From Casetext: Smarter Legal Research

Atkinson v. Carter

Court of Appeals of Texas, Houston, Fourteenth District
Mar 1, 1990
785 S.W.2d 449 (Tex. App. 1990)

Opinion

No. A14-90-112-CV.

March 1, 1990.

W. Troy McKinney, Houston, for relator.

John F. Carter, III, Bonnie J. Fitch, David F. Webb, Houston, for respondent.

Before J. CURTISS BROWN, C.J., and JUNELL and MURPHY, JJ.

OPINION


Justice Holmes was referring to the common law when he observed, "The life of the law has not been logic; it has been experience." O.W. HOLMES, THE COMMON LAW 1 (1881). Jurisprudence under the Texas Election Code would make Holmes happy. This case is a mandamus action which involves the first amendment, federalism, forum selection, statutory requirements for judicial candidacy, and the permissible remedies for failure to satisfy those requirements.

I. FACTS

At issue is the candidacy of Bonnie Fitch for the office of judge, Harris County Criminal Court at Law Number Thirteen. Relator Mark D. Atkinson, the incumbent judge of that court, faces re-election this year. He brought this original proceeding against Jack Carter, Chairman of the Harris County Democratic Executive Committee, seeking a writ of mandamus ordering Carter to remove Fitch's name from the primary election ballot. Atkinson contends that Fitch failed to gather the requisite number of signatures from qualified voters. To put it more precisely, he claims many of the signatures on her petitions either are not those of registered voters, see TEX.ELEC.CODE ANN. § 141.063(1) (Vernon 1986) (signer must actually be a registered voter), or do not comply with the statute's demand for certain data: voter registration number, street address, zip code, and so on. See id. § 141.063(2). In support of this claim he has submitted copies of Fitch's petitions, plus a detailed compilation by a professor of political science. That compilation is a computer generated tabulation of all signers, cross-referenced by alleged defect, and designed to account for signatures with more than one alleged defect so that the final tally does not appear artificially low by virtue of double-counting the signatures to be disregarded.

There is no dispute over the statute's requirements. A would-be candidate for the office sought must either pay a filing fee and tender a petition with 250 signatures, or tender a petition with 750 signatures. TEX.ELEC.CODE ANN. § 172.021(e) (Vernon Supp. 1990); see Cohen v. Rains, 745 S.W.2d 949 (Tex.App. — Houston [14th Dist.] 1988, orig. proceeding); Plummer v. Veselka, 744 S.W.2d 347 (Tex.App. — Houston [1st Dist.] 1988, orig. proceeding). There is strong disagreement, however, over the law's constitutionality. Fitch makes a first amendment challenge to application of this part of the Election Code to her, and she also denies the invalidity of certain signatures. In an effort to find a more favorable forum to hear her constitutional claim, she ostensibly removed this case to federal court. We say ostensibly, because it is not entirely clear whether a nonparty can exercise the right of removal.

These numbers do not apply statewide, but only in counties having a minimum specified population. See TEX.ELEC.CODE ANN. § 172.024(a)(8, 10, 12) (Vernon Supp. 1990).

Relator first filed his motion for leave to file a petition for writ of mandamus in our court on Friday, February 9, 1990. We granted leave and scheduled oral arguments for 3:00 p.m. Friday, February 16. Upon appearance of the parties — relator, respondent, and the real party in interest — Fitch's counsel advised us of her recent invocation of federal authority. Hearing this, we inquired whether we could even entertain arguments, given that an improper removal is nevertheless an effective one until the federal court orders a remand. All attorneys graciously assured us they had no objections to our hearing their presentations. One item which became clearer at argument was the comparative positions of the participants: relator Atkinson had just learned of the purported removal, real party in interest Fitch had sought the removal, and respondent Carter was "not opposed" to the removal, an explainable circumstance given his absence from the country.

She had taken the appropriate steps in federal court, and a freshly filed notice of removal found its way from our clerk's office to the courtroom by the time arguments ended.

Counsel informed us they had located a federal judge to hear arguments on the removal question that same Friday afternoon as soon as our hearing came to an end. But with the following Monday being President's Day, no one seriously anticipated a ruling from federal court until Tuesday, February 20, a deadline with its own significance for reasons we shall explain later. As expected, a remand order came on Tuesday from the United States District Court for the Southern District of Texas. Accordingly, the case is properly here for decision and we turn to the merits.

The Honorable Kenneth Hoyt (formerly our colleague on the First Court of Appeals) held the writ proceeding not to be a civil action for purposes of the removal statute, 28 U.S.C. § 1441. That holding had two collateral effects beyond simply returning the case to us. First, it foreclosed the possibility of distinguishing the respondent from the real party in interest within the context of removal practice. Therefore the fact that the removal request came from Fitch instead of Carter made no difference: under Judge Hoyt's reasoning the result would have stayed the same no matter who tried to invoke federal jurisdiction.
Second, it leaves open the possibility that we may choose not to follow Fifth Circuit law on the constitutionality of the Texas Election Code scheme. Had this case remained in federal court, much of that decision would have been controlled by Pilcher v. Rains, 853 F.2d 334 (5th Cir. 1988).

II. STATUTORY COMPLIANCE

Atkinson sees various irregularities in Fitch's petitions, and he alleges those defects to be fatal. First there is the matter of multiple signatures. Fitch tendered a total of 1087 signatures. Of those 1087, 304 signatures appear more than once; some appear twice, a few three times, and one four times. All sides agree on the impermissibility of duplication. See Cohen v. Rains, 745 S.W.2d at 954-55 (prohibiting double-counting). Thus there remain only 783 unique entries.

Relator next offers an array of alleged defects among those 783, claiming that enough signatures are invalid to drop the number below the necessary 750. He says 41 of the persons listed are not registered voters as required by § 141.063(1). This contention is supported by an affidavit from the Voter Registrar's office. In response, however, we have been given an affidavit from the assistant primary director of the Harris County Democratic Executive Committee. There the affiant states that on the basis of his research 16 of those 41 entries are actually registered voters, with only a slight variance in name or address between the records of the Voter Registrar's office and the petition signature. For example, consider the case of Martha Algenita Scott Davis, one of the 41. The Voter Registrar shortened her name to Martha Scott Davis, but she signed the petition as Algenita Scott Davis. Plainly, Fitch may not be penalized for such a variance. Moreover, we are simply not equipped to resolve this sort of factual dispute, and we will not do so in a mandamus action. See Cobra Oil Gas Corp. v. Sadler, 447 S.W.2d 887, 895 (Tex. 1968); Bigham v. Sutton, 565 S.W.2d 561, 562-63 (Tex.Civ.App. — Austin 1978, orig. proceeding); Stevens v. Link, 433 S.W.2d 779, 781 (Tex.Civ.App. — Texarkana 1968, orig. proceeding). Those 16 entries cannot therefore be excluded on the basis of unregistered status. Nevertheless the 25 remaining signatures must be excluded, because the law requires a signer to be registered.

Atkinson's remaining challenges may be taken together. They rest on the following bases:

(a) missing voter registration number (27 entries),

(b) incorrect voter registration number (25),

(c) missing city and missing zip code (9),

(d) missing city only (232),

(e) missing zip code only (11), and

(f) missing date of signature (4).

Although the Election Code nowhere states whether a given defect is fatal, caselaw has filled that gap. At least two types of irregularities have been held not to be fatal: omission of "Texas" as part of the residence address when city, county, and zip code were given; and omission of the year as part of the signing date where day and month were given and other indicia allowed inference of the year. And we held only a few days ago that a signature cannot stand in the face of an attack on any of the first three grounds listed above (missing registration number, incorrect registration number, or missing both city and zip code). Dunn v. Slagle, 783 S.W.2d 953 (Tex.App. — Houston [14th Dist.] 1990, orig. proceeding). There we declined to decide the effect of a missing city alone or a missing zip code alone, saving those questions for another day. Today is not that day. We need do no more to decide this case than follow our prior decision.

Love v. Veselka, 764 S.W.2d 564 (Tex.App. — Houston [1st Dist.] 1988, orig. proceeding); Bacon v. Harris County Rep. Exec. Comm., 743 S.W.2d 369 (Tex.App. — Houston [14th Dist.] 1988, orig. proceeding); Cohen v. Strake, 743 S.W.2d 366 (Tex.App. — Houston [14th Dist.] 1988, orig. proceeding).

Leal v. Mather, 709 S.W.2d 269, 271 (Tex.App. — Houston [14th Dist.] 1986, orig. proceeding).

Our decision in Dunn v. Slagle has become the subject of a counter-mandamus by the real party in interest in that case, Elaine Brady. She has filed an original proceeding in the supreme court seeking to overturn our ruling. Brady v. Fourteenth Court of Appeals.

Applying the rule of Dunn v. Slagle, we hold invalid the signatures supported by either no registration number (27), a wrong number (25), or a missing city and missing zip code (9). A total of 61 entries therefore must fall. Adding to this the 25 names we have already found invalid for lack of registration, we find a grand total of 86 signatures to be deducted from the 783 non-duplicate signatures. Since that group of 697 fails to satisfy the statutory minimum, Respondent Carter should apparently have rejected Fitch's application for a place on the ballot.

III. COUNTERARGUMENTS A. Laches

Our task is not complete, however, as several defensive arguments require attention. First we consider the claim that we should refuse mandamus relief because of laches. It is black-letter law that an unreasonable and prejudicial delay may operate to bar equitable remedies. The question is, how much delay is unreasonable in the context of the election process? Fitch filed her application on Friday, December 29, 1989, shortly before the deadline of Tuesday, January 2, 1990. See TEX.ELEC.CODE ANN. § 172.023(a) (Vernon Supp. 1990) (setting the filing deadline as 6 p.m. Jan. 2). Soon thereafter, her petitions undoubtedly became available for public inspection. Yet relator waited until February 9 to bring this action. To be sure, he needed a certain amount of time to ascertain the validity of those signatures, and the thorough tabulation of names has been enormously helpful to us. The problem is that we cannot wait long to decide this case; primary election day is scheduled for March 13. Moreover, the law establishes a deadline for challenging an application for a place on the ballot, namely the day before the beginning of absentee voting by personal appearance. TEX.ELEC.CODE ANN. § 141.034 (Vernon Supp. 1990). This year the deadline for relief falls on Tuesday, February 20 — the first working day after oral arguments in the present proceeding — which makes a laches argument look rather attractive.

Nevertheless we cannot bring ourselves to dispose of issues as important as these on the basis of laches. Relator could assuredly have acted with more dispatch, but part of the delay is also attributable to the unsuccessful removal attempt, and the real party in interest might have filed her petitions earlier than the last working day before the filing deadline. Verifying signatures is tedious business: the exhibits on file prove that. Relator's computer tabulation must have required considerable time, time which saved us from having to struggle through certain counting exercises unaided. We referred earlier to the affidavit of a Democratic party official who personally investigated the registration status of the 41 voters alleged by relator to be unregistered, 16 of which may very well be registered. That document says the research took a full eight hours. It would be inappropriate to close the courthouse doors simply because gathering evidence takes time. There is virtually no authority for refusing to enforce the Election Code because of laches, and the supreme court has continued to make election rulings even at this late date without a hint of such a conclusion. Last minute mandamus actions may well become an intolerable problem, but they have not quite reached that point yet.

Exacerbating the time pressure problem is the fact that ours is not the only court making election mandamus rulings in areas of unsettled law. On February 14, two days before we heard this case, the supreme court handed down an opinion in Sears v. Bayoud, 786 S.W.2d 248 (Tex. 1990). Sears held a candidate constitutionally ineligible to run for a judicial office, but found further that (for reasons we will not go into) after a certain time a candidate's name cannot be removed from the ballot even though the candidate could not validly win election. As a result, one candidate might have a valid complaint in a mandamus proceeding yet run into certain limitations on the available remedies. Precisely that seems to have happened in Dunn v. Slagle. We were informed on the morning of arguments in this case that the supreme court had issued a stay order against that part of our decision in Dunn v. Slagle which purported to order Elaine Brady removed from the ballot; given the crush of election cases pending before them, however, the justices were apparently in no position to produce an accompanying opinion.

In addition, there is apparently an even more recent ruling from the supreme court. Before discussing that development it is necessary to bring up the second defensive argument, namely the unconstitutionality of requiring Fitch to collect 250 signatures (plus an additional 500 in lieu of a filing fee). In support of this contention two cases are cited: Anderson v. Celebrezze, 460 U.S. 780, 103 S.Ct. 1564, 75 L.Ed.2d 547 (1983), and its leading progeny in the Fifth Circuit, Pilcher v. Rains, 853 F.2d 334 (5th Cir. 1988).

B. Constitutionality

Anderson struck down an Ohio law which would have required John Anderson, an independent candidate for President, to collect 5000 signatures by March 20, whereas the two major parties could choose their candidates at a later date. Basing its decision on the first and fourteenth amendments, the Court devised a balancing test that called for consideration of (1) the character and magnitude of the asserted injury, (2) the state's interests in upholding its rule, and (3) the extent to which those interests necessitate burdening the plaintiff's rights. 460 U.S. at 789, 103 S.Ct. at 1570. That case differed from this one in certain critical respects. One was the law's limitation on "political participation by an identifiable group whose members share a particular viewpoint, associational preference, or economic status." Id. at 792-93, 103 S.Ct. at 1571-72. No such limitation operates here. Instead of impairing ballot access by a party or identifiable group, the Texas law at issue makes its demands only of the person who seeks to run for office. Another feature recognized by the Court was the extraordinary nature of the Presidential election process, "a uniquely important national interest." Id. at 794-95, 103 S.Ct. at 1572-73.

The Pilcher decision also prohibited the denial of ballot access to a minority political party. Relying on Anderson, the Fifth Circuit upheld an injunction against enforcement of the Texas statute which required voter registration numbers on the party's petitions. That law would have called for the Libertarian Party to collect over 34,000 signatures, including voter registration numbers, by signers who had not voted in the same year's Republican or Democratic primaries. But the Pilcher rationale rested on the inclusion of residence address along with the signature. Thus, any signers who happened to have similar names "would be sufficiently distinguished by address." 853 F.2d at 337. The attack on requiring residence address is accordingly unsupported by Pilcher, because the case affirmatively relies on the inclusion of that information. Furthermore, the Libertarian Party proved at trial "that the voter registration number requirement imposed a serious burden," id. at 335, because

That figure is based on a requirement of 1% of the total vote for governor in the last election. TEX.ELEC.CODE ANN. s 181.006(b)(2) (Vernon 1986).

(1) less than two percent of petition signers know their voter registration number by heart or carry their voter registration card with them, so as to be able to give the number at the time the petition is signed; (2) petition collectors, as a result, spend from fifty to seventy percent of the petition drive looking up voter registration numbers on county records and adding them to petitions; and (3) no other state requires voter registration numbers on minor party petitions.

Id. at 336. Not one of these factors is present here. Nothing has been proven at trial or elsewhere. Nor is it conceivable that only two percent of petition signers would be unable to remember their residence address: writing one's address is hardly a herculean task. Accordingly, respondent cannot establish what Anderson requires, and what Pilcher found, namely a "serious burden."

Moreover, it is by no means clear that Pilcher reflects a correct — or at least complete — understanding of constitutional law. The supreme court has in fact endorsed petition requirements precisely because they allow fairer access to the ballot than do filing fees, which unduly burden the poor:

[T]here are obvious and well-known means of testing the "seriousness" of a candidate which do not measure the probability of attracting significant voter support solely by the neutral fact of paying a filing fee. States may, for example, impose on minor political parties the precondition of demonstrating the existence of some reasonable quantum of voter support by requiring such parties to file petitions for a place on the ballot signed by a percentage of those who voted in a prior election. Similarly, a candidate who establishes that he cannot pay the filing fee required for a place on the primary ballot may be required to demonstrate the "seriousness" of his candidacy by persuading a substantial number of voters to sign a petition in his behalf.

Lubin v. Panish, 415 U.S. 709, 718, 94 S.Ct. 1315, 1321, 39 L.Ed.2d 702 (1974) (citation and footnote omitted, emphasis added); see also American Party of Texas v. White, 415 U.S. 767, 94 S.Ct. 1296, 39 L.Ed.2d 744 (1974) (upholding a 1% support requirement for minor political parties); Jenness v. Fortson, 403 U.S. 431, 91 S.Ct. 1970, 29 L.Ed.2d 554 (1971) (upholding a 5% support requirement). Against this background the Court could say in 1986:

Jenness and American Party establish with unmistakable clarity that States have an "undoubted right to require candidates to make a preliminary showing of substantial support in order to qualify for a place on the ballot. . . " Anderson v. Celebrezze, 460 U.S. 780, 788-89, n. 9, 103 S.Ct. 1564, 1569-70, n. 9, 75 L.Ed.2d 547 (1983). We reaffirm that principle today.

. . . .

To require States to prove actual voter confusion, ballot overcrowding, or the presence of frivolous candidacies as a predicate to the imposition of reasonable ballot access restrictions would invariably lead to endless court battle over the sufficiency of the "evidence" marshaled by a State to prove the predicate. Such a requirement would necessitate that a State's political system sustain some level of damage before the legislature could take corrective action. Legislatures, we think, should be permitted to respond to potential deficiencies in the electoral process with foresight rather than reactively, provided that the response is reasonable and does not significantly impinge on constitutionally-protected rights.

Munro v. Socialist Workers Party, 479 U.S. 189, 107 S.Ct. 533, 537-38, 93 L.Ed.2d 499 (1986). Since Munro allowed a state to demand a 1% showing of support, it is inconceivable that the constitution forbids requiring only 250 signatures to be drawn from a county with millions of citizens. We therefore reject the constitutional challenge made today.

We note, but do not rely on, the recent decision in Shipley v. Harris County Democratic Executive Committee and Carlos L. Correa, No. 01-90-122-CV, 1990 WL 19197 (Tex.App. — Houston [1st Dist] Feb. 16, 1990, orig. proceeding). That court found a candidate had failed to submit 250 valid signatures. A majority of the panel rejected the candidate's constitutional attack. A dissenting justice would have found § 172.021(e) unconstitutional because it treats large counties differently from all other counties. Id. (O'Connor, J., dissenting). Yet her conclusion hardly seems to follow from the body of her dissent, which chronicles the well-known problem of what she calls "frivolous filers," where questionable candidates would enter the process in Dallas and Harris Counties just before 6:00 p.m. on the date of the filing deadline. The fate of those candidates then rose or fell with the success of their party in that year's elections.
This factual predicate — which the Supreme Court emphatically said in Munro was not a prerequisite for setting up reasonable restrictions on ballot access — seemingly lends dispositive support to the Texas scheme. So thought the majority in Shipley. But the Texas supreme court has overturned that decision, without opinion. Possible reasons for reversal might include the propriety of relief granted (as in Sears v. Bayoud and Dunn v. Slagle ), the actual number of valid signatures, or the constitutional permissibility of demanding 250 signatures. The only hint that any member of this court would credit a constitutional challenge to the signature scheme came in a separate opinion in Leal v. Mather, 709 S.W.2d 269 (Tex.App. — Houston [14th Dist.] 1986, orig. proceeding) (Ellis, J. concurring). That judge's view related only to vagueness, and he obviously abandoned his position two years later when he would have barred a candidate from running, on the ground that some signatures failed to contain "Texas" as part of the signer's residence address. Cohen v. Strake, 743 S.W.2d 366 (Tex.App. — Houston [14th Dist.] 1988, orig. proceeding) (Ellis, J. dissenting).

All that remains is the issue of relief. When we granted the writ (Tuesday, Feb. 20), we did so with language tracking the majority opinion's final paragraph in Shipley, with explanation that our opinion would follow. Because there has been no supervening opinion by the supreme court, we adhere to that formula. Respondent is ordered to reject Fitch's application, to deliver to her written notice of the reason for rejection, and to certify in writing a declaration of ineligibility to the canvassing authority for the election. The real party in interest has filed motion asking us to withdraw our order granting relief; that motion is denied.


Summaries of

Atkinson v. Carter

Court of Appeals of Texas, Houston, Fourteenth District
Mar 1, 1990
785 S.W.2d 449 (Tex. App. 1990)
Case details for

Atkinson v. Carter

Case Details

Full title:Mark D. ATKINSON, Relator, v. Jack CARTER, Chairman of the Harris County…

Court:Court of Appeals of Texas, Houston, Fourteenth District

Date published: Mar 1, 1990

Citations

785 S.W.2d 449 (Tex. App. 1990)

Citing Cases

Scott v. Rees

Under such circumstances the court is necessarily without jurisdiction to do anything but dismiss the cause.…

Carter v. Fourteenth Court of Appeals

They further contended that the construction and application of the Election Code provisions urged by…