The Illinois Racing Board promulgated a rule that requires jockeys and other participants in horse races in Illinois to submit to random drug testing not founded on any suspicion of wrongdoing. A class action on behalf of these participants was brought against the Board, charging that the rule violated their Fourth Amendment right to be free from unreasonable searches. The district court granted a preliminary injunction. 721 F. Supp. 958 (N.D.Ill.1989). A panel of this court, by a divided vote, affirmed the district court, agreeing that the rule violated the Fourth Amendment. 924 F.2d 664 (7th Cir.1991). We granted rehearing en banc to enable the full court to consider the unclear, delicate, and important question of where the Fourth Amendment should be deemed to strike the balance between the interest of the state in using drug testing as a regulatory instrument and the interest of persons in preserving their physical privacy.

The operative word in the Fourth Amendment is "reasonable," the legal standard therefore is reasonableness, and the decision whether a particular public program that invades interests protected by the amendment is nonetheless reasonable, and therefore lawful, requires a judgmental, forward-looking, balance-striking, probabilistic assessment, rather than, as the plaintiffs would have it, a conclusive demonstration of measurable harms certain to be inflicted if the program is struck down. New Jersey v. T.L.O., 469 U.S. 325, 337, 105 S. Ct. 733, 740, 83 L. Ed. 2d 720 (1985); Treasury Employees v. Von Raab, 489 U.S. 656, 674, 109 S. Ct. 1384, 1395, 103 L. Ed. 2d 685 (1989); International Brotherhood of Teamsters v. Department of Transportation, 932 F.2d 1292, 1304-05 (9th Cir.1991); Willner v. Thornburgh, 928 F.2d 1185, 1187-88 (D.C.Cir.1991); Harmon v. Thornburgh, 878 F.2d 484, 487-88 (D.C.Cir.1989). The weaker the interest asserted, therefore, the less showing of countervailing harms the government must make. Skinner v. Railway Labor Executives' Ass'n, 489 U.S. 602, 633, 109 S. Ct. 1402, 1421, 103 L. Ed. 2d 639 (1989); Willner v. Thornburgh, supra, 928 F.2d at 1188 ("decreasing levels of intrusiveness require decreasing levels of justification"), 1190; Taylor v. O'Grady, 888 F.2d 1189, 1199 (7th Cir.1989); Thomson v. Marsh, 884 F.2d 113, 115 (4th Cir.1989) (per curiam). And since the plaintiff's interest--the privacy interest--cannot be quantified, neither need the regulatory interest be quantified. Although the appeal is from a preliminary injunction, the parties have asked us to decide the ultimate question, which is whether the drug-testing program violates the Fourth Amendment. Chicago Observer, Inc. v. City of Chicago, 929 F.2d 325, 329 (7th Cir.1991); Cronin v. U.S. Department of Agriculture, 919 F.2d 439, 445 (7th Cir.1990).

The facts that bear on the balance of the competing interests in this case, and therefore on the reasonableness of the challenged rule, are set forth in the panel majority opinion, and can be summarized briefly. Horse racing in Illinois, as everywhere else in the civilized world (as far as we know), is a heavily regulated activity, and this for three reasons. It is highly dangerous to jockeys and to their counterparts in harness racing, called drivers; it is a magnet for gambling; and it has an unsavory, or at least a shadowed, reputation, growing out of a long history of fixing, cheating, doping of horses, illegal gambling, and other corrupt practices. Phillips v. Graham, 86 Ill. 2d 274, 286, 56 Ill. Dec. 355, 360, 427 N.E.2d 550, 555 (1981); Garifine v. Monmouth Park Jockey Club, 29 N.J. 47, 56-57, 148 A.2d 1, 5 (1959); U.S. Commission on the Review of the National Policy Toward Gambling, "Second Interim Report" 52, 54 (July 1976); cf. Marrone v. Washington Jockey Club, 227 U.S. 633, 33 S. Ct. 401, 57 L. Ed. 679 (1913). The second and third points are of course related. Gambling on horse races as on other sports and games has generally been illegal in this country, and illegal activities create and attract unsavory characters and methods: especially horse racing, because of the enormous sums bet on it. Illinois allows parimutuel betting (where the odds are determined automatically by the amount bet on each horse rather than set by bookmakers), but betting through bookmakers continues to flourish though illegal and the industry has never been able wholly to dispel an aura of scandal.

The Illinois Racing Board has a dual concern with the use of illegal drugs by participants in horse races. First is a concern with the personal safety of those participants, who might be injured or killed in accidents that would not have occurred but for such use. Second is a financial concern. Illinois derives tens of millions of dollars in tax revenues annually from parimutuel betting. Those revenues would fall if betting declined as a result of a belief by the public that the fairness of the races was being impaired because jockeys and other participants were using drugs. Pelling v. Illinois Racing Board, 214 Ill.App.3d 675, 158 Ill. Dec. 322, 325-26, 574 N.E.2d 116, 119-120 (1991).

Members of the Jockeys' Guild first approached the Illinois Racing Board in 1984 with expressions of concern about drug use by participants in horse races. In 1985 the Board adopted a pilot drug test screening program for jockeys and harness drivers; 17 percent tested positive for cocaine, marijuana, or both. The validity of the test methodology is challenged but there is doubtless some drug use among horse-race participants, for the Jockeys' Guild has instituted a drug counseling program for its members. In 1988 the Board, acting pursuant to a statute that gives it broad regulatory authority over horse racing in Illinois, adopted the rule challenged in this case. The plaintiffs do not argue that the enabling statute is invalid or that the rule is not authorized by it. Nor do they challenge the entire rule. They do not challenge the part that forbids horse-race participants to use on the grounds of any race track any illegal drug (technically, any "controlled substance" not lawfully prescribed by a physician). They challenge the method of enforcement--random drug testing, up to five times a year per participant. The individual is permitted to give his urine specimen in the (relative) privacy of a toilet stall, with a representative of the Board standing by but not actually watching the individual urinate.

The Fourth Amendment, as interpreted in the modern cases, protects privacy. Katz v. United States, 389 U.S. 347, 88 S. Ct. 507, 19 L. Ed. 2d 576 (1967); United States v. Janik, 723 F.2d 537, 547-48 (7th Cir.1983). Urination is generally a private activity in our culture, though, for most men, not highly private. Men urinate side by side in public restrooms without embarrassment even though there is usually very little, and often no, attempt to partition the urinals. In hospitals and physicians' offices, urine samples of both men and women are generally taken by female nurses or technicians under conditions of privacy similar to those prescribed by the racing board's rule (there are female as well as male jockeys). The affront to the cluster of emotions that define the sense of privacy that is caused by the giving of a urine sample is not the same for everybody and of particular relevance here it is slight for people who for whatever reason are subject to frequent medical examinations. Boxers receive complete medical examinations before each match, which may be several times a year. Many Americans have an annual physical examination in which they provide a urine sample, expose their most private parts to inspections, and are poked, squeezed, and kneaded in these and other private places--all this with a minimum sense of embarrassment and certainly none of affront. Athletes (not limited to boxers), actors, and airline pilots are illustrative of the many types of worker whose job is of a character that requires the worker to submit to frequent medical examinations. As Hamlet said, "The hand of little employment hath the daintier sense." The less habituated a person is to undergoing medical or other intrusions into his private realm, the more sensitive he is apt to be to such intrusions; the more habituated he is to them, the less sensitive he is apt to be. A further point, which distinguishes the person who has frequent medical examinations because of illness from the person who has frequent medical examinations because his job requires it, is that the latter voluntarily trades away some of his privacy for other goods. Cf. Willner v. Thornburgh, supra, 928 F.2d at 1190, 1193. Self-selection will tend to allocate jobs in which privacy is limited to persons who value privacy less.

The plaintiffs do not doubt that the Board could if it wanted require them to undergo a complete medical examination before each race, and such an examination would require the giving of a urine specimen. The only difference would be the supervising personnel. On balance the full examination would be substantially more intrusive than the test required by the challenged rule. Also more numerous, since jockeys race more than five times a year.

Like other losses, losses of privacy should be evaluated at the margin. The issue is the incremental loss of privacy caused by the Board's rule. International Brotherhood of Teamsters v. Department of Transportation, supra, 932 F.2d at 1300; see also Willner v. Thornburgh, supra, 928 F.2d at 1190-92. That increment is slight, and the burden on the state of establishing the need for the rule is correspondingly lightened. Certainly the state need not, as it might in other circumstances have to do, demonstrate a danger of mass disaster--the sort of danger that a drug-using airline pilot or missile silo commander might pose. It need not point to specific accidents caused by a drug-using jockey or some other horse-race participant. It need only demonstrate that the state interest in random drug testing is substantial, not that it is transcendent.

The state's interest here has, as we have said, two components. One is the personal safety of the participants, the other the financial interest of the state. With respect to the first, Judge Wood's panel majority opinion was eloquent on the dangers of racing (including harness racing) and on the potential for accidents that is posed by drug-using by any of the participants. The jockeys, the harness drivers, the assistant starters (who help the jockey control the horse in the starting stall), and the outriders (also called parade marshals: they ride the horses from the paddock to the track before the race) are personally in danger, as well as a danger to persons in the vicinity. The starter is in no danger himself, but if he starts the race before all the horses are facing forward in the starting stalls, poised to run, he can provoke an accident that might crush a jockey or an assistant starter. (A race horse weighs half a ton.) The persons at greatest risk are the jockeys and harness drivers; they are at risk from each other but also from the other participants, and that is why it is important that all the participants be careful and alert. Drug use impairs care and alertness, slows reflexes, impairs judgment.

The more dangerous an activity is, the more dangerous is drug use by participants in it. Horse racing is the most dangerous of the common sports, other than auto racing. An average of 2 jockeys are killed each year in this country, out of some 2,000 (the membership of the Jockeys' Guild, 1 Encyclopedia of Associations 2221 (25th ed., Deborah Burek ed. 1991)), and another 100 are injured seriously enough to be disabled for at least a week. The Jockeys' Guild has 40 permanently disabled members--one out of every 50. The annual death toll of 1 per 1,000 implies that a jockey who races for 10 years has a 1 percent chance of dying in a race. How much the use of illegal drugs contributes to this toll is unknown, but cannot be assumed to be trivial.

To the danger to personal safety must be added the danger to the state fisc. As we have said, the State of Illinois derives substantial revenues from horse racing, and does so in an era of financial stringency for state and local government. It can ill afford a drop in those revenues. Yet they would drop if parimutuel betting declined, and such betting might decline if the public suspected widespread use of drugs by horse-race participants. We do not want to place undue weight on this consideration. The primary losers from any decline in revenues from horse racing are the owners of the race tracks, who being private entities could require random drug-testing of the users of their facilities without coming within the scope of the Fourth Amendment at all, and yet they have not done so. Maybe they have not done so because the Illinois Racing Board has acted for them, but their failure to act is some evidence that the fiscal benefits of the challenged rule are smaller than the Board claims. Smaller, but surely not negligible, given the public suspicion of the honesty of racing and other sports; and they reinforce the concerns about the personal safety of the horse-race participants if drug use is not deterred with the aid of a vigorous program of random drug-testing.

The district judge was influenced by the fact that there have been no proven cases of lethal or other serious accidents caused by drug-using horse-race participants, or any other public scandals resulting from such use (as distinct from the doping of the horses themselves--an old problem in horse racing). But government is not limited to addressing public safety problems after serious accidents reveal its want of foresight. Menora v. Illinois High School Ass'n, 683 F.2d 1030, 1034 (7th Cir.1982). Dissenting in Von Raab, Justice Scalia pointed out that "neither frequency of [drug] use nor connection to harm is demonstrated," 489 U.S. at 681, 109 S.Ct. at 1398, and the majority acknowledged that the "testing scheme was not implemented in response to any perceived drug problem among Customs employees" and "that the program actually has not led to the discovery of a significant number of drug users"--in fact, "no more than 5 employees out of 3,600 have tested positive for drugs." Id. at 673, 109 S.Ct. at 1394. (That is a little more than one-tenth of 1 percent. Compare the 17 percent who tested positive in the Racing Board's pilot test--albeit this may not be a reliable number.) Yet still the program was upheld.

When we compare the plausible dangers, both to safety and to revenue, that the challenged rule aims to combat with the very moderate incremental infringement of privacy that the rule brings about, we conclude that the rule is not unreasonable, and therefore that it does not violate the Fourth Amendment. In so concluding we join the only other federal court of appeals to have considered the legality of a drug-testing program for horse-race participants. Shoemaker v. Handel, 795 F.2d 1136, 1141-43 (3d Cir.1986). Horsemen's Benevolent & Protective Ass'n, Inc. v. State Racing Comm'n, 403 Mass. 692, 532 N.E.2d 644 (1989), invalidated a drug-testing program for jockeys similar to the one challenged in this case, but the primary ground was the state constitution rather than the Fourth Amendment and the decision preceded the Supreme Court's drug-testing cases, though so did Shoemaker. Serpas v. Schmidt, 827 F.2d 23 (7th Cir.1987), rejected Illinois's claim to be allowed to conduct random searches of the living quarters of "backstretchers," who take care of horses at race tracks but do not participate in the race. It was not a drug-testing case but an old-fashioned search case, and the persons to be tested were not actual race participants. So the intrusion on privacy was greater and the state interest weaker.

Random drug-testing has been adopted and challenged in a variety of other settings of course, and it is from the cases dealing with these other drug-testing programs that we have distilled the principles that have guided us in this opinion. Their factual differences from each other and from this case are too great for the other cases to control our decision, but a brief summary may help to demonstrate the conformity of our decision with the developing case law.

Almost all the cases fall into one of three categories: transportation workers, government employees, and sports participants. Random drug-testing of transportation workers is consistently upheld, whether they are railroad employees, truck drivers, airline pilots, or bus drivers. The leading case is Skinner v. Railway Labor Executives' Ass'n, supra, but there is a host of others, including Bluestein v. Skinner, 908 F.2d 451 (9th Cir.1990), and Transport Workers' Union v. Southeastern Pennsylvania Transportation Authority, 884 F.2d 709 (3d Cir.1989). The public safety interest in such testing is obvious.

The government employee cases are a mixed lot. Following the lead given by the Supreme Court in Treasury Employees v. Von Raab, supra, the courts uphold such testing where the government employee is armed and therefore potentially dangerous, where he has a security clearance and therefore poses a potential threat of compromising national security should he become addicted to drugs, where he works with dangerous materials, where he is involved in the enforcement of the drug laws themselves and might therefore be tempted into illegal activity involving drugs, or where, as in the case of prison guards, he is in direct contact with drug offenders. Illustrative cases are American Federation of Government Employees v. Skinner, 885 F.2d 884 (D.C.Cir.1989); National Federation of Federal Employees v. Cheney, 884 F.2d 603 (D.C.Cir.1989); Thomson v. Marsh, supra; and our own Taylor v. O'Grady, supra. The cases do not, however, permit random drug-testing of administrative and back office personnel, who though employed by the armed forces or other government agencies do not themselves pose any great threat to safety or other interests. Our decisions in Taylor and Serpas (the "backstretcher" case) are of this type. Analogous is Harmon v. Thornburgh, supra, which held that criminal prosecutors in the Justice Department, unless they hold security clearances or are involved in drug prosecutions, may not be subjected to random drug-testing.

Last are the sports cases. Two are the horse-racing cases that we have already discussed (Shoemaker and Horsemen's Benevolent). Schaill v. Tippecanoe County School Corp., 864 F.2d 1309 (7th Cir.1988), upheld random drug testing of high school athletes, but we based the decision on the idea that the Fourth Amendment applies with diminished force in schools, an idea with no apparent relevance to the present case. O'Halloran v. University of Washington, 679 F. Supp. 997, 1005-07 (W.D.Wash.), rev'd on other grounds, 856 F.2d 1375 (9th Cir.1988), upheld random drug testing of athletes participating in intercollegiate competition, emphasizing the prevalence of drug abuse in competitive sports. See generally Note, "Drug Testing and the Student-Athlete: Meeting the Constitutional Challenge," 76 Iowa L.Rev. 107 (1990).

The present case differs from the transportation cases in that those cases involve a danger to passengers and bystanders as well as to the transportation workers themselves. The present case is equally remote, however, from the government employee cases in which random drug-testing was struck down, because in those cases the employees in question were dangerous neither to themselves nor to third parties. The salient facts in the present case, which have no direct counterpart in any other cases except the other horse-racing cases, are that the incremental invasion of privacy is very slight; the physical danger of drug use--not only (or always) to the user himself, but also to other participants albeit not to the broader public--is acute; and there is in addition a substantial state financial interest (the parimutuel revenues), which is comparable with the interest in maintaining an efficient and productive work force, stressed in Willner v. Thornburgh, supra, 928 F.2d at 1192-93. In conjunction, these factors persuade us that the state interests outweigh the very limited privacy interest and therefore that the program is lawful.

Obviously this is a more difficult judgment with regard to the starters, assistant starters, and parade marshals than with regard to the jockeys and harness drivers. Not only are the jockeys and drivers in greater danger, but they are the athletes, whom the Racing Board could reasonably decide to make submit to physical examinations before each race. However, the other participants who are in no or little danger themselves--the starters and the outriders--can endanger other participants, while the assistant starters can endanger both themselves and the jockeys or drivers; and all these other participants can, if impaired or corrupted by a drug habit, ruin the fairness of the race. It is desirable, too, not to multiply legal distinctions indefinitely or engender internal frictions within the racing community by allowing random drug-testing of jockeys and drivers but not of the other participants in the race.

We conclude that the Racing Board's program of random drug testing of participants in horse racing does not violate the Fourth Amendment. The judgment is therefore reversed and the case remanded with instructions to dismiss the suit.

REVERSED AND REMANDED.

HARLINGTON WOOD, Jr., Circuit Judge, with whom BAUER, Chief Judge, and FLAUM, and RIPPLE, Circuit Judges, join, dissenting.

Judge Posner for the majority illuminates the issue with balance and fairness, but I continue to believe that the Board rule is an unreasonable and unnecessary encroachment on the privacy and constitutional protections of horse race participants.

Illegal drugs are the scourge of modern society, but hopefully the various efforts of the "war on drugs" will in time succeed. The culprits are the smugglers, dealers, and all the others who conspire out of greed to spread the destruction of illicit drugs. In drug prosecutions we carefully guard all the constitutional rights of those principal players. However, in this civil context, far from that circle of culprits, the majority has little difficulty in diminishing the constitutional protections of those involved in horse racing.

The majority says that if you are a participant in Illinois horse racing you deserve fewer constitutional protections than the rest of us. These participants will now be subject to warrantless body searches, searches without any personalized suspicion. That is held to be both reasonable and practicable because horse racing is already regulated, is dangerous, and is a money maker for the state because of the contributions gamblers make to state revenues. It is feared that the gamblers, who are spared these intrusions, might lose faith in the integrity of horse racing. Those reasons are not without some merit, as Judge Posner claims, but there is more to be considered in the reasonableness balancing. The majority is racing way out ahead of the Supreme Court on a very slippery legal track with no known finish line. Its reasoning can carry us far beyond horse racing. On this kind of uncertain and dangerous track there can be no winners, only losers--losers of constitutional protections. Two recent Supreme Court cases that consider random urine drug testing in different factual contexts-- National Treasury Employees Union v. Von Raab, 489 U.S. 656, 109 S. Ct. 1384, 103 L. Ed. 2d 685 (1989), and Skinner v. Railway Labor Executives' Association, 489 U.S. 602, 109 S. Ct. 1402, 103 L. Ed. 2d 639 (1989)--deserve more consideration and application than the majority gives them.

In Von Raab, the Court held that the United States Customs Service could implement an employee drug-testing program for those employees aspiring to positions involving the interdiction of illegal drugs or to positions requiring the incumbent to carry firearms. Those employees, without a warrant or individualized suspicion, were required to produce urine samples to be analyzed for evidence of illegal drug use. The Court approved the program as a reasonable fourth amendment intrusion because of special governmental needs beyond the normal needs of law enforcement. It reached that conclusion by balancing individual privacy expectations against any impracticability the government's interests might suffer if first required to secure a warrant or offer some degree of individualized suspicion. 489 U.S. at 665-66, 109 S.Ct. at 1390-91.

Central to the Court's decision in Von Raab was its finding that there existed a national crisis in law enforcement caused by the smuggling of illicit drugs into this country. The Customs Service was characterized as "our Nation's first line of defense against one of the greatest problems affecting the health and welfare of our population." Id. at 668, 109 S.Ct. at 1392. Those employees might also have their physical safety threatened, or be tempted by traffickers' bribes; the record revealed some deaths as well as bribery and other integrity problems among Customs employees. The majority explained that the "national interest in self-protection could be irreparably damaged if those charged with safeguarding it were, because of their own drug use, unsympathetic to their mission of interdicting drugs"; the consequences could be disastrous. Id. at 670, 109 S.Ct. at 1393. The government therefore had a compelling interest in ensuring the unimpeachable integrity and judgment of our national front-line interdiction employees. The government also had a compelling interest in ensuring that drug-impaired employees authorized to carry guns would not jeopardize the lives of others. When viewed as a whole, the situation involved "extraordinary safety and national security hazards," id. at 674, 109 S.Ct. at 1395, that were sufficient to substantiate a testing program even though there was no evidence of an existing drug problem.

Against those great public interests the majority weighed the employees' interests in having to submit to urine tests and approved the regulation. Customs employees were found to be "[u]nlike most private citizens or government employees in general" in that they reasonably expected their privacy interests to suffer. Id. at 672, 109 S.Ct. at 1394. In that work with those responsibilities the expectation of employee privacy was found to be diminished. Id.

Von Raab was a 5-4 decision, the closeness of which may not be so significant now in view of subsequent changes in Court personnel. Nevertheless, it remains current precedent with a strong dissent by Justice Scalia, joined by Justice Stevens. Justice Scalia characterized the urine test imposed on Customs employees as a "type of search particularly destructive of privacy and offensive to personal dignity." Id. at 680, 109 S.Ct. at 1398. He then dissented because of the lack of evidentiary foundation; in his view neither the frequency of drug use nor its connection to possible harm to others had been demonstrated. He saw no real evidence of a problem that would be solved by urine testing. Id. at 681, 109 S. Ct. at 1398. That is in contrast to his view under different facts in Skinner, a case we shall next examine, where he agreed that there was sufficient justification.

In Skinner, the Supreme Court also arrived at the conclusion that the searches of certain railroad employees are constitutional by balancing governmental "special needs" with the privacy interests of the individual. 489 U.S. at 619, 109 S.Ct. at 1414. It found that the government's interest in regulating the employees for public safety reasons presented special needs that may justify departure from usual constitutional requirements. The Court saw fit to dispense with the warrant requirement in the railroad context and to permit a search without individualized suspicion. Id. at 624, 109 S.Ct. at 1416.

The analysis in Skinner begins with a recognition that the fourth amendment "guarantees the privacy, dignity, and security of persons against certain arbitrary and invasive acts by officers of the Government or those acting at their direction," id. at 613-14, 109 S.Ct. at 1410-11, and that "the collection and testing of urine intrudes upon expectations of privacy that society has long recognized as reasonable." Id. at 617, 109 S.Ct. at 1413. The analysis of urine "can reveal a host of private medical facts about an employee including whether he or she is epileptic, pregnant or diabetic." Id. Urine tests were accordingly deemed searches under the fourth amendment and urine testing was characterized as a more difficult question than breath testing or even blood testing because of the particular privacy concerns implicated by urine testing. Id. at 626, 109 S.Ct. at 1417.

The Court recognized that an improperly operated locomotive could be lethal. Id. at 628, 109 S.Ct. at 1419. The particular railroad employees to be tested therefore had duties fraught with risk of injuries to others besides railroaders with possible disastrous consequences. The evidence in Skinner showed that there were forty-five train accidents between 1975 and 1983 resulting from the errors of alcohol- or drug-impaired employees. Some of the railroad accidents caused the release of hazardous materials and in one instance caused the evacuation of an entire Louisiana community. Id. at 608, 109 S.Ct. at 1408.

The Skinner Court further noted that rail employees have diminished expectations of privacy by reason of participation in an industry pervasively regulated to insure safety, id. at 627, 109 S.Ct. at 1418, but cautioned that merely being in a regulated industry is not enough to view privacy as minimal. Id. at 628, 109 S.Ct. at 1419. It also noted that the urine specimen was collected in a medical environment by personnel unrelated to the railroad employer, and was thus not unlike regular physical examination procedures. Id. at 626-27, 109 S.Ct. at 1417-18.

Surely after considering Von Raab and Skinner it can be seen that horse racing is not a category of dangers of the same nature or magnitude. It is hard to imagine what could happen during a horse race that could raise concerns analogous to the "extraordinary safety and national security hazards" in Von Raab or the potential for "great human loss" in Skinner. It is certainly not possible to explain the necessity for testing the plaintiffs as due to any state crisis to protect the safety, health, and welfare of its population. The race track is not the front line of defense against any public danger. With more and more people crowding our privacy there will be less and less of it, but in this horse-racing case there is no need to sacrifice privacy for a speculative symbolic cause without some reasonable foundation in fact, even if we were to assume the other Supreme Court criteria could be met in the state context. There are no compelling underlying concerns sufficient to support a "no evidence" approval of testing in this case.

In contrast to Skinner, moreover, there is not the slightest evidence of drug-related accidents in this horse-racing case. The basis the Board uses to justify this intrusive program is speculative and unfounded. There was a pilot drug test in 1985 that the Board relied on in part to impose random testing. In opposition the plaintiffs offered evidence to demonstrate specifically that there were substantial flaws in the manner that the pilot program was conducted so as to make it unreliable. The Board recognizes the flaws, but claims the tests were adequate for its purposes. Even given the 1985 test, however, the Board failed to offer evidence of a single Illinois narcotics-related accident or integrity lapse, but nevertheless drug testing was seen by it as the thing to do. The Board's good intentions, however, should not be a substitute for misguided intentions.

Obviously horse racing can be dangerous, but there is some risk even when a single horse is standing in a stall. Racing is by no means a jockey or driver demolition derby. The errors of jockeys or drivers may occasionally cause accidents. Some risk is inherent in this sport. Those involved know that and accept it, but the serious horse race accidents are more often caused by the physical failure of the horses, not the fault of the participants. It is with the horses where the real dangers of horse racing are hidden. Even the plaintiffs in this case might prefer not to have that issue raised, but those concerned with the real dangers and abuses in horse racing would do well to consider the horses. There might then be no excuse to intrude on the privacy of the participants without more to go on than there has been offered in this case.

Because the dangerousness of horse racing is an issue in this case with the implication that randomly tested participants will make horse racing safer we need to consider briefly the significant causes of horse-racing accidents.1 These include, among others, the use of both legal and illegal drugs to increase race horse speed, or to keep an unfit horse in the race by lessening its otherwise disabling pain. A controversy exists about the use of prerace legal drugs in order to keep unfit horses temporarily able to compete. (In the recent past it is reported that one Illinois horse-racing official resigned because the state would not implement a recommendation to ban all prerace horse medications.) There is also, for instance, some blame placed on using two-year-old horses not fully developed, and also on excessive racing of a particular horse, as well as racing without regard to track conditions.

Some experts estimate that 90% of race horses have physical conditions that can lead to accidents. Thoroughbred race horses weigh over 1000 pounds, but are racing on legs thinner than a human's while carrying the weight of a jockey, and sometimes added weights. Standardbred trotters and pacers are pulling drivers, usually weighing more than jockeys, in two-wheel sulkies. In addition to horses' legs there is great strain put on hearts and lungs. The press recently reported three thoroughbred horses, Go For Wand, Mr. Nickerson, and Shaker Knit, collapsed for physical reasons and died on the track during the 1990 Breeders Cup Races at Belmont Park.2

If this state's concern were really with danger in sports it might look for drug problems in auto, motorcycle, motorboat, and stock car races where the speed is greater. Those vehicles are filled with gasoline that in accidents may explode into flames with auto parts flying all over and injuring spectators as well as drivers. The same situation exists with air shows. Polo or rodeo events also have serious dangers, not only to the participants but to spectators. It is rare, however, that jockeys, drivers, or their horses ever clear the fence and injure spectators.

The majority argues that there is no need to wait for a disaster to happen. That is true in ordinary situations, but speculation about such an event is no grounds for dispensing with constitutional protections of privacy, dignity, and security. I doubt that preventing the worst horse race accident in history would justify setting aside the Constitution.

The state's real interest appears to be in the money that horse racing generates, not the accidents. The majority does not put much weight on the alleged danger to the "state fisc" as a basis to justify the intrusion, and I can understand that, but I believe that factor strongly motivates the state. I have great doubt, however, that the possible and speculative loss of revenue in any amount from a sporting event can justify a breach of the fourth amendment.

The majority analysis does not use the term "search," only "privacy." That suggests that privacy may be something less. Our privacy is of very great importance, but it can be encroached upon without a search, for instance, by peeking over one's shoulder to see what is being read. In one way or another our privacy is continually being eroded. We have almost grown accustomed to it. This case, however, should be seen for what it is. Privacy is violated by reason of a search prompted without the least suspicion. Taking a urine sample by a search under these particular factual circumstances is an extreme violation of privacy.

As the first of several illustrations to show that men in general are less sensitive to privacy intrusions (most jockeys and drivers are male), the majority injects into this argument the use of public restroom urinals. That use is usually one of practical necessity and cannot mean that men would not prefer more privacy. In fact the use of such facilities may increase, not diminish, the individual's appreciation and value of his privacy. In any event when men urinate side by side in public restrooms one man seldom asks his urinal neighbor to provide a specimen to send to a laboratory.

When we visit doctors' offices for an examination, another majority illustration, we do so voluntarily out of our personal health concerns, and we can rely on the doctor-patient privilege. That type of medical setting, however, is not what is envisioned by the Board's rule. It is the racing stewards who notify the participants that they are to be tested. Those chosen must present themselves at some collection site at the track secured by Board representatives. The participant uses a stall, suggesting that a bathroom at the track is used. The Board representative is close by, but does not observe the act of urination. The representative receives the specimen from the participant and delivers it to a laboratory. In contrast to Skinner, the Board's testing involves neither a medical environment nor people unrelated to the race meet. See 489 U.S. at 626-27, 109 S. Ct. at 1417-18.

The majority also seems to assume that the plaintiffs are subject to frequent medical examinations and therefore that taking the urine sample is merely a minor invasion of their privacy. That is not the case. The record shows that there are no physical examinations as a condition of licensure for starters, assistant starters, outriders, or parade marshals. Physicals for those participants would be very hard to justify. Jockeys and drivers can be required to take licensure physical examinations under present rules, but the record shows that historically physicals have seldom been required even for them. I also see little to the majority's argument that the Board could require each participant to take a complete physical before each race so the taking of an occasional urine sample should be of no concern. That would, of course, not be practical or justified and would result in more physicals at the track than horse races.

The majority also injects a bit of economic theory by suggesting that jobs subject to this kind of intrusion will be chosen only by those who value privacy less so therefore the intrusion makes no difference. That may be true to an extent, but it should not be allowed to narrow horse race employment to those people. I am not sure what kind of people they would be who do not value their privacy. There are many reasons for those who choose horse racing as a career: money; excitement; fame; and even a love and knowledge of horses. If their privacy meant anything they would have to go elsewhere to keep their privacy. Horse racing would suffer.

The majority analysis seems to be broadly based on the assumed premise that those involved in horse racing are not entitled to the same degree of privacy as the rest of us and that the intrusion is a minor one. I believe that to be wrong. Horse racing is certainly regulated, but that is not enough. The Supreme Court has explained to us what it takes to justify an intrusion on the privacy of government employees in positions affecting great public and national interests. Those involved in horse racing--private citizens regulated by the government--should be entitled to that same high threshold set by the Supreme Court in Von Raab and Skinner. They should not be needlessly pushed over it.

There is, as there should be, concern about the possible use of drugs by those involved in horse racing. These plaintiffs, however, object only to random testing without any suspicion or cause whatsoever to single out the person to be tested. The district court found that a racing steward could require testing upon "corroborated evidence or observable phenomena, such as direct observation of drug use or possession and/or physical symptoms of being under the influence of alcohol or drugs such as erratic behavior." A rule provision of that sort requiring something to go on is reasonable and is not the basis of plaintiff's objection. In this closely regulated and supervised activity that ought to be enough. Races are filmed and conducted under the close scrutiny of race stewards and thousands of spectators. Plaintiffs' privacy should be invaded "no more than is necessary." New Jersey v. T.L.O., 469 U.S. 325, 343, 105 S. Ct. 733, 743, 83 L. Ed. 2d 720 (1985).

The plaintiffs also offered some constructive alternative and less intrusive suggestions. One was that a comprehensive system of trained supervision would be a sufficiently effective tool to detect any impairment during racing programs. The plaintiffs also recommended neuropsychological testing, a type of performance test that, unlike a urine test with its limited value,3 directly measures impairment of behavioral functions. Another possibility would be drug education programs. Nor is there any reason to believe that the usual law enforcement techniques, such as informants, undercover agents, and sting operations would not be effective if there is in fact a drug problem at the race track. These suggestions in some combination ought to be sufficient at least against a background of little or no demonstrated need.

The majority claims that plaintiffs would require "a conclusive demonstration of measurable harms certain to be inflicted if the program is struck down." If that is what plaintiffs seek then I would not support that degree of certainty. In between those two evidentiary positions of none and much, I believe there is reasonably a middle ground. I believe there should be at least some reliable evidence, not necessarily conclusive, to suggest a reasonable need to encroach on fourth amendment rights. I do not concede, however, that the interests here involved are nearly significant enough to meet the tests of Von Raab and Skinner.

The Board rule has to be weighed and judged for reasonableness. The majority says this calls for a "judgmental, forward-looking, balance-striking, probabilistic assessment," then takes a pessimistic view of the future, puts its thumb on the scale, and opts in favor of the testing. The majority cites the number of horse-racing accidents nationwide and concedes that it cannot say how much the use of illegal drugs contributes to that record. Nevertheless then it says that this percentage "cannot be assumed to be trivial." That constitutes its "judgmental, forward-looking, balance-striking, probabilistic assessment," an impressive way of saying that no evidence of need has been offered and then excusing that lack by saying that no evidence is needed anyway.

The plaintiffs' privacy interest is treated as trivial, but that is the majority's slippery track. Where will this convenient disregard of constitutional principles for some appealing cause actually end? Some may see good reasons to apply the same test to many other occupations from judges4 to cab drivers. Teachers are molding our next generation so why not them, too? There is no end to the future corrosive effects of the precedent here established of dispensing with the Constitution when it is troublesome and impracticable without even some reasonable showing of need. I, too, would like to achieve a drug-free society, but would accomplish it with more constitutional restraint than the majority espouses.

The majority has some trouble reining in for random testing those race participants other than jockeys and drivers, and understandably so. There is not the slightest evidence, only speculation, with or without their use of drugs, that those other participants could in the ordinary course cause serious damage of any kind, or ever have, and certainly not to the public. These other race participants are found guilty and subjected to drug testing primarily because of their association with jockeys and drivers who are themselves not shown to be guilty. In view of the broad sweep of this drug rule approved by the majority I fail to see why the racing stewards who have nearly absolute control over horse racing are not also included. And why not the Board members themselves?

If it is all placed on the scale and weighed for reasonableness, as we are to do, that scale should tilt in favor of the Constitution and the rights of these participants.5 It is too bad we cannot ask the horses about horse racing, but they are not talking.6

I respectfully dissent.