POSNER, Circuit Judge.
Twice in 1987 Carlos Villanova was civilly committed against his will to Chicago Read Mental Health Center, a state facility. Claiming that the commitment was an unreasonable seizure of his person and therefore violated his rights under the Fourth and Fourteenth Amendment and that the failure to release him within 24 hours, as required by state law, violated his Fourteenth Amendment right to due process of law, he brought this suit under 42 U.S.C. § 1983 against the psychiatrists employed by Chicago Read and by another public body, the Psychiatric Institute of Cook County, who he claims were responsible for his commitments. There is a little more to the suit but this is enough for our purposes. The judge dismissed it on the pleadings.
Illinois' mental health code authorizes involuntary commitment of "a person who is mentally ill and who because of his illness is reasonably expected to inflict serious physical harm upon himself or another in the near future" or is unable to care for himself. Ill.Rev.Stat. ch. 91½, ¶ 1-119. The plaintiff does not argue that this standard violates the Constitution. Nor that the statutory procedures for applying the standard violate the Constitution: before a person can be committed, there must be a petition stating that commitment is proper under the statutory standard and a certificate, signed by a mental-health professional who has examined the person sought to be committed, also stating that the commitment is proper under the standard. These documents allow commitment for only 24 hours. For the commitment to be extended beyond that, a different mental health professional must sign a second certificate. Ill.Rev.Stat. ch. 91½, ¶¶ 3-601 through 3-604, 3-610.
The record is scanty, as one expects when a case is decided on the pleadings. At the time of the first commitment Villanova was in jail, apparently awaiting trial for theft, when he requested a psychiatric examination to determine his fitness to stand trial. The judge ordered him sent to the Psychiatric Institute for an examination. A petition to commit Villanova was drawn up a few days after the order. It is apparent from the language of the petition that the author was a psychiatrist or other mental health professional, doubtless employed by the Psychiatric Institute, but beyond this the author's identity is unknown. The petition states that Villanova "presents today with tangential irrelevant responses to inquiries. He is superficially oriented, quite grandiose, and has delusions of thought projection. Insight is lacking and judgment is poor. He is currently unable to care for himself and is a danger to self and others. He is in need of inpatient hospitalization." A psychiatrist at the Institute
The plaintiff's lawyer asks us to analogize a petition for commitment to an application for an arrest warrant, a certificate of commitment to the warrant itself, and the defendants who filled out and signed the certificates in this case to the arresting officers. Where all this analogizing gets Villanova we have no idea; and the analogy is in any event impossibly strained because in a civil commitment there is nobody who corresponds to the magistrate who issues a warrant.
We analyze the case in more straightforward terms. A civil commitment is a seizure, and may be made only upon probable cause, that is, only if there are reasonable grounds for believing that the person seized is subject to seizure under the governing legal standard, not here challenged. Chathas v. Smith, 884 F.2d 980, 987 (7th Cir.1989); Gooden v. Howard County, 954 F.2d 960, 968 (4th Cir.1992) (en banc); Baltz v. Shelley, 661 F.Supp. 169, 178 n. 36 (N.D.Ill.1987). There is no requirement of a warrant issued by a judicial officer. For that matter, even an arrest warrant is required only when a person is to be arrested in his home. Payton v. New York, 445 U.S. 573, 583-90, 100 S.Ct. 1371, 1378-82, 63 L.Ed.2d 639 (1980); McKinney v. George, 726 F.2d 1183, 1188 (7th Cir.1984).
There was probable cause. Villanova himself or his counsel believed he was sufficiently mentally disturbed not to be fit to stand trial. The Psychiatric Institute confirmed that he had symptoms of mental illness, and opined that he was dangerous to himself or others. Since he was not only awaiting trial on a criminal charge but had a record of violent crime — robbery is a crime of violence, and we do not understand Villanova to be denying the truth of the statement in the first certificate that he had been imprisoned for robbery — an inference of dangerousness from his display of symptoms of mental illness was a reasonable one that mental health professionals whose competence and credentials are not challenged were entitled to draw. Villanova's lawyer makes fun of the terms that
"For a brief period...." The shorter the detention, the less compelling is the evidence of the necessity for it that the authorities need to produce. This is the theory of the "Terry stop" and of much else in Fourth Amendment law. Delaware v. Prouse, 440 U.S. 648, 663, 99 S.Ct. 1391, 1401, 59 L.Ed.2d 660 (1979); United States v. Martinez-Fuerte, 428 U.S. 543, 557-60, 96 S.Ct. 3074, 3082-84, 49 L.Ed.2d 1116 (1976); United States v. Chaidez, 919 F.2d 1193, 1197-98 (7th Cir.1990). In mathematical terms, the test of a reasonable commitment can be expressed by the inequality C>PH, where C is the cost of confinement to the person confined, H is the harm he might do if released, and P is the probability of his doing that harm if released. The resemblance to Learned Hand's famous formula for negligence (B>PL, where B is the burden of precautions, L the loss if there is an accident that the precautions could have prevented, and P the probability of an accident if the precautions are not taken, United States v. Carroll Towing Co., 159 F.2d 169, 173 (2d Cir.1947); see United States Fidelity & Guaranty Co. v. Jadranska Slobodna Plovidba, 683 F.2d 1022 (7th Cir.1982)) is no accident. The test of negligence at common law and of an unlawful search or seizure challenged under the Fourth Amendment is the same: unreasonableness in the circumstances.
An algebraic formulation of legal rules, which has been used not only to define negligence but also to summarize the standard for granting or denying preliminary injunctions, American Hospital Supply Corp. v. Hospital Products Ltd., 780 F.2d 589, 593-94 (7th Cir.1985), has value in expressing rules compactly, in clarifying complex relationships, in identifying parallels between diverse legal doctrines, and in directing attention to relevant variables that might otherwise be overlooked. It is not, however, a panacea for the travails of judicial decision-making. In practice, the application of standards that can be expressed in algebraic terms still requires the exercise of judgment, implying elements of inescapable subjectivity and intuition in the decisional calculus. Abbott Laboratories v. Mead Johnson & Co., 971 F.2d 6, 12 (7th Cir.1992).
Now it is obvious (returning to our formula, duly qualified by the remarks in the preceding paragraph) that the higher is P, the probability that the person sought to be committed would do harm if allowed to remain free, the stronger is the case for commitment. That is the variable that most of the cases stress. Chaidez, however, in the course of explicitly adopting the sliding-scale approach sketched above, 919 F.2d at 1199, adds that the longer the detention and hence the higher C is, the higher must P be in order to justify the detention. Chaidez also points out, citing Llaguno v. Mingey, 763 F.2d 1560, 1565 (7th Cir.1985) (en banc) (plurality opinion), that the higher is H, the magnitude of the harm the person may do if left at large, the stronger is the case for commitment. 919 F.2d at 1199. A person who might poison a city's water supply if left at large is more dangerous than one who with the same probability would merely slash his own wrists; so the case for committing the former is stronger. Despite the reservations about "a multifactor balancing test of `reasonable
C — the cost of confinement to the person confined — is smaller the shorter the detention, and the smaller C is the more likely it is to be outweighed by the harm (discounted by the probability of its occurrence) if commitment is denied, i.e., by PH. While it is clear that the evidence of dangerousness in this case was sufficient to warrant a 24-hour commitment, it is much less clear that it was sufficient to warrant a significantly longer one — and we don't know how long Villanova was committed to Chicago Read on either occasion. It is true that often when issues of duration are in play, the legal pigeonhole changes from the Fourth Amendment to the due process clause. The seizure is complete upon arrest, and the Fourth Amendment falls away. Wilkins v. May, 872 F.2d 190 (7th Cir.1989). The deprivation of liberty brought about by confining the arrestee, or a person civilly committed, is tested under the due process clause. United States v. Salerno, 481 U.S. 739, 746, 107 S.Ct. 2095, 2101, 95 L.Ed.2d 697 (1987); Addington v. Texas, 441 U.S. 418, 425, 99 S.Ct. 1804, 1809, 60 L.Ed.2d 323 (1979). However, Gerstein v. Pugh, 420 U.S. 103, 95 S.Ct. 854, 43 L.Ed.2d 54 (1975), had analyzed the question under the Fourth Amendment, and while subsequent cases may have seemed to abandon that approach, County of Riverside v. McLaughlin, ___ U.S. ___, 111 S.Ct. 1661, 114 L.Ed.2d 49 (1991), shows that it lives. The two lines of cases can be reconciled by recognizing that the Fourth Amendment governs the period of confinement between arrest without a warrant and the preliminary hearing at which a determination of probable cause is made, while due process regulates the period of confinement after the initial determination of probable cause. The cases are generally though not uniformly consistent with this distinction. Compare Coleman v. Frantz, 754 F.2d 719, 721-25 (7th Cir.1985), and United States v. Fernandez-Guzman, 577 F.2d 1093, 1097-98 (7th Cir.1978), with Sivard v. Pulaski County, 959 F.2d 662, 666-67 (7th Cir. 1992).
Does the distinction between the two legal categories (search and seizure, and due process) make any practical difference? Probably not, since whether the issue of duration is addressed under the Fourth Amendment or under the due process clause, the benefits of confinement to the government must be compared with the costs to the person confined. Mathews v. Eldridge, 424 U.S. 319, 335, 96 S.Ct. 893, 903, 47 L.Ed.2d 18 (1976). It is true that Gerstein v. Pugh and County of Riverside v. McLaughlin establish in the name of the Fourth Amendment a 48-hour limit on confinement prior to a determination of probable cause and that there is no corresponding hard-and-fast deadline for subsequent confinement challenged under the due process clause. But the difference, as it seems to us, is not between the Fourth Amendment and the due process clause. It is between the situation in which a person is being held pursuant to a judicial determination, and a situation in which he is being held without such a determination. Naturally there should be a tighter limit on confinement in the latter situation; the probability of error is greater, and needlessly so because there was time to have held a hearing and obtained a judicial determination.
There are additional obstacles to this suit. One is the Parratt doctrine, that in the case of random and unauthorized deprivations of liberty or property by public employees a state provides all the process that is due by furnishing adequate state judicial remedies for the deprivation. On the scope of the doctrine, see Zinermon v. Burch, 494 U.S. 113, 110 S.Ct. 975, 108 L.Ed.2d 100 (1990), and Easter House v. Felder, 910 F.2d 1387, 1400-05 (7th Cir. 1990) (en banc). Furthermore, as far as we can make out the only significance to Villanova of the fact that he was not released after 24 hours failing the issuance of the second certificate required by the statute is that he thinks this requirement of state law gave him a "liberty" interest in freedom after 24 hours that he can enforce under the due process clause of the Fourteenth Amendment. This belief reflects the persistent fallacy that procedural requirements create substantive entitlements, although the Supreme Court and this court keep insisting they don't. See Smith v. Shettle, 946 F.2d 1250, 1254 (7th Cir.1991), and cases cited there.
This point may bear amplification. If a state provides, as Illinois has done, that a person cannot be committed who is not a menace to himself or to others, this creates a "liberty" interest of which the state cannot deprive a person without giving him due process of law. But if it specifies the particular process — one, two, or fifty certificates of dangerousness — to which the committed person is entitled, it creates no additional liberty interest. State law is a source of substantive entitlements to property and liberty that the due process clause protects, though not the only source; the clause protects life and natural liberty, as well as state-created rights. Jones v. United States, 463 U.S. 354, 361, 103 S.Ct. 3043, 3048, 77 L.Ed.2d 694 (1983). But it is not the source of the procedural entitlements that constitute due process of law; the clause itself prescribes the procedures that the holder of the substantive entitlements that the clause protects is entitled to demand. Smith v. Shettle, supra, 946 F.2d at 1254.
The question Villanova should have raised, therefore, is whether the due process clause requires a second certificate in a civil commitment case within 24 hours of the initial commitment. But the answer would be no, so he would still lose. No doubt some provision must be made for reexamining the committed person's condition. He cannot be committed for life on the basis of psychiatric determinations made at the outset of the commitment. Psychiatry is not so exact a predictive science. But neither is the due process clause a straitjacket. Id. at 1255; cf. County of Riverside v. McLaughlin, supra, ___ U.S. at ___-___, ___, 111 S.Ct. at 1668-69, 1671. It leaves the details of the procedure for continuing evaluation of the committed person's condition to the states. If 48 hours is not too long to detain a person on a police officer's say-so, we do not think that 24 hours is too long to detain a person on the basis of a certification of two psychiatrists or other mental health professionals that he is an imminent danger to himself or others.
We note one more fallacy in the plaintiff's submission. He argues that the only thing that counts in determining probable cause is what the defendants who signed the commitment certificates knew. The contents of the petitions for commitment are therefore irrelevant, because there is no indication that the psychiatrists who signed the certificates had read the petitions, thus rendering inapplicable the doctrine that "when a group of officers is
The statute requires presentation of both the petition and the certificate to the "facility director" (e.g., the director of Chicago Read), Ill.Rev.Stat. ch. 91½, ¶¶ 3-601(a), 3-602, and if he reads both, this establishes the path of communication required by the doctrine of collective knowledge. We don't know whether he read both here, but it doesn't matter; communication was not required. We find no constitutional violation. Other issues are raised but none that matters in light of the discussion in this opinion.