavit establishes that a manual recount of the Cast Vote Records generated by eslate and the paper ballots would produce a different result than the tabulation for eslate alone. (R. at 294.) Despite his qualifications and methods, the district court excluded the affidavit under Rule 702. (R. at 818.) C. The Affidavit of William Brannon Corroborates the Existence of Emphasis Voters. 28

29 TDP further submitted the affidavit of William Brannon, who is the Regional Field Director for the Texas Democratic Party, and his area includes Madison County. (R. at 303.) Brannon s affidavit is based on his several years of experience dealing with elections, as well as with recounts. (R. at 303.) Brannon essentially confirms the existence and occurrence of emphasis voting and he further opines that the only method to discern the intent of the voter with eslate is to print the Cast Vote Records, which is a burdensome additional expense that many candidates, especially local candidates, cannot shoulder. (R. at ) D. The General Election Data from Nueces County Shows a Disparity Between eslate and Other Systems TDP also submitted General Election Data by precinct for the 2006 general election in Nueces County. (R. at ) In Nueces County, as in most others, the absentee balloting is done by paper, whereas the early voting and Election Day voting are done electronically. (R. at 684.) Even a cursory review of the undervote percentages between the paper absentee ballots and the early and Election Day electronic ballots shows a much higher percentage of undervotes than the electronic voting. (R. at ) A reviewer will note that the paper ballots record almost no undervote in one precinct after another. For example, in Precinct One, absentee ballots submitting casting votes for Kay Bailey Hutcheson had a 0% undervote rate, whereas the early and Election Day eslate 29

30 voting had 4.48% and 2.99% undervotes respectfully. (R. at 306.) The numbers are even higher in other precincts. Consistent with the other evidence TDP submitted, the records from Neuces County show that eslate fails to record voter intent in Nueces County just like it does in Madison County. (Compare R. at 306 with R. at 678.) E. The Affidavit of Daniel J. Francis Confirms that eslate has a Statistically Significant Higher Undervote Rate Than Paper Ballots To support its claim that the increase in eslate undervotes is higher than on other systems, TDP submitted the expert Affidavit of Daniel J. Francis ( Francis Affidavit ). (R. at ) Francis performed a statistical analysis on the submitted election data. (R. at ) The Francis Affidavit establishes with scientific certainty that there is a statistically significant increase in undervotes when eslate is used, in comparison to paper balloting. (R. at ) The Francis Affidavit includes detailed descriptions of the data relied on and the statistical methodology, which included running detailed computer analyses using ANOVA (Analysis of Variance), which is one of the most widely used and peer reviewed statistical techniques in academia and in the corporate realm. (R. at 667.) The ANOVA showed that there was an extremely low probability of chancecorrelations between undervotes and eslate. (R. at 670.) In other words, something real, definite, and repairable is causing the higher eslate undervote. TDP offered 30

31 sufficient evidence to raise a fact issue that the undervote is caused by disenfranchised emphasis voters. The district court s opinion does not address the Francis Affidavit. 31

32 SUMMARY OF THE ARGUMENT TDP sought declaratory relief and an injunction to prohibit the Secretary from further certifying the eslate voting system for use in Texas unless, or until, it is reprogrammed to address the problems with its confusing and misleading voter interface. TDP also sought an injunction requiring the Secretary to decertify eslate so that it may not be employed in any General Election until it has been redesigned. The fundamental problems with the eslate voting system manifest themselves twice. The first manifestation is in the actual entry of the ballot selections. The selection/deselection process is confusing to voters, and the warning screens are either ineffective or they cause further confusion. The user interface makes it less likely that a vote will be cast and is proven by the submitted election results. The second manifestation occurs during tabulation, particularly during a manual recount. When the voter sees the ballot summary at the end of the process, an emphasis voter would potentially see that the straight party option is selected, but that in each individual race; there are no selections. The eslate system counts this as a blank ballot. If the emphasis voter was using a paper ballot in the same county, the paper ballot would look the same, yet it would be counted as a party vote for each race on the ballot. That is, there is no uniform rule to discern the voter s intent. 32

33 First, the district court abused its discretion when it excluded the Parten Affidavit, as Parten is a well-respected election official who based his conclusions on data that he collected and on reliable principles. Second, the eslate, as it presently operates, burdens the fundamental right to vote in violation of the Equal Protection and Due Process Clause of the Fourteenth Amendment because eslate treats similarly situated voters differently and the user interface makes it more likely that eslate voters will cast an inaccurate and ineffective ballot than voters using other systems. Third, and for the same reasons, the Secretary violated the Due Process Clause of the Fourteenth Amendment. Fourth, the Secretary is in violation of the Federal Help America Vote Act ( HAVA ), by failing to adopt uniform and nondiscriminatory standards that define what constitutes a vote and what would be counted as a vote for each category of voting system used in this state. Fifth, Secretary has violated the Texas Election Code by certifying an election system that fails to count a voter s intent when it is clearly ascertainable. 33

34 ARGUMENT & AUTHORITY This case is not a broad-based indictment of the electronic voting platforms used in Texas, nor is it about a right to emphasis vote. Rather, this case is about requiring election machines to record the voters intent, and the user interface should help not hinder the process. There are three specific, fundamental defects in the process which TDP aims to correct: 1. After a straight-party has been selected, eslate should not by default deselect a candidate who is selected again, or it should provide a no selection option; 2. The warning screen should inform a voter explicitly that a choice is being de-selected and that no vote has been secured; and 3. The summary screen should not indicate that a straight party ticket is being cast when in fact the ballot is blank. In the alternative, Hart InterCivic could likely correct the problem by inserting an effective warning screen at the conclusion of the balloting that informs the voter that a 34

35 blank ballot is being cast. 4 The Secretary had advanced no evidence of any downside to making these programming changes. I. The District Court Erred When it Excluded the Affidavit of Robert Parten Under Federal Rule of Evidence 702 In response to Defendant s Motion for Summary Judgment, TDP submitted the Affidavit of Robert Parten as Exhibit A ( Parten Affidavit ). (R. at ) In the affidavit, Parten states: I was the election administrator in Tarrant County, Texas for 20 years. I have been involved in the administration of elections for 42 years. I have extensive experience in administering elections. I have been involved in many recounts and have observed countless ballots. (R. at 291.) Parten s qualifications to testify as an expert on election matters is buttressed by Hart InterCivic s own corporate representative testifying under oath that Parten is one of the most recognized people in [the election] field. (R. at 507.) Parten ultimately opines that the way eslate handles straight-party and emphasis voters explains the high undervote rate in eslate. (R. at ; see also R. at 818.) 4 At the very least, TDP should have been allowed to conduct discovery from the various counties using eslate, and it should have been allowed by the district court to submit its claims to the fact finder to determine the fact issues that were raised in the summary judgment response. 35

36 The district court s ruling that the Parten Affidavit was inadmissible was an abuse of discretion. See Boyd v. State Farm Ins. Cos., 158 F.3d 326, 331 (5th Cir.1998). Contrary to the district court s ruling, Parten did collect data on undervotes and emphasis votes. (R. at 818.) Parten states that he reviewed Cast Vote Records from the eslate voting machines in Madison County, reflecting undervotes. (R. at 292, referencing R ) Parten states that he conducted a similar study following the 2000 General Election and that he found similar errors in that contest. (R. at ) Parten then recites the operation of the eslate interface concerning emphasis votes, indicating that the summary screen of eslate would look like a paper ballot but that, unlike the paper ballot, the eslate would record no vote. (R. at ) Based on his 42 years of experience in election administration, Parten ultimately concludes: [R]eviewing the Cast Vote Records for Madison County, I believe voters could be confused by the review screen and believe they are casting a vote for all of the candidates in a particular party when instead they are casting a vote for no one. In the event a voter had emphasis voted in only a few races, I believe a voter could be confused by the review screen, showing no selection in those few races, but still observing a straight-party choice at the top of the review screen. A voter who casts a ballot after seeing such a review screen may believe they were voting for every candidate in the political party of their choice. (R. at 294.) 36

37 The district court expressed concern that Parten did not connect undervotes to emphasis voters any more than to other possible explanations.... (R. at ) On the contrary, Parten specifically states in his affidavit that the high undervote of eslate is not readily explained by voters merely choosing not to vote in every race because voters undervote on paper balloting systems and on eslate, but eslate s undervote rate is higher: The eslate s operation concerning straight-party and emphasis voters is likely to explain some or all of the undervote observed when comparing paper and eslate voting systems. (R. at 294.) The subtext of the district court s exclusion of the Parten Affidavit was essentially Parten s credibility. (R. at 818.) Exclusion on that basis is improper and manifestly erroneous. See, e.g., Viterbo v. Dow Chemical Co., 826 F.2d 420, 422 (5th Cir.1987) (holding that questions regarding the scientific basis of an expert s opinion affects the weight to be assigned that opinion rather than its admissibility and should be left for the jury's consideration. ) Furthermore, the district court based its grant of summary judgment, in large part, on its exclusion, making the error harmful and reversal is warranted. See St. Romain v. Industrial Fabrication and Repair Service, Inc., 203 F.3d 376, 381 (5th Cir. 2000). 37

38 II. The District Court Erred When it Dismissed TDP s Claims Arising Under the Equal Protection Clause of the Fourteenth Amendment For two related reasons, the district court erred when it summarily dismissed TDP s equal protection claims seeking declaratory and injunctive relief. 5 First, the district court applied the Anderson-Burdick rational basis test when it should have applied the Reynolds strict scrutiny analysis. Compare Anderson v. Calabrese, 460 U.S. 780, 784 (1982) (ballot access); and Burdick v. Takushi, 504 U.S. 428, (1992) (ballot access) with Reynolds v. Sims, 377 U.S. 533, 554 (1964) (voting rights). Second, even if mere rational basis applies, TDP raised genuine issues of material fact on each element of its constitutional claims. The Fourteenth Amendment provides: No State shall make or enforce any law which shall abridge the privileges or immunities of citizens of the United States; nor shall any State deprive any person of life, liberty, or property, without due process of law; nor deny to any person within its jurisdiction the equal protection of the laws. 5 The eslate system inflicts a concrete harm on TDP, as it is forced to then re-educate voters on different types of voting platforms across the state. Political parties have a vested interest in getting voters to select the straight ticket, and TDP uses a significant portion of its economic resources encouraging Texas voters to do so. The eslate puts the TDP in the awkward position of motivating voters to cast straight ticket ballots but not motivating the voters so well that they emphasis vote. TDP also has associational standing because its members who emphasis vote and do so on eslate machines are less likely to have his or her vote counted. See Texas Democratic Party v. Benkiser, 459 F.3d 582 (5th Cir. 2006). 38

39 U.S. CONST. amend. XIV 1. The Fourteenth Amendment applies to protect the right to vote in both state and federal elections. Reynolds, 377 U.S. at 554 It is firmly established that the right to vote extends beyond casting of the ballot, and extends to the right to have the vote counted with equal weight as other votes. Id. at ( [I]t is as equally unquestionable that the right to have one s vote counted is as open to protection... as the right to put a ballot in a box. ) (internal quotations omitted). right to vote: The Reynolds Court did not mince words when it articulated the gravity of the The right to vote freely for the candidate of one's choice is of the essence of a democratic society, and any restrictions on that right strike at the heart of representative government. And the right of suffrage can be denied by a debasement or dilution of the weight of a citizen s vote just as effectively as by wholly prohibiting the free exercise of the franchise. Id. at 555. And in an earlier decision: [T]he right of individuals to associate for the advancement of political beliefs, and the right of qualified voters, regardless of their political persuasion, to cast their votes effectively. Both of these rights, of course, rank among our most precious freedoms. Williams v. Rhodes, 393 U.S. 23, (1968). Again, this is not a case about restricting a candidate s access to the ballot; it is about the right of the voter to cast an 39

40 effective ballot for the candidate of his or her choosing. In other words: does the voting system serve the voters or do the voters serve the voting system? The Secretary s certification of eslate dilutes the weight of the straight-party citizen s vote because it: (1) uses a confusing summary screen; (2) lacks effective warnings; and (3) tabulates votes differently than paper ballots. Ultimately, eslate fails to record the intent of the voter. A. The Standard of Review In the interests of thrift, TDP asks the Court to note that this standard of review applies to all points of error concerning the grant of summary judgment. Rule 56(c) of the Federal Rules of Civil Procedure is clear. Summary judgment shall be entered as a matter of law only if the summary judgment evidence shows that there is no genuine issue as to any material fact. See FED. R. CIV. P. 56(c). This Court reviews the grant of a summary judgment de novo, applying the same standard of review that the district court did, viewing the facts and resolving all inferences in favor of the non-movant. Hainze v. Richards, 207 F.3d 795, 798 (5th Cir. 2000). If a rational trier of fact could find for the non-moving party, summary judgment is inappropriate. James v. Sadler, 909 F.2d 834, 837 (5th Cir.1990). B. The District Court Erred When It Used the Anderson-Burdick Test Because This is Not a Ballot Access Case 40

41 In its opinion, the district court erroneously concluded this was a ballot access case, and that the impact on voter s constitutional rights was minor. The Court then applied the Anderson-Burdick test in its evaluation of TDP s claims arising under the Fourteenth Amendment. As illustrated below, the proper standard in a voting rights case is strict scrutiny as articulated in Reynolds and its progeny. This case is about having one's ballot counted not getting on it. 6 Using the incorrect legal standard, the district court erroneously held that the use of eslate voting machines advances the state=s important interest in regulating elections, and any infringement of an emphasis voter s right to vote is permitted because the impact is slight. (R. at ) See Anderson v. Calabrese, 460 U.S. 780, 784 (1982) (noting the ability of states to regulate access to the ballot for candidates). This was error. The lesser standard of Anderson-Burdick is not the correct analysis in a right to vote case. To illustrate the distinction, the Supreme Court has notably defined the focus of a ballot access case: Our ballot access cases, however, do focus on the degree to which the challenged restrictions operate as a mechanism to exclude certain classes of candidates from the electoral 6 Oddly, the district court began its analysis by citing Reynolds for the proposition that the right to vote in federal and state elections is protected by the Constitution, but then it switched to the Anderson-Burdick analysis. (R. at ) 41

42 process. The inquiry is whether the challenged restriction unfairly or unnecessarily burdens the availability of political opportunity. Clements v. Fashing, 457 U.S. 957, 964 (1982) (emphasis added), citing Lubin v. Panish, 415 U.S (1974). In contrast, a right to vote case focuses on the right of qualified voters within a state to cast their ballots and have them counted. Reynolds 377 U.S. at 555, citing United States v. Classic, 313 U.S. 299, 315 (1941). Cases involving the right to vote have been treated with more animus by the courts than have those cases involving ballot access, so the distinction is critical. Compare, e.g., Reynolds, 377 U.S. 533 (apportionment case); Gomillion v. Lightfoot, 364 U.S. 339 (1960) (racial gerrymandering) with Clements, 457 U.S. 957 (candidate access to ballot); Anderson, 460 U.S. 780 (candidate access to ballot). TDP does not allege that the Secretary is restricting access to the ballot because the certification of eslate has no effect on who is on the ballot, the ballots shown on an eslate lists the same candidates as do the paper ballots used in other counties for the same races, and nothing about the use of eslate affects the ability of a candidate to be named on the ballot. 1. This right to vote case is unlike those ballot access cases that apply a rational basis inquiry. 42

43 The facts of this case are readily distinguishable from Anderson and Burdick, and the district court s reliance on them was misplaced. In both cases, the Supreme Court found the challenged restriction was proper because it did not excessively burden access to the ballot and held the strictness of the review of access depended on the severity of the restriction the greater the restriction, the stricter the review. See Burdick, 504 U.S. at 433; Anderson, 460 U.S. at Fifth Circuit cases recognize these holdings. Texas Indep. Party v. Kirk, 84 F.3d 178, 182 (5th Cir. 1996); Strayhorn v. Williams, 430 F. Supp. 2d 661, (W.D. Tex. 2006) (holding signature verification for petition to be placed on ballot is permissible). The Sixth Circuit has also observed that ballot access cases have a Apractical@ or Aprudential@ dimension (i.e., the states have to prepare ballots in time to have them distributed for an election, the ballots must be of a workable size, etc.) that is absent from right to vote cases, and so strict scrutiny is the level of review that ought to be applied to right to vote cases. Stewart v. Blackwell, 444 F.3d 843, (6th Cir. 2006), vacated as moot, 473 F.3d 692 (6th Cir. 2007) (per curiam). Anderson arose out of the independent candidacy of John Anderson in the 1980 Presidential election. Anderson, 460 U.S. at Anderson challenged certain deadlines imposed by the State of Ohio for including a candidate on the ballot. Id. at 783. Anderson challenged the constitutionality of Ohio s filing deadline to submit the necessary paperwork to get a place on the ballot. Id. at 783. The case did not concern 43