Grimes v. Smith, 585 F. Supp. 1084 (N.D. Ind. 1984)

US District Court for the Northern District of Indiana - 585 F. Supp. 1084 (N.D. Ind. 1984)
May 7, 1984

585 F. Supp. 1084 (1984)

Douglas M. GRIMES, Eleanor B. Jackson, and Cleo L. Cusic, Plaintiffs,
William (Bill) SMITH, Jr., John A. Grimes, Charles H. Graddick, Robert (Bob) Lacy, and Michael Brown, Defendants.

Civ. No. H83-383.

United States District Court, N.D. Indiana, Hammond Division.

April 25, 1984.

As Amended May 7, 1984.

*1085 Jerald S. Meacham, Gary, Ind., for plaintiffs.

Frederick T. Work, Gary, Ind., for Graddick.

Terry C. Gray, Gary, Ind., for Grimes.

J. Michael Katz, Merrillville, Ind., for Smith and Lacy.

David Capp, Merrillville, Ind., for Brown.


POSNER, Circuit Judge, Sitting by Designation.

I was designated to conduct the trial in this suit brought by Douglas Grimes, a defeated candidate for the Democratic nomination for Judge of the City Court of Gary, Indiana, and two voters in the primary against Charles Graddick and others, alleging that by putting up a same-name candidate (defendant John A. Grimes) in another race the defendants conspired to defraud the plaintiffs of their constitutional rights to participate in electoral politics, in violation of 42 U.S.C. §§ 1983 and 1985. The suit seeks damages and equitable relief. A six-day jury trial, concluded on March 19, 1984, resulted in verdicts totaling $100,004 in favor of the plaintiffs and against all of the remaining defendants but Graddick. I had directed a verdict for another defendant, Robert Deloney, at the end of the plaintiff's case, because there was no evidence connecting him to the conspiracy, at the same time dismissing the section 1983 charge because he was the *1086 only public-employee defendant remaining in the case. The Lake County Election Board and its members had been dismissed earlier. I now have before me the defendants' Rule 59(a) motions for new trial (based on alleged errors in the instructions, and on the size of the damage award), Rule 50(b) motions for judgment notwithstanding the verdict, and Rule 60(b) motions for relief from judgment; the plaintiffs' motion that I enjoin Judge Graddick from continuing in office; and Judge Graddick's motion for an award of attorney's fees under 42 U.S.C. § 1988.

In Gary, winning the Democratic primary is tantamount to winning the general election. Last year the Democratic primary was held on May 3. The major offices at stake were Mayor, City Clerk, and City Judge. The favorites in these races, all incumbents, were Richard Hatcher, for Mayor; Barbara Wesson, for City Clerk; and Douglas Grimes, for City Judge. Mayor Hatcher is the long-time Mayor of Gary, and the ticket of Hatcher, Wesson, and Grimes was the Administration ticket. But there were other candidates in each of these races. Entering the voting booth, a voter interested in just these three offices would have seen, reading from left to right on the voting machine, a row with 10 names in it, in groups of three, four, and three, each group being under the name of an office (in order, Mayor, City Clerk, and City Judge). To the right of all these names were the names of candidates in aldermanic races. The first three names on the ballotThomas Crump, Robert Gordon, and Richard Gordon Hatcherwere the candidates for Mayor. The next four Kenneth S. Bass, John Damian, John A. Grimes, and Barbara Leek Wessonwere the candidates for City Clerk. And the next threeLloyd Buford Fisher, Charles H. Graddick, and Douglas H. Grimes were the candidates for City Judge. In the Mayor's race the "serious" candidates were Crump and Hatcher. They got 23,150 and 27,835 votes respectively, and the third candidate got 1,217. In the City Clerk's race Miss Wesson was the only serious candidate, and she got 26,069 votes. Bass and Damian got 3,343 and 4,192 votes respectively, and John Grimes a surprising 8,957. In the City Judge's race Douglas Grimes and Graddick were the serious candidates, and they got 19,373 and 22,174 votes respectively, and the third candidate 3,285. Graddick went on to win the general election (as did the other Democratic nominees), and he took office on January 1, 1984. Douglas Grimes brought this lawsuit shortly after the primary election and asked for a preliminary injunction to prevent Graddick from running in the general election as the Democratic nominee for City Judge; but this was denied and the matter was set for trial on the merits.

Although John Grimes did not win the City Clerk's race, the fact that this political unknown who had done no significant campaigning got almost 9,000 votes was surprising. And since he had the same last name as Douglas Grimes, and Douglas Grimes, also contrary to expectations, lost the election for City Judge, Douglas Grimes suspected foul play and began to investigate. The investigation revealed that John Grimes (who is now 43 years old) had been at the time of the campaign and election an unemployed carpenter, almost blind as a result of glaucoma, with little education, no steady source of income (though he picked up a few dollars a week doing handyman work for his landlord, and now receives social security disability benefits), and no previous involvement in politics. He had never in his life voted in a primary or general election, and did not vote in the March 1983 primary in which he was a candidate. He has never had the faintest idea of the duties of a City Clerk. It is evident, and no one denies, that John Grimes did not stand for election to this post on his own initiative. He was put up by somebody; the question is by whom.

Defendant Michael Brown, an unemployed 27-year-old, testified that he decided to get into politics by managing someone's campaign. He knew John Grimes, though not well; and he testified that he thought that Grimes would be a good candidate for City Clerk in the May 1983 primary. He *1087 was unable to give any explanation of why he thought so, beyond remarking that he didn't think Grimes' handicap should be disqualifying. The deadline for filing as a candidate for the May primary was noon on March 4. Brown testified that he borrowed a car and drove John Grimes to the place for filing. He testified that he knew that a candidate had to be a registered voter, and learned somehow that Grimes was not registered, so the first thing they did when they arrived was to have Grimes register. He did so and then signed the filing form, which was time-stamped 11:59 a.m. on March 4.

Grimes and Brown then went their separate ways. Both testified that they campaigned for Grimes' election, but admitted that their campaigning was limited to sporadic personal canvassing in their respective neighborhoods. Brown's campaigning did not go much beyond buying an occasional drink for an acquaintance in the bars that he frequented, and urging them to vote for Grimes. There were no campaign rallies or speeches, campaign leaflets, or billboards or other advertising; there were no campaign workers. The campaign expenditure form that Grimes filed after the election disclosed expenditures of zero.

John Grimes and Michael Brown were connected to the other defendantswho are, besides Judge Graddick, his campaign manager Bill Smith and his campaign coordinator Bob Lacyin several ways. Grimes and Smith have known each other for many years. Grimes has a daughter by Smith's sister, and Grimes and Smith regard each other like the brothers-in-law they would be if Grimes had married Smith's sister. Grimes also has known Lacy and Lacy's wife well for many years. Brown is a friend of Smitha particularly good friendand of Lacy, and he worked under Lacy in the Graddick campaign, in security, until he began managing John Grimes' "campaign"; and there was testimony that he continued working in the Graddick campaign for a short time after he became Grimes' campaign manager. Grimes encountered both Smith and Lacy several times during the campaign, and afterward Grimes asked Smith to give him a ride to Grimes' lawyer's office so that he could get help with his campaign expenditure form, and Smith obliged him.

As part of his investigation of the election outcome, Douglas Grimes went looking for John Grimes to talk to him about the results, and one of John Grimes' sisters, Betty Grimes, according to her testimony, called Bob Lacy to find out what she should do. When she looked in the phone book (apparently referring to a personal book of phone numbers rather than a telephone directory), Bob Lacy's name "just fell out," and she called him. He denied that she had called him but the jury was entitled to disbelieve his denial, especially since Betty Grimes was a friendly witness to her brother and the other defendants. Also after the election, a journalist tried to interview John Grimes (the "mystery Grimes" as he was then called, because he had done so well in the election yet no one had heard of him) in a parking lot. The journalist testified that when he insinuated laughingly that Grimes had been put up to run for City Clerk in order to help Graddick's campaign against Douglas Grimes, Brown, after a moment's hesitation, joined in the laughter. Finally, John Grimes' bank book, which was put into evidence, showed two deposits totaling $700 in mid-March, which was six weeks before he began receiving social security disability benefits. He was unable to account convincingly for these receipts. In fact, he denied having received this money, despite what the bank book (which he acknowledged was his) showed.

Although there was no direct evidence that Graddick's campaign directors, Smith and Lacy, put up Grimes to run for City Clerk, the circumstantial evidence reviewed above provided an adequate basis for the jury to infer that they did soboth of them; for the jury was entitled to find that Smith and Lacy, in view of their acquaintance with Grimes, must have discussed the matter with each other. The jury was also justified in finding that Graddick had not joined the conspiracy. Although Smith *1088 and Lacy testified that they discussed all strategy matters with Graddick, the jury was not required to believe that, or indeed any, part of their testimony. Graddick testified in his own behalf and the jury may have believed that he did not know about the scheme, and was not the kind of man to engage in such tricks. Whether the jury was right or wrong, there was no inconsistency in its exonerating Graddick but finding that the other four defendants were participants in the conspiracy.

As to the effects of the conspiracy: lay and expert testimony established a reasonable probability that many people, enough to swing the election to Graddick, voted for John Grimes (whose name appeared before Douglas Grimes' on the ballot, reading from left to right) thinking they were voting for Douglas Grimes. Douglas Grimes' co-plaintiffs so testified, as did another voter in the May primary. And assisted by testimony of the plaintiffs' well-qualified expert witness, Dr. Gerald Wright, a political scientist who teaches at the University of Indiana at Bloomington, the jury was entitled to conclude that the confused voters who testified were typical of many Gary voters. Granted, the evidence regarding the effects of the conspiracy was not entirely satisfactory. The plaintiffs actually presented two inconsistent theories of how the voters were confused. The lay witnesses testified that they voted for John thinking he was Douglas, and did not vote in the City Judge's race at all; the expert speculated that voters who voted for John thinking he was Douglas went on and voted for Graddick in the City Judge's race because they recognized his name. The lay witnesses' theory seems the more plausible but is inconsistent with the heavy turnout in the City Judge's race. Their testimony would lead one to expect that the turnout should have been very light because thousands did not vote in the race at all, thinking that they had already voted in that race when they pulled the lever for John Grimes. But no plausible alternative to confusion was advanced that could explain how John Grimes, without campaigning, managed to get several thousand votes more than the other token candidates in the Clerk's raceboth of whom did more campaigning than John Grimes.

Thus, I do not think that the jury was unreasonable in finding that the defendants conspired to mislead voters in the May primary and that they succeeded to the extent of probably affecting the outcome of the City Judge's race; and I therefore deny the defendants' motions for judgment notwithstanding the verdict insofar as those motions are based on an alleged insufficiency of evidence. But I must also consider their argument that the facts do not make out a cause of action under 42 U.S.C. § 1985(3), which after I dismissed the section 1983 claim was the only basis on which the plaintiffs could get a judgment.

This section, which originated in the Ku Klux Klan Act of 1871, punishes, so far as is relevant to this case, conspiracies to deprive, "either directly or indirectly, any person or class of persons of the equal protection of the laws, or of equal privileges and immunities under the laws...." Until a few months ago, an insuperable obstacle to the maintenance of this action under section 1985(3) would have been the absence of any involvement of state officers in the conspiracy. Our court of appeals had taken the position that if the right of which the plaintiff is deprived is one created by the Fourteenth Amendment, which limits only state action, the plaintiff cannot prevail under section 1985(3) unless he proves that the conspiracy involved state officers. See Williams v. St. Joseph Hospital, 629 F.2d 448, 451-52 (7th Cir. 1980), and cases cited there. Any right to vote in a state or local election, or to run for state or local office, must be based on either the due process or the equal protection clause of the Fourteenth Amendment, at least in the absence of racial discrimination (not present in this case, as we shall see), which might bring the right under the Thirteenth Amendment. But last July the Supreme Court, in United Brotherhood of Carpenters & Joiners v. Scott, ___ U.S. *1089 ___, 103 S. Ct. 3352, 3358, 77 L. Ed. 2d 1049 (1983), made clear that the burden on the plaintiff in a section 1985(3) case based on the deprivation of a right under the Fourteenth Amendment is to "prove that the state was somehow involved in or affected by the conspiracy." (Emphasis added.) The State of Indiana was affected by the conspiracy in the present case; the conspiracy caused (at least, probably caused) the election of a public official who was not the voters' real choice.

A more serious obstacle to the cause of action here is the Court's discussion in Scott of whether "Congress meant to forbid wholly nonracial, but politically motivated conspiracies...." Id. 103 S. Ct. at 3359. "[W]e find difficult the question whether § 1985(3) provided a remedy for every concerted effort by one political group to nullify the influence of or do other injury to a competing group by use of otherwise unlawful means. To accede to that view would go far toward making the federal courts, by virtue of § 1985(3), the monitors of campaign tactics in both state and federal elections, a role that the courts should not be quick to assume. If respondents' submission were accepted, the proscription of § 1985(3) would arguably reach the claim that a political party has interfered with the freedom of speech of another political party by encouraging the heckling of its rival's speakers and the disruption of the rival's meetings." Id. at 3359-60. The Court went on to acknowledge that there was some support in the legislative history of the Ku Klux Klan Act for a broad view of section 1985(3), and it stopped short of holding that the section cannot reach political conspiracies. See id. at 3360. Moreover, any such "holding" would really have been dictum, because the case did not involve a political conspiracy, but rather a conspiracy by union workers to drive nonunion workers out of town. Nevertheless, the language quoted above is a good predictor of how the Court probably would rule if presented with the question whether this case is actionable under section 1985(3); and writing as a district judge, as I am doing in this case, I am more than willing to take my cue from the Court's considered dictum, especially in the absence of any authority in our court of appeals for the existence of a cause of action on the facts of this case. There is some court of appeals authority, though it is pre-Scott, in other circuits. See Means v. Wilson, 522 F.2d 833, 837-40 (8th Cir. 1975); Cameron v. Brock, 473 F.2d 608 (6th Cir. 1973). In those cases public officers were members of the conspiracies; but, as I have noted, this is inessential under Scott.

It is true that Smith v. Cherry, 489 F.2d 1098 (7th Cir.1973) (per curiam), found a cause of action on facts rather similar to those of this case. But Cherry was not a 1985(3) case, it was a 1983 case, so it was necessary only to find that the Fourteenth Amendment creates, as against the states, rights to vote, and to stand for office, that can be taken away by election fraud. (On the contours of a section 1983 election fraud case see also Griffin v. Burns, 570 F.2d 1065, 1076-79 (1st Cir.1978).) One of the defendants in Cherry was an incumbent state senator, which created the necessary state action. The court did say, in a footnote, that the case had been brought under section 1985 as well as section 1983. See 489 F.2d at 1100 n. 1. But there was no further mention of section 1985, and none at all of subsection 3; and the opinion twice mentions section 1983 as the statutory basis of the suit. See 489 F.2d at 1100, 1102. Cherry is not a section 1985(3) precedent.

As none of the defendants who remained in the present case when it went to the jury held public office at the time of the alleged conspiracy, this is a 1985(3) case or it is nothing (there is no pendent state-law claim). And far from having ever taken an expansive view of section 1985(3), the Seventh Circuit, in Williams v. St. Joseph Hospital, supra, and the earlier cases cited in Williams, such as Cohen v. Illinois Institute of Technology, 524 F.2d 818 (7th Cir.1975), was a pioneer in the process of contraction that has culminated in the Supreme Court's decision in Scott. I would *1090 be most surprised therefore if our court of appeals disregarded the broad hint in Scott that political conspiracies as such are not actionable under section 1985(3). It goes without saying that there is no racial or similar hostility behind the conspiracy in the present case that would take it out of the category of purely political conspiracies. Both Douglas Grimes and Charles Graddick, and also their respective supporters, are black, which is not significant in itself but only as showing that they are of the same race. A conclusion that there is no federal remedy for purely private, purely political conspiracies need not leave the victims of such conspiracies totally without remedy. There are a variety of state law remedies for election fraud, see Developments in the LawElections, 88 Harv.L. Rev. 1111, 1294-1331 (1975), but there is, as I have said, no pendent state-law count in the complaint. I do not know whether the plaintiffs had or have any judicial or administrative remedies under Indiana law in respect of the fraud.

As an original matter, there is a strong argument from the text of 1985(3) against recognizing a cause of action on the facts of this case. The emphasis in the section on deprivation of equal protection or equal privileges or immunities suggests that the victims of the election fraud must be some identifiable group, such as blacks; the supporters of an incumbent official of the same race as the challenger seems not to be that kind of identifiable group. The Supreme Court in Griffin v. Breckenridge, 403 U.S. 88, 102, 91 S. Ct. 1790, 1798, 29 L. Ed. 2d 338 (1971), while reserving the question whether the statute is limited to racial discrimination, interpreted the repeated use of the word "equal" in the statute to mean "that there must be some racial, or perhaps otherwise class-based, invidiously discriminatory animus behind the conspirators' action." Such animus is not present in a garden-variety election fraud of the sort involved in this case, where the only object of the challenger whose supporters engaged in fraud was to get into office, and not to oppress some identifiable group in the community.

There are other arguments against recognizing a cause of action in this case. They have a different domain from the argument that the victims of purely political conspiracies are not a class protected by section 1985(3). That argument would put all purely political conspiracies beyond the reach of section 1985(3), unless the conspirators included a state officer or officers, but it would be irrelevant to a conspiracy that was actionable under section 1983 because state officers were involved. The arguments I am about to consider are arguments about why the particular type of political conspiracy alleged and proved in this case should not be a federal tort, under any statute, but they do not exclude the possibility that other types of political conspiracy might be actionable. The first of these arguments is that the particular kind of fraud involved in this case is easily prevented by voters' taking elementary precautions; just reading the ballot with a modicum of care would have dispelled any impression that John Grimes was a candidate for City Judge. But this argument cuts against the grain of the common law concept of intentional torts in general and fraud in particular, which does not allow a defense of contributory negligence. See, e.g., Cenco Inc. v. Seidman & Seidman, 686 F.2d 449, 454 (7th Cir.1982). It may be cheap for victims to avoid being defrauded, but it is cheaper for the defrauders to avoid defrauding; indeed, it costs them less than nothing, since the fraud itself is costly to arrange and carry out. Besides, I am not so sure that it is cheap for all voters to avoid being defrauded. Now that voter qualifications are minimal, realism requires one to recognize that many people who vote are not well educated or well informed. The cost to them of obtaining the information they need in order to vote intelligently may be very high, and therefore their right to vote might be seriously impaired if they had no legal protection against attempts to mislead them, even though more alert and informed voters would not be misled.

*1091 Second, to recognize a right of action on the facts of this case would interfere with free competition in the political arena, and this in two ways. First, the issue of causation is bound to bring the merits of the election race into the courtroom, as indeed happened in the trial of this case. In order to show that the alleged conspiracy had not changed the outcome of the election, the defendants put in evidence (over the plaintiffs' vigorous but unavailing objections) that the people of Gary were getting tired of Mayor Hatcher and his minions, that Douglas Grimes had as City Judge offended the electorate by actions he took with regard to gun control and by his investment in a company that had a contract with the city, and that Charles Graddick received strong support as an anti-machine, and home-grown, candidate. This testimony was relevant and probative, but it meant that the local election race was being refought in the federal courthouse. Second, to make the putting up of a sham candidate actionable as fraud is to erect a barrier of some potential significance to becoming a candidate for public office. Even if the constitutional tort alleged in this case could be confined (as I think it could be, cf. Rudisill v. Flynn, 619 F.2d 692, 695 (7th Cir.1980)) to cases where the voters were confused by a candidate's name, as distinct from his campaign claims, this would still create a barrier to any candidate for public office who had the same name as or a similar name to that of another candidate not necessarily for the same office but running in the same election. True, it is no defense to an action for trademark infringement that the infringer is using his own name (which just happens to be, let us suppose, Calvin Klein); but he can always do business under another name. If your name happens to be Grimes, you can hardly run for office under a different name. If the tort alleged in this case were actionable, then any time someone named Grimes ran for public office in an election where another candidate was named Grimes he would run the risk of being accused of having conspired with the campaign staff of the opponent of the other Grimes to swing the election to that opponent.

But this kind of riskthe risk of legal erroris endemic in our, or in any, legal system; and a similar risk did not deter our court of appeals in Smith v. Cherry, supra, from holding that an election fraud was actionable, albeit under section 1983 rather than 1985(3). The fraud in Cherry was that the candidate who won the primary election had run in it never intending to serve; by agreement with his fellow defendants, he stepped aside before the general election and one of the defendants was appointed, by a party committee controlled by the others, to take his place. (The main plaintiff in the action was the candidate who had run against the stand-in in the primary, and had lost.) Cherry countenances a risk similar to that in this case of inhibiting candidature; a candidate who wins a primary and for any reason steps aside becomes vulnerable to a charge that he conspired with his replacement. To make any campaign conduct (other, perhaps, than physically stealing votes) actionable creates a risk of discouraging lawful activity, which can be mistaken as being conspiratorial and fraudulent; but the risk was deemed acceptable in Cherry. True, the risk of discouraging same-name candidacies that would be created by recognizing a cause of action in the present case seems greater than the corresponding risk in Cherry. But there is a pretty good method of preventing liability: do some campaigning. If John Grimes had waged any sort of campaign, not only would it be difficult to regard him as a sham candidate (and I charged the jury that they must find that he was a sham candidate to find any of the defendants liable), but many fewer voters would have been confused. A big source of the confusion here was that the voting public did not know there were two Grimes running in the May primary, and John Grimes would have dispelled that confusion automatically by campaigning. Thus the only, or at least the principal, same-name or similar-name candidates that would be discouraged by a tort of conspiring to put up a sham candidate would be *1092 candidates who did not plan to do any campaigning. Their discouragement may be an acceptable price (as they are, virtually by definition, marginal candidates) to pay for giving legal remedies to voters who are deprived of their right to vote in a practical sense by conspiracies to run sham candidates. But subsequent cases in this circuit, notably Rudisill v. Flynn, supra, 619 F.2d at 694, have read Cherry very narrowly, and maybe therefore our court of appeals would draw a line between the facts of Cherry and the facts of this case.

However all the competing considerations balance out, I consider myself bound by my sense of the direction of the law in this area to hold that the complaint did not state a cause of action and must therefore be dismissed. But since the plaintiffs are very likely to appeal my decision, I shall go on and discuss the other issues raised in the various post-trial motions, none of which I think has merit, so that if the court of appeals disagrees with me and holds that the plaintiffs did state a cause of action, it can (if it agrees with my other rulings) reinstate the jury's verdict and so terminate this litigation without necessity for a retrial.

The first of these issues is whether (assuming as I shall from here on out that the plaintiffs did state a cause of action) I should order a new trial because of errors in the instructions. Some of the defendants' objections go rather to the existence of a cause of action than to the soundness of the instructions, assuming such a cause of action exists; and I need not discuss these. I reject the suggestion that the instruction, "You must find that the conspiracy probably changed the outcome of the election," was erroneous because the plaintiffs (in the defendants' words) "were required to show that the conspiracy did, in fact, change the outcome of the election." That is too high a standard; once the jury finds that there was an unlawful conspiracy, it is entitled to resolve doubtful questions of causation against the conspirators. This is a very generally followed principle of tort law, see, e.g., Sindell v. Abbott Laboratories, 26 Cal. 3d 588, 163 Cal. Rptr. 132, 607 P.2d 924 (1980); Summers v. Tice, 33 Cal. 2d 80, 199 P.2d 1 (1948), and I see no reason why it should not be followed in a constitutional-tort case as well. I also reject the suggestion, which the defendants make in their motion under Rule 60(b), that the phraseology of the instruction ("You must find ...") in effect directed a verdict for the plaintiffs. In context, it was clear that the "You must find" locution referred to what the jury must find in order to bring in a verdict for the plaintiffs; causation was one of the elements of the tort.

The defendants argue that no instruction should have been given on punitive damages because the plaintiffs did not include a punitive-damages instruction in the proposed instructions that they submitted with their pretrial brief. This is true, but the defendants' equities in objecting are weak, since contrary to my express request at the pretrial conference they failed to submit any proposed instructions until the trial was well under way. In any event, since the plaintiffs from the beginning charged the defendants with willful misconduct, it was predictable that they would ask for an instruction on punitive damages; the defendants could not have been surprised. The defendants also object that I "failed to inform the jury that a compensatory damage award must be based upon actual, proven damages." This objection is groundless. I instructed the jury not only that, "In assessing damages, you must choose figures that will reasonably compensate each plaintiff for the actual injury he has suffered as a result of the defendants' violations of his constitutional rights," but also that "You must not base an award of damages on mere speculation or conjecture, but if you find that a plaintiff's rights were violated and you cannot determine the amount of damages, you may award, as actual damages, nominal damages of $1." In fact the jury awarded $1 in compensatory damages (and $1 in punitive damages) to each of the voter plaintiffs.

*1093 However, the jury awarded $50,000 in actual and $50,000 in punitive damages to Douglas Grimes, and the defendants complain that, quite apart from whether the damage instructions were correct, these amounts are too much, and require a new trial. I disagree, bearing in mind the limits of a trial judge's power to set aside a verdict because the damages are excessive, on which see, e.g., Abernathy v. Superior Hardwoods, Inc., 704 F.2d 963, 971-72 (7th Cir.1983); Crane v. Consolidated Rail Corp., 731 F.2d 1042, 1050-51 (2d Cir.1984). It is true that I instructed the jury not to consider, in assessing plaintiff Grimes' damages, any money he may have spent on his campaign or may have lost as a result of losing his job. There was no evidence that any of the campaign expenditures on his behalf came out of his own pocket, and when he left his post as City Judge he was immediately hired as Corporation Counsel of Gary at a higher salary. Although both of these are part-time jobs, and it is possible that being a City Judge has a greater spillover effect on one's private practice than being Corporation Counsel, no evidence to this effect was introduced.

But I instructed the jury that "you may consider the intangible injury to a person who is wrongfully deprived of a position of public prominence and trust." This is an allowable item of actual damages and $50,000 was a reasonable effort to put a price tag on it. Although the availability of general damages in constitutional-tort cases involving the deprivation of a substantive right was said in Owen v. Lash, 682 F.2d 648, 657-59 (7th Cir.1982), to be an unsettled question in this circuit, and the view in most of the circuits that have addressed the question is that they are recoverable, see id. at 658-59, a subsequent decision of this court, Crawford v. Garnier, 719 F.2d 1317, 1324-25 (7th Cir.1983), indicates that they are not. Although as an original matter I would incline to the view that they are recoverable in an election case such as this, as otherwise the tort remedy would have very little valuethe benefits from voting are not pecuniary, and many people drawn to public office are not drawn by hope of pecuniary gain but by the more intangible rewards of public serviceI am bound by Crawford to hold that they are not recoverable. But Crawford makes clear that damages are available for humiliation, see 719 F.2d at 1324; cf. Kincaid v. Rusk, 670 F.2d 737, 746 (7th Cir.1982), and I am inclined to view damages for the intangible injury from loss of an office of public prominence to be sufficiently akin to damages for humiliation to be recoverable in this action. Obviously there is an inherent arbitrariness in putting a price on such a loss, when it has no financial consequences. But as there is only one City Judge in Gary, the jury could reasonably find that the job was a source of prominence, prestige, and self-esteem to the lawyer that held it and therefore that its loss was a substantial blow to Douglas Grimes. Although no evidence was introduced on Grimes' combined income from his public jobs and his private practice, it is unlikely that the award of compensatory damages exceeded a year's income for him. Such an award is not so excessive as to warrant setting aside the jury's verdict.

I also do not think that $50,000 was an excessive amount of punitive damages. The misconduct of the defendants, if actionable at all, was serious, and $50,000 is not so large a penalty that I can infer that the jury was carried away by emotion in awarding it. As a matter of fact I was favorably impressed by the jury. The jurors were attentive during the trial and deliberated for six hours before bringing in their verdicts.

The defendants object to a few of the procedural instructions, but these objections have no merit at all and I move on to the plaintiffs' motion that (always assuming the judgment for the plaintiffs stands) I enjoin Judge Graddick from continuing to serve in his office. Although the jury exonerated Judge Graddick and I am bound by its verdict in his favor, which the plaintiffs do not challenge, I have the power, sitting as a court of equity, to enjoin Judge Graddick from continuing in office. Because of the way I instructed the jury, its verdicts against the other defendants must be assumed to be based on a finding *1094 that the fraud in which they engaged probably changed the outcome of the election; in other words, that but for the fraud Judge Graddick would probably have lost the election. If, therefore, he probably owes his election to fraud, and to a fraud carried out by the two senior officials of his campaign committee (Smith and Lacy), a court of equity can require him to disgorge the fruits of the fraudand thus give up his office. Cf. 3 Pomeroy, A Treatise on Equity Jurisprudence § 918 (Symons ed. 1941).

But assuming therefore that I do have the power to put Judge Graddick out of office until a new election is held, I would not exercise it on the facts of this case even if I thought that the other defendants had violated federal law (if they did not, I have no power to act against Graddick, in view of the absence of any state law-claim in this case). Considerations of comity and federalism should make a federal judge most reluctant to eject state or local judges from office, especially when the judge sought to be ejected has been found innocent of any personal wrongdoing and his ejection is sought merely as an incident to proceedings against the actual wrongdoers. The jury awarded the principal plaintiff a substantial amount of damages, and, more important, vindicated his contention that the election was "stolen" from him by Judge Graddick or Judge Graddick's staff. This puts Douglas Grimes in a good position to campaign for the office when the next election is held. As a matter of fact he might be hurt rather than helped by my ordering Graddick to vacate his judgeship. I could not order the City of Gary, which is not a party to this suit, to reinstate Grimes. I could only declare the office vacant. The governor has the appointing power in the event of a vacancy, see Ind.Code § 3-2-10-5.5(a), and if someone other than Grimes were appointed he might be a more formidable candidate in the next election than Graddick, since he would not be tainted by the scandal created by the misconduct of Graddick's staff.

The last matter before me is Judge Graddick's request for an award of attorney's fees under 42 U.S.C. § 1988. Such an award would be proper only if the plaintiffs' inclusion of him as a defendant in this action had been frivolous. See Mary Beth G. v. City of Chicago, 723 F.2d 1263, 1277-78 (7th Cir.1983). It was not. The jury did not find, but could reasonably have found, that Judge Graddick was aware of and participated in the conspiracy of his principal campaign aides. I do not mean that he was; but the possibility that a jury would find that he was is not so remote as to entitle him to an award of attorney's fees under the grudging standard that the courts follow when the defendant in a civil rights suit is requesting such an award. And although I have found that the plaintiffs never had a cause of action against Graddick, this does not make their case against him frivolous; the question whether they have a cause of action is a difficult one.

To summarize, all of the motions pending before me are denied except the defendants' motions for judgment notwithstanding the verdict. Those motions are granted and the suit is dismissed with prejudice. The clerk is directed to enter judgment accordingly.