IN THE UNITED STATES DISTRICT COURT
FOR THE EASTERN DISTRICT OF TEXAS
MARSHALL DIVISION
LEAGUE OF UNITED LATIN
AMERICAN CITIZENS, ET AL. §
vs. § CIVIL ACTION NO. 2:03-CV-354
RICK PERRY, ET AL. § Consolidated
APPLICATION FOR LEAVE TO FILE PETITION IN INTERVENTION
Applicant submits this intervention to supply deficiencies in the arguments made, and expected to be made, by other litigants that fail to properly address the concerns of the United States Supreme Court in its remand of June 28, 2006. That remand was an invitation to present standards to the Court by which it could decide when congressional district maps are drawn in a manner that is partisan to a degree that make the constitutional issues justiciable. Applicant presents such standards that, if adopted by the Court, can lead the way to ending the corrupt practice of gerrymandering after more than 200 years, and entirely remove undue or excessive partisan influence on the election of members of the U.S. House of Representatives.
Jon Roland, in propria persona
7793 Burnet Road #37
Austin, Texas 78757
Telephone: (512)374-9585
Email: jon.roland@constitution.org
July 12, 2006
IN THE UNITED STATES DISTRICT COURT
FOR THE EASTERN DISTRICT OF TEXAS
MARSHALL DIVISION
LEAGUE OF UNITED LATIN
AMERICAN CITIZENS, ET AL. §
vs. § CIVIL ACTION NO. 2:03-CV-354
RICK PERRY, ET AL. § Consolidated
GRANT OF LEAVE TO FILE PETITION IN INTERVENTION
Leave to file Petition in Intervention is hereby granted to Jon Roland in this case.
_____________________________________________ __________________________
Judge presiding Date
IN THE UNITED STATES DISTRICT COURT
FOR THE EASTERN DISTRICT OF TEXAS
MARSHALL DIVISION
LEAGUE OF UNITED LATIN
AMERICAN CITIZENS, ET AL. §
vs. § CIVIL ACTION NO. 2:03-CV-354
RICK PERRY, ET AL. § Consolidated
PETITION IN INTERVENTION OF JON ROLAND
Jon Roland, in propria persona
7793 Burnet Road #37
Austin, Texas 78757
Telephone: (512)299-5001
Email: jon.roland@constitution.org
Summary
In this petition Intervenor, a citizen of the State of Texas, will offer:
1. A standard of proof by which the Court may determine what is and is not constitutionally impermissible adoption of district maps for the U.S. House of Representatives.
2. Argument that it is not the maps per se but the intent and method of creating and adopting them that may be in violation of the Constitution and Voting Rights Act.
3. Argument that the manual revision of a few of the districts of the 2003 map, even if done by the Court, is just as improper as the drawing of those districts in the first instance, and that only a new map, produced by a new method, is compliant with both the Constitution and the Voting Rights Act.
4. Argument that however defective might be the 2003 map, it is too late to revise it for the 2006 election, and the Court should adopt a new method for producing maps for subsequent elections.
5. A method for producing maps for subsequent elections, that satisfies the standard of proof offered and also the U.S. Supreme Court in it’s decision of June 28, 2006.
Standing
In this and similar cases the courts have recognized citizens or voters of a state as having standing to challenge legislative district maps, and Intervenor appears as a citizen of the State of Texas. However, corporate entities other than the State itself should not be considered to have standing in their corporate capacity, and Intervenor moves that they be restyled by prepending “Texas Members of” or “Residents of” to their corporate names, to emphasize this point.
Litigant-intervenors have been arguing this case as though they are asserting a property
right in the district maps, but there can be no property right in district maps. Their standing in this
case must be based on the ancient principle of having the right to privately prosecute a public right,
and that may require us to re-examine the precedent set in Frothingham v. Mellon, 262 U.S. 447
(1923).
Background
This case has emerged as a contest between two groups of litigant-intervenors, one supporting the Map of 2003, gerrymandered in favor of the Republican Party, and the other the undoing of the Map of 2003 and a return to what we may call the Map of 2001, which is gerrymandered in favor of the Democratic Party. Now the election of 2004 has already been conducted using the Map of 2003, and members of Congress elected from districts of that map.
Attorneys for the litigant-intervenors opposing the Map of 2003 made cogent arguments against the Map of 2003 in their filings, but those arguments apply equally well to the Map of 2001.
In its decision of June 28, 2006, the U.S. Supreme Court offered diverse opinions, most of which expressed dissatisfaction with the course of this case, but a majority could be assembled only on the narrow finding that District 23 of the 2003 map is in violation of the “dilution” clause of the Voting Rights Act, and on that basis remanded. However, it is not possible to redraw just one district, and any redrawing of just the districts adjoining District 23, while it might reduce the “dilution” of Hispanics in that district, can only do so by “diluting” those adjoining districts, and that would also be in violation of the Voting Rights Act. Therefore, the only way to comply with that Act is to redraw the entire map.
The present situation is that both major parties have already conducted their primaries, and the minor parties their district conventions, based on the 2003 map. No funds are appropriated to hold new primaries. Nominees have raised funds, filed reports, made media purchases, rented office space, and otherwise committed to the 2003 map. As loathe as Intervenor is to argue in support of reliance interests in a constitutional case, he sees no remedy at the present time that is not worse than continuing with the 2003 map for the 2006 election, and while he makes no motion on redrawing the present map, will move to adopt a method for redrawing maps for subsequent elections.
In the previous brief of the Jackson plaintiffs and Democratic Party intervenors the duty of this court and all parties is defined well by the characterization of the position of the U.S. Supreme Court in a previous remand by a “majority of the Supreme Court Justices have now held that partisan-gerrymandering claims remain justiciable, all members of the Court have now agreed that severe partisan gerrymanders violate the Equal Protection Clause, ...” but that the plurality “despaired of finding a workable standard”, citing Vieth v. Jubelirer, 124 S. Ct. at 1785.
Based on the briefs Intervenor has seen, none of the litigant-intervenors has properly or adequately addressed the concerns of the United States Supreme Court in its first remand and direction that the case be reconsidered in the light of Vieth v. Jubilirer. That remand was an invitation to present the court standards by which it could decide when congressional district maps are drawn in a way that is partisan to a degree that the constitutional issues are justiciable. Nor can it be expected they will do so in response to the remand of June 28, 2006. The failure of litigant-intervenors to thus respond borders on contempt, and presented the court with the unenviable choice between two alternatives, both of which are unconstitutional. A third map, developed using the same methods used to produce the first two, would not be a solution, because the process is the problem. Intervenor presents an entirely different alternative, and standards by which the corrupt practice of gerrymandering can be ended after more than 200 years, bringing to a close not just partisan influence that is undue or excessive, but removing it altogether.
What has heretofore been missing from these lines of argument is the recognition that partisan gerrymandering is not an attribute of maps so much as of the process of producing and adopting them. Such undue influence, like ethnic discrimination, is a mental act, and the maps are evidence of that mental act. When maps are drawn by partisan human beings, no matter how conscientious they might be to avoid partisan bias, there will inevitably be undue influence exerted upon those doing the drawing and adopting, and also by those doing the drawing and adopting on those with a stake in the outcomes. There is a problem even if no one yields to such undue influence and it is only a matter of appearance and not of substance. Corruption works both ways.
The new situation
For more than 200 years partisan gerrymandering has been taking place in the United States. It is named for Elbridge Gerry, one of the Founders. This legacy is due in no small part to the fact that there was no alternative method available of drawing maps than to have it be done by human beings, and the process was so laborious that it was impractical to produce large numbers of maps and have the final map be selected at random rather than by legislative debate and vote on a particular map.
That situation has changed. The technology now exists to have a computer program draw maps randomly that soon satisfy almost any combination of mathematical constraints one might want to impose on them, without any manual intervention. Even though the constraints remain the same, each time the program is run, a quite different map is produced. Some such programs are commercial products, readily available at a modest cost. But perhaps the best is called TARGET, written by staff programmers of the Texas Legislative Council. It works well, feeding its results to the REDAPPL program used now to manually draw maps, which also produces the legal descriptions of the districts.
It is no longer necessary for courts to be asked to examine maps as possible instances of severe partisan gerrymandering. The technical means now exists to end forever the corrupt process that produces gerrymandered maps, by letting machines produce them, mechanically, without influence by partisan factors. Courts or legislatures need only specify the constraints and administrative procedures, generating many and adopting one at random, without debate over adoption of the final version.
Three years ago the court could have visited the staff offices of the Texas Legislative Council at 1501 N. Congress Avenue, Austin, Texas 78711 tel. (512)463-1143 for a demonstration of the TARGET program used to generate maps according to the constraints proposed in this brief. Unfortunately, since then changes have been made in the structure of the voter database and staff resources were not allocated to modifying TARGET to continue to be able to work with the revised database. Staffers estimate it would take them about two months to do so if they were asked. However, that would be to enable use of all the fields of the database, and for the purpose of this proposal, it would only need to use the voter counts and locations, and not voting history, ethnicity, or other demographic factors. Intervenor expects for this restricted use TARGET could be made ready in less time.
Proposed solution
Intervenor moves the Court adopt the following solution:
(1) Constraints on the maps. The smallest unit of area shall be the voting precinct, as presently established by law, which shall be of equal population within a county and not differ in population from county to county by more than necessary to accommodate counties of low population.
(a) Equipopulous. The population of each district shall not differ from that of any of the other districts by more than a factor of 0.0001 or the margin of error of the census count, whichever is less.
(b) Aligned. Only counties with a population of more than a factor of between 0.1 and 1/3, initially 1/3 unless or until amended by the State Legislature, of the average population of a congressional district, may be split between districts, unless a larger number of counties must be split to meet the specification (a) above, and a smaller number of counties shall be split if specification (a) can be met.
(c) Contiguous. Districts must be contiguous, so that there is always at least one continuous line of points connecting any two points within the district, and no connection between parts consists only of a line or point.
(d) Simply connected. Districts must be simply connected, so that any continuous loop of points within the district may be shrunk to a point within the district without crossing boundary lines. This means no holes in districts, and no surrounding of one district by another.
(e) Compact. Districts shall be maximally compact, resulting from a running time of at least 6 and not to exceed 24 hours, adjusted for improvements in processor speed, with compactness defined by minimizing the value of p²/4πA, where p = perimeter and A = area of the district, with all other values remaining constant or improving. For an area bounded by a circle the value of this expression is 1.
(2) Procedures. District maps shall be generated and finally adopted mechanically with minimal human intervention using a computer program.
(a) The software to be used initially shall be the TARGET software already developed, but may be modified or replaced at the discretion of the State Legislature thereafter. But source code for the production version of any computer redistricting software and the database shall be made accessible for downloading from the web site of the State for public examination and comment, and to be shared with other states and communities, at no cost other than storage media.
(b) The State shall establish and maintain adequate safeguards to insure that no unauthorized alterations are made in the software or interventions made in the running of it that might bias the output. As soon as feasible, a version of the present database containing only information needed to satisfy the public constraints established herein or by act of the State Legislature shall be prepared, and made the only database accessible to the redistricting program during the generation of maps for official selection.
(c) A commission or jury, hereinafter called the “Commission”, consisting of at least twelve and not more than 23 individuals, initially the Texas Legislative Council unless or until the State Legislature shall provide otherwise, shall supervise the redistricting process.
(d) Initially, and thereafter during one month every two years prior to each Congressional election, and after the most recent decenniel census results are available, the Commission shall cause to be randomly generated at least twice as many maps as there are members of the Commission.
(e) Each Commission member shall have the right to reject or strike one map from among the maps randomly generated during the current redistricting session.
(f) One map shall be selected at random from among the randomly generated maps that remain after strikes, and that map shall become the district map for the next election without amendment or debate.
(g) The biannual schedule of redistricting may revert to a decenniel schedule after 2010, if the State Legislature so enacts.
Standard of Proof
Intervenor moves the Court declare the following to be the Standard of Proof:
A method of producing and adopting district maps will be deemed to violate Equal Protection if the mean predicted outcome of the candidate pool from which a randomly selected candidate map is drawn differs by one or more from a norm defined as follows:
1. A statistically large enough pool of maps, probably at least 20, is generated at random, to produce a standard pool.
2. Using voting history data, a prediction is made of the outcome of the election for each map in the standard pool and averaged to yield a pool mean, which is the norm. If there is only one candidate map, then it is a candidate pool of one.
The Standard of Proof is illustrated by Exhibit A, in which the blue curve represents the 2001 map, the red curve the 2003 map, and the green curve the standard for comparison. Each curve is a normal or Gaussoid distribution of predicted outcomes of multiple runs of the prediction algorithm. The mean of the blue curve is set, for purposes of illustration, at15 Republican seats, and the red curve at 21, based on actual outcomes. The mean of the green curve is 18, as an educated guess. By this method, in this illustration, the blue 2001 map and red 2003 map both differ from the norm by 3, in different directions, and therefore fail the standard. However, these curves are for illustration purposes only, as and the procedure would have to be actually run to more accurate curves and their means.
This standard would have to be adjusted for other states, with fewer counties or more districts, on elements such as how many counties to split. But it should be possible to eventually arrive at a formula.
The proposed solution satisfies the Standard of Proof because it involves producing maps the same way they are produced in the Standard.
The proposed solution satisfies a proper application of the Voting Rights Act because it affords no opportunity for the exercise of discriminatory intent.
Satisfaction of other or previous constraints
The question must be addressed of what to do about compliance with such constraints as provisions of the Voting Rights Act that proscribe dilution of minority-majority districts. While this can be made an additional constraint on the maps to be generated by the redistricting software, there are several difficulties with that requirement that must be addressed by Congress and the courts.
(1) Demographic trends may make it difficult or impossible to attain. Integration is working. Ethnic groups are becoming increasingly dispersed and intermingled, such that if present trends continue, it may become impossible to find any locality were a minority is a majority, or even concentrated to a high degree, at the level of a congressional district.
(2) The requirement presumes manually-drawn maps, which have historically been susceptible to ethnic gerrymandering. Replacement of a manual process by a blind mechanical process which has no information about ethnicity effectively eliminates the factor of ethnic discrimination. If minority representation remains a valid political goal, and discrimination is no longer possible, then it will be time to initiate public debate on the issue.
(3) For the above reasons, it should not be deemed a violation of the Voting Rights Act if the map adopted by the above process happens to dilute minority-majority districts, because such a chance result cannot be evidence of discriminatory intent.
Further argument
Now all of this would seem to suggest that the court “legislate from the bench”, something neither the court nor Intervenor would favor under any but extraordinary circumstances, but we have to ask whether the State Legislature is going to do its duty to properly legislate in this area and not continue to abuse its legislative discretion, and if it is not, and in the absence of any legislation from Congress, whether the court has any choice but to intervene with a solution as complex as the situation requires to satisfy the letter and spirit of the Constitution and the intent of the Voting Rights Act.
Intervenor drafted two alternative bills for the Texas Legislature, but after canvassing nearly the entire membership, could not find a single member willing to sponsor legislation that would accomplish what is being proposed in this petition. He received a great deal of favorable comment to the effect that his was the most rational, sensible, and statesmanlike alternative, but members were unwilling to defy their party leadership, some expressing real fear. It would seem that the Republican majority wants to amplify its power to the maximum extent possible, the Democrat Party wants an issue it thinks will help it regain the majority after the next election, and neither seems devoted to the Constitution or to doing what is best for the people. In the face of this relentless dereliction, it would seem the court has little choice but to consider a judicial solution like this, nor has the Intervenor any choice but to propose it in this somewhat unconventional petition.
Alternatives to Single-Member Districts
Although not a question before this Court, it should be pointed out that there are alternatives to single-member districts that, if adopted, would avoid the problems confronted in this case. One of them, proxy voting, would not require a constitutional amendment, but only changes in both state legislation and in the rules of procedure of the U.S. House of Representatives. All candidates would run at large in the state. The n candidates, where n is the number of members allocated to that state, who receive the most votes would be seated as members of the House, where each would cast a number of votes in all proceedings equal to the number of votes he got in the last election. That would enable any group of voters larger than 1/n of the population of the state to elect a member. With modern electronic voting vote counts should not delay proceedings.
About the Intervenor
Jon Roland is founder and president of the Constitution Society, established in 1994, with website
at //constitution.org, and editor of the digital online editions of most of the more
important works of constitutional history, law, and government, which can be found at that site.
He is the nominee in 2006 for the office of Texas Attorney General of the Texas Libertarian Party.
By profession he is a computer programmer, but has not contributed to the development of the
software discussed in this petition.
Respectfully submitted,
_____________________________________________
Jon Roland, in propria persona
7793 Burnet Road #37
Austin, Texas 78757
Telephone: (512)299-5001
Email: jon.roland@constitution.org
July 12, 2006
____________
