Norman L. Berthiaume was a Maine nurse practitioner licensed by the Maine Board of Nursing. In July 1990, Berthiaume pleaded guilty to the violation of 18 U.S.C. § 1462, which prohibits the importation of obscene materials. Berthiaume had ordered and received from a United States Customs Service undercover operation in Mexico, through the mails, a videotape depicting children engaged in sexual activity. He was given two years' probation and a $2,000 fine; the sentencing judge said that the crime was an aberration.

While the charges were still pending, Berthiaume notified the Board of the situation and applied for a renewal of his nursing license. He said that he had purchased the child pornography out of professional interest. In November 1990, the Board met with Berthiaume and his attorney at an informal conference. Afterwards, the Board renewed his license on a probationary basis pending a psychological evaluation by an independent psychologist.

James Bivins, a government lawyer who was advising the Board, contacted two psychologists in an effort to find someone who could evaluate Berthiaume. One of the psychologists recommended Dr. William O'Donohue, a psychologist who was then an assistant professor at the University of Maine specializing in the evaluation of sex offenders. Jean Caron, the Board's executive director (a non-voting staff position), contacted O'Donohue, who told her and Bivins about the tests he employed and considered appropriate in such an evaluation.

These tests, O'Donohue told them, included interviews, filling out surveys, and a penile plethysmograph test. He explained that in the plethysmograph test, the subject places on his penis a device that measures its circumference and thus the level of the subject's arousal as he is shown sexually explicit slides or listens to sexually explicit audio "scenes." Caron and Bivins reported their conversation to Betty Clark, the chair of the Board, who told them to retain O'Donohue's services.

Bivins arranged the evaluation and informed Berthiaume that it would include a penile plethysmograph test. Berthiaume was told that if he refused to take the test, the Board would take steps to revoke his nursing license. Berthiaume expressed reservations orally and in writing but ultimately agreed to the test and signed the informed consent form he was given by O'Donohue. Berthiaume says he signed it under duress and disputes whether he was given enough information about the test to make his consent truly informed.

O'Donohue administered the battery of tests, surveys, and interviews that he had discussed with Bivins and Caron and sent a report to the Board. In his report, O'Donohue stated that the plethysmograph test was "inconclusive." However, O'Donohue made a diagnosis of probable pedophilia, relying primarily on Berthiaume's specific request for pornography depicting boys of particular ages and an admission (later retracted) by Berthiaume during an interview with O'Donohue that he had regular sexual fantasies about boys and girls between the ages of 12 and 15.

In December 1990, the Board voted to renew Berthiaume's license on a probationary basis for two years. He was required to tell employers and supervisors about his agreement with the Board, limit his clientele to patients older than 18, and receive psychological counseling. Berthiaume entered into a consent agreement with the Board in February 1991 agreeing to these conditions.

Berthiaume then brought suit in the federal district court in Maine, seeking damages under 42 U.S.C. § 1983 and the Maine Civil Rights Act, Me.Rev.Stat. Ann. tit. 5, § 4682, alleging violation of his rights under the United States and Maine constitutions. The defendants included Clark, Caron, Bivins, and O'Donohue.1 The defendants countered with a summary judgment motion asserting defenses of absolute immunity for officials involved in quasi-judicial proceedings and qualified immunity for officials who do not violate clearly established rights.

The district court rejected the absolute immunity defense, which we need not reach. As to qualified immunity, the district court denied the defendants' request to dismiss, relying on our decision in Harrington v. Almy, 977 F.2d 37 (1st Cir.1992); the court said that "factual issues" remained open that were necessary to the decision whether the defendants violated a clearly established right and thus forfeited immunity. These open issues included:

(1) whether Defendants acted reasonably in failing to consider alternative approaches to meeting the Board's need for additional information without requiring Plaintiff to take the penile plethysmograph test; (2) whether the test was shocking, degrading, and humiliating and, if so, whether Defendants acted reasonably in failing to consider the impact of the procedure on Plaintiff; (3) whether the penile plethysmograph was scientifically capable of meeting the legitimate state interest in this case; and (4) whether the extent of Defendants' inquiry into the scientific validity of the penile plethysmograph was reasonable in light of the intrusiveness and invasiveness of the test.

The defendants have now appealed. At the threshold, Berthiaume argues that we do not have jurisdiction to consider these appeals. Ordinarily, the district court's rejection of a qualified immunity defense is immediately reviewable under the collateral order doctrine, for reasons explained by the Supreme Court in Mitchell v. Forsyth, 472 U.S. 511, 524-30, 105 S. Ct. 2806, 2814-18, 86 L. Ed. 2d 411 (1985). However, the rule is subject to a recent exception established in Johnson v. Jones, 515 U.S. 304, 115 S. Ct. 2151, 132 L. Ed. 2d 238 (1995), which Berthiaume invokes here.

In Johnson, the Supreme Court held that interlocutory appeals would not be permitted to challenge one type of ruling denying qualified immunity. Where qualified immunity turns on a question of fact and the district court declines to grant summary judgment because it says that the fact is genuinely disputed, there is no interlocutory review. 515 U.S. at 313, 115 S.Ct. at 2156. The Court's reason for the exception was prudential; it wanted to foreclose a narrow, time-consuming inquiry whose resolution by interlocutory appeal was a game not worth the candle. See id. at 316-17, 115 S.Ct. at 2157-58.

Conversely, a defendant who concedes arguendo the facts found to be disputed is not barred by Johnson from taking an interlocutory appeal on a legal claim that the defendant is nevertheless entitled to qualified immunity on facts not controverted. This view of Johnson, set forth in Stella v. Kelley, 63 F.3d 71, 74 (1st Cir.1995), has been confirmed by Behrens v. Pelletier, 516 U.S. 299, 313, 116 S. Ct. 834, 842, 133 L. Ed. 2d 773 (1996). The Court in Behrens recognized that determining which facts are uncontroverted may require a certain amount of appellate effort including review of the record. Id. at 313, 116 S.Ct. at 842.

One other point needs to be borne in mind. The genuine disputes insulated from immediate review under Johnson are those involving facts, such as what happened. Questions of law application--for example, whether a set of assumed facts add up to a constitutional violation--are not so insulated and are ordinarily subject to de novo review. Behrens, 516 U.S. at 312-13, 116 S.Ct. at 841-42. This is equally true in deciding whether the assumed facts show a violation of "clearly established" law.

We turn now to the "merits" of the immunity dispute: whether, assuming that facts found controverted were resolved in favor of Berthiaume, the defendants were nevertheless entitled to qualified immunity as a matter of law based upon facts not controverted. In section 1983 cases, the classical formulation is that an official is immune if his or her actions did not violate the plaintiff's "clearly established" constitutional rights. Harlow v. Fitzgerald, 457 U.S. 800, 818-19, 102 S. Ct. 2727, 2738-39, 73 L. Ed. 2d 396 (1982). Subsequent case law has elaborated upon this concept in two significant ways:

First, the case law makes clear that the question is not whether some right has been established clearly at a highly abstract level, for example, the right to be free from unreasonable searches and seizures. Rather, the question is whether, under the circumstances that confronted the official, "a reasonable official would understand that what he is doing violate[d] that right." Anderson v. Creighton, 483 U.S. 635, 640, 107 S. Ct. 3034, 3039, 97 L. Ed. 2d 523 (1987); see also Hunter v. Bryant, 502 U.S. 224, 228-29, 112 S. Ct. 534, 536-37, 116 L. Ed. 2d 589 (1991) (per curiam); Brown v. Ives, 129 F.3d 209, 211-12 (1st Cir.1997), cert. denied, --- U.S. ----, 118 S. Ct. 1307, 140 L. Ed. 2d 471 (1998).

Second, qualified immunity may exist even though in hindsight a court might determine that the action of the official violated the Constitution. The doctrine of qualified immunity leaves "ample room for mistaken judgments" and protects "all but the plainly incompetent or those who knowingly violate the law." Malley v. Briggs, 475 U.S. 335, 341, 343, 106 S. Ct. 1092, 1096, 89 L. Ed. 2d 271 (1986). This is consistent with the purpose of qualified immunity to avoid unduly chilling official action by exposing officials to personal damage liability for good faith judgments. Harlow, 457 U.S. at 814-15, 102 S.Ct. at 2736-37.

Berthiaume describes the defendants' actions as ones that "shock the conscience" and thereby violate his right to substantive due process. Rochin v. California, 342 U.S. 165, 172, 72 S. Ct. 205, 209-10, 96 L. Ed. 183 (1952). Alternatively, he claims that the plethysmograph test violated his right to be free of unwanted bodily searches under the Fourth and Fourteenth Amendments.2 Depending on circumstances, these might be two quite different claims; but in this case we think the considerations are similar, although the two constitutional standards are different.

The "shock the conscience" test is, here as in the ordinary case, largely directed to the substance of what the defendants did, including the nature of the conduct and the reasons for it. The same issues of intrusiveness and justification are pertinent in examining the matter through a Fourth Amendment lens, where the touchstone is reasonableness. In some cases procedural issues dominate the Fourth Amendment inquiry--the securing of a warrant from a neutral magistrate, for example--but Berthiaume makes no procedural claim in this instance.

Starting with the conduct itself, Berthiaume's main objection is to the plethysmograph test. Of course, the procedure was being carried out in the course of a psychological examination, and there are plenty of ordinary medical procedures that are disagreeable or upsetting to the patient. Competency examinations in criminal cases are similarly intrusive. But we think that the nature of the test in this case would strike most people as especially unpleasant and offensive.

The problem of justification is more complicated--indeed, it is the heart of the case--and it needs to be considered at two different levels. One is the decision to insist that Berthiaume subject himself to psychological screening, and the other is the decision to use the plethysmograph test in the course of that examination. The former involves an administrative judgment, and the latter implicates some measure of the expertise and the practice of a profession.

Starting with the administrative judgment, we think it was not unreasonable, and certainly not shocking, for the Board to require Berthiaume--as a condition of retaining his nursing license--to submit to psychological testing by an expert. Berthiaume had pled guilty to the federal felony of importing obscene materials, and the materials involved children engaged in sexual activity. There was no reason why the Board had to accept Berthiaume's unvarnished assertion that he had acquired the child pornography out of professional interest.

The Board is charged under Maine law with responsibility for the fitness of persons who hold themselves out as nurses. See Me.Rev.Stat. Ann. tit. 32, § 2101. Like doctors, nurses are a part of the medical profession and are entrusted with patient care, where reliance is normally placed on the competence of the nurse or doctor. Given that children were the subject of Berthiaume's pornography, many people would have regarded the Board as downright remiss if it had not insisted on further review.

It is no answer to assert, as Berthiaume does, that government employment or other benefits cannot be conditioned on a surrender of constitutional rights. In some contexts this is true, see Blackburn v. Snow, 771 F.2d 556, 568 (1st Cir.1985); but in others, government employment or other benefits may be conditioned on steps that could not be demanded as of right. E.g., Rankin v. McPherson, 483 U.S. 378, 388-89, 107 S. Ct. 2891, 2899-2900, 97 L. Ed. 2d 315 (1987); Wiley v. Mayor and City Council of Baltimore, 48 F.3d 773, 777-78 (4th Cir.), cert. denied, 516 U.S. 824, 116 S. Ct. 89, 133 L. Ed. 2d 45 (1995). See J. Nowak & R. Rotunda, Constitutional Law § 14.46, at 958 (5th ed.1995).

The more troublesome decision is the use of the penile plethysmograph, but here the main focus must be upon O'Donohue rather than the other defendants. There may be procedures so bizarre that lay persons and lawyers could not conceivably stand behind a doctor or psychologist who proposed their use. But this hardly appears to be such a case. Indeed, we conclude that O'Donohue himself has not engaged in a violation of clearly established law. It follows that the other defendants did not do so either.

Whatever its scientific origins, pedophilia does not have revealing physical features like the measles, and the condition is one that the patient himself would be likely to deny. Given the need for inquiry, O'Donohue thus had good reason to consider the use of any legitimate tool of investigation. The critical question is whether the test was so far outside the range of legitimate professional judgment that its use in the present case was shocking to the conscience, or constitutionally unreasonable, and also violated clearly established law.

O'Donohue said in his deposition that he regarded the test as a legitimate one. But O'Donohue's version of the scientific scene was in some degree controverted by Berthiaume's expert; just how far is a different question. Given the usual standard on summary judgment, St. Hilaire v. City of Laconia, 71 F.3d 20, 24 (1st Cir.1995), cert. denied, 518 U.S. 1017, 116 S. Ct. 2548, 135 L. Ed. 2d 1068 (1996), which is doubly appropriate in light of Johnson, we will assume that Berthiaume's expert is to be believed as to facts, although not necessarily as to legal labels.

The difficulty is that Berthiaume's own expert does not even begin to assert that the penile plethysmograph is quack science and that its use would be considered outlandish in the medical community. Rather, he admitted that the device is an accepted tool in the assessment and treatment of, for example, known child molesters. He called it "pretty much a standard practice in treatment programs for sex offenders," and said that by the late 1980s, the test "had gone from a procedure that was experimental ... [to one] being used all over North America by large numbers of professionals."

This expert's testimony shows a dispute in the scientific community as to how far the device is useful in assessing people not previously shown to have deviant sexual interests. The expert said that the plethysmograph is not useful to identify pedophilia because the test has a high rate of false negatives, that is to say, it will pass as normal many persons who are actually pedophiles. The expert said that this lack of "sensitivity," combined with its intrusiveness, made it inappropriate for use in cases (like this one) where it is used for screening rather than treatment.

But Berthiaume's expert admitted that, while he had criticized the use of the device for screening purposes in a monograph, the issue had not been definitively resolved in the scientific community. Further, he conceded that the device was the most sensitive of all tools that are employed to screen for pedophilia--the least worse of a bad lot, in his opinion. Finally, he agreed with O'Donohue's lawyer that the decision whether to administer the plethysmograph test was a question of "professional judgment."

In short, the situation amounts to this: there was good reason for the Board to determine whether Berthiaume was afflicted with pedophilia and only limited means available for that task; the plethysmograph is widely used in the scientific community for the treatment of pedophilia; its use for screening is debatable and the scientific community is not of one mind; and whether to administer the test was a question of professional judgment. It is doubtful that this testimony would show negligence in a simple case of medical malpractice.

Be that as it may, this testimony would not permit the conclusion that O'Donohue's conduct was so outrageous or unreasonable as to establish a violation of "clearly established" law. We need not suppose that a court would necessarily have upheld a denial of Berthiaume's license if he had refused to take the test. It is enough that there is no prior case law condemning the conduct as unconstitutional, nor is it so plainly beyond the bounds of civilized or rational behavior as to make it an obvious constitutional violation even without direct precedent.

This brings us to Harrington v. Almy, 977 F.2d 37 (1st Cir.1992), relied on by the district court in denying summary judgment. There, on the basis of doubtful testimony by a child, a town police officer became one of 170 individuals accused of child abuse. The police officer was neither formally charged nor convicted of a crime, but the town's city manager insisted on a psychological examination, which included the penile plethysmograph, as a condition of continued police duties. When the police officer sued under section 1983, this court held that the claim of qualified immunity by the same manager could not be resolved on summary judgment.

The court's original opinion asserted that Harrington was "the unusual case in which the question whether a constitutional right has been violated and the question whether that right was clearly established are essentially coincident." Id. at 45. This proposition is in some tension with the Supreme Court's view that qualified immunity leaves "ample room for mistaken judgments," Malley, 475 U.S. at 343, 106 S.Ct. at 1097, and with later cases in this circuit. Veilleux v. Perschau, 101 F.3d 1, 3 (1st Cir.1996) (en banc) (per curiam); Roy v. Inhabitants of the City of Lewiston, 42 F.3d 691, 695 (1st Cir.1994).

Indeed, in Harrington itself, the court's treatment of qualified immunity and the merits as coextensive was attacked by a petition for rehearing urging that the court had not faced up to the question whether the right was "clearly established." The court then wrote a memorandum denying rehearing, asserting that the record on summary judgment was insufficient to decide whether the device was an impermissible intrusion and whether the city manager could reasonably have believed it was permissible. See Harrington, 977 F.2d at 46.

Whatever the state of the record in Harrington--and the opinion reveals nothing about what evidence had been offered by either side--we think the plaintiffs' own expert in this case virtually conceded that the use of the test, even for screening, is a matter of debate and professional judgment. Further, Harrington itself did not condemn the procedure outright, so it is doubly difficult to describe its use by O'Donohue as shocking the conscience or unreasonable under clearly established law. This is especially so since the basis for any required testing in Harrington was far slimmer than Berthiaume's guilty plea and confessed conduct.

One final point remains. We have thus far appeared to ignore Berthiaume's purported consent to O'Donohue's examination, including the plethysmograph test. Berthiaume, despite his objections, gave a written consent and had the advantage of a lawyer's advice. While the Board was threatening Berthiaume's license, it would have been easy enough to refuse the test and seek an immediate injunction against the Board's attempt to revoke the license or compel the test. Ordinarily, consent would end the matter.

Nevertheless, Berthiaume has made allegations of duress and said he had inadequate information about the nature of the test, and it might be difficult to resolve these issues on the existing record. Thus, we have assumed arguendo that Berthiaume's claim is not barred by his consent. Still, the appearance of consent is highly pertinent. Forcible administration of a test of this kind would be an entirely different case.

We conclude that the defendants are entitled to qualified immunity on the claims under the federal Constitution. As the federal claims must be dismissed prior to trial, and there is no independent basis for jurisdiction over the state claims, the state claims should be dismissed without prejudice. See, e.g., Martinez v. Colon, 54 F.3d 980 (1st Cir.1995); Brennan v. Hendrigan, 888 F.2d 189 (1st Cir.1989).

Reversed.