EASTERBROOK

Thomas C. Crooks (argued), Attorney, Chicago, IL, for Plaintiff–Appellant. Julie M. Koerner, Michael J. Victor (argued), Attorneys, O'Halloran, Kosoff, Geitner & Cook, Northbrook, IL, for Defendant–Appellee.

Before EASTERBROOK, Chief Judge, and POSNER and WOOD, Circuit Judges.

EASTERBROOK, Chief Judge.

Ruby Graham cashed some checks at a currency exchange and with $5,000 in her purse drove to the public library in Bellwood, Illinois. She was accompanied by her mother, Elizabeth Graham. Just as Ruby stepped through the library's door, her purse was grabbed from behind. Ruby turned and struggled with the snatcher. Elizabeth saw what was happening, yelled “no, no, not my baby,” and ran to give aid. The robber shot Elizabeth in the chest, then shot Ruby in the head, ran to the car, grabbed Elizabeth's purse, and fled. Elizabeth was seriously injured, but Ruby suffered only a graze to her temple. Both were taken to the hospital. Ruby was released after receiving a few staples.

Ruby described the attacker to two police officers—one at the library, another at the hospital. She did not tell either officer that she had known her assailant. The day after the robbery, officer Jimenez Allen went to Elizabeth's hospital room to question her. Allen found Ruby, along with James Bufkin, visiting Elizabeth. Bufkin told Allen that he had heard a rumor that Wydrick Phillips, who lived in the Grahams' neighborhood, had been staking out currency exchanges and robbing people who cashed tax-refund checks (as Ruby had done). Later that day Ruby examined photographs at the police station to see whether she could identify the robber. All photographs had been chosen to meet the descriptions Ruby gave. She spent about 15 minutes examining five or six sheets of photographs, each containing six pictures (but no names). She concentrated on the fourth sheet. She told Allen that she knew the sixth person on the sheet but did not think him the robber. Ruby also told Allen that the first photo on this sheet was similar to the robber—but eventually she selected the fifth person on this page and told Allen that she was “sure” that he was the culprit. Ruby had selected a photograph of Wydrick Phillips.

Police arrested Phillips, who was charged with robbery and attempted murder. He was acquitted when no evidence corroborated the testimony of Ruby and Elizabeth (both of whom picked Phillips from a lineup and identified him at trial). A search of Phillips's home did not turn up any proceeds or weapons; a test of the clothing that he had been wearing on the day of the robbery was negative for gunpowder residue; and the defense introduced a manifest of FedEx deliveries showing (if it was accurate) that Phillips, a courier for that company, could not have been at the Bellwood library when the robbery and shootings occurred. Phillips then filed this suit under 42 U.S.C. § 1983 against the Village of Bellwood and seven police officers who had participated in the investigation and prosecution, contending that he had been arrested without probable cause. The district court granted summary judgment to the defendants on this federal claim and relinquished supplemental jurisdiction of all state-law claims. 743 F.Supp.2d 931 (N.D.Ill.2010). The judge concluded that Ruby's selection of Phillips's picture established probable cause for his arrest.

Phillips's principal argument is that Allen spoiled the identification procedure by speaking with Bufkin where Ruby could overhear, which primed her to finger Phillips. This made her identification unreliable, Phillips insists. He advances several other arguments—such as a contention that the police should have done more to follow up other leads before arresting him. (Police received tips implicating three other persons. They conducted a second photo array including a picture of one of these three; Ruby did not select it. They did not show her pictures of the other two.) Phillips also contends that police should not have arrested him before his mother had a chance to show them a copy of the FedEx manifest that was produced as an alibi at trial. These are weak arguments. Police need not run down all leads before making an arrest—especially not when a crime is violent and leaving the perpetrator at large may endanger other persons. Nor need police wait for alibis, which even when presented they need not believe. A delivery manifest could be fabricated, or the times and locations of deliveries could be altered before the document is shown to police. See Hernandez v. Sheahan, 455 F.3d 772 (7th Cir.2006) (presenting prison officials with a document supposedly demonstrating innocence does not require the prisoner's immediate release). Probable cause is established by a reasonable belief that a person committed a crime. See Illinois v. Gates, 462 U.S. 213, 103 S.Ct. 2317, 76 L.Ed.2d 527 (1983). Police are entitled to leave to the criminal process the full examination of potential defenses. See, e.g., Hebron v. Touhy, 18 F.3d 421 (7th Cir.1994).

The only subject that requires extended discussion is Phillips's contention that Ruby's selection of his photo was the result of an unreliable process. Identification by a single eyewitness who lacks an apparent grudge against the accused person supplies probable cause for arrest. See, e.g., Gramenos v. Jewel Companies, Inc., 797 F.2d 432 (7th Cir.1986); Askew v. Chicago, 440 F.3d 894 (7th Cir.2006). But Phillips maintains that an identification should not count if it was the result of unduly suggestive procedures. He relies on decisions such as Neil v. Biggers, 409 U.S. 188, 93 S.Ct. 375, 34 L.Ed.2d 401 (1972), which holds that a person's selection of a suspect from a photo spread or lineup cannot be used as evidence at trial if the selection was the result of unduly suggestive procedures that create an excessive likelihood of irreparable misidentification. See also Simmons v. United States, 390 U.S. 377, 384, 88 S.Ct. 967, 19 L.Ed.2d 1247 (1968).

Biggers and similar decisions don't support Phillips's position directly. They concern the admissibility of evidence at criminal trials, not claims for damages against arresting officers. Similarly, decisions such as United States v. Johnson, 859 F.2d 1289, 1294 (7th Cir.1988), and United States ex rel. Hudson v. Brierton, 699 F.2d 917, 923–24 (7th Cir.1983), whose significance the parties debate, concern the admissibility of evidence at a criminal trial. But Phillips contends that the Biggers approach should be extended from trials to arrests, and from a rule of evidence to a rule of damages. We think that the proposed extension would be improvident—though with an important proviso that we discuss later.

The Supreme Court stressed in Gates that evidence need not be admissible at trial in order to support a finding of probable cause. A conclusion that there is probable cause for arrest (or indictment) just gets the criminal process started. Many later steps shape what evidence can be used at a trial and how much is required for conviction. Doctrines, such as that of Biggers, that are designed to reduce the chance of erroneous conviction do not have any greater function at the arrest stage than do doctrines such as the requirement of proof beyond a reasonable doubt, the defendant's entitlement to confront his accusers, or the hearsay rule. Probable cause is established, and arrests are made, without an adversarial presentation. Application of the Biggers framework is possible, however, only after evidence has been gathered and an adversarial hearing held.

Phillips wants such a proceeding held now and the results used to determine the validity of the arrest. But hindsight is not an appropriate basis for awarding damages against police. The validity of an arrest depends on what is known at the moment of the arrest, not on evidence that may be developed years later. Phillips believes that he can get damages without the benefit of hindsight. He tells us that officer Allen must have known that he was steering Ruby to pick Phillips and that a reasonable officer must have known that priming a witness in this way would produce an inaccurate selection. The first part of this argument is cut short by the rule that the officer's motives do not count; probable cause is objective. See, e.g., Ashcroft v. al–Kidd, ––– U.S. ––––, 131 S.Ct. 2074, 2080–83, 179 L.Ed.2d 1149 (2011); Devenpeck v. Alford, 543 U.S. 146, 153, 125 S.Ct. 588, 160 L.Ed.2d 537 (2004); Whren v. United States, 517 U.S. 806, 116 S.Ct. 1769, 135 L.Ed.2d 89 (1996). And the second part of Phillips's argument fares no better.

Whether mentioning a name and a rumor in a victim's presence influences a later identification is a difficult question of psychology, not something about which police officers (and lawyers) know the right answer instinctively. Bufkin's statements may have meant nothing to Ruby. She says that she knew neither the name nor Phillips's appearance; he contends, to the contrary, that she must have known his name and face because the two lived in the same neighborhood. In court, on a motion for summary judgment, we must assume that Phillips is right about this; but in a hospital room years ago, Allen was not bound to know that this was so. (Allen does not concede it even today, nor has Ruby ever admitted knowing plaintiff's name and appearance before she selected his photograph.)

Suppose Ruby is lying and did know Wydrick Phillips by name and appearance; would it follow that her knowledge of a tip that he had been robbing people would lead her to identify him as her assailant? Phillips's lawyer treats an affirmative answer as obvious, but nothing is obvious about the psychology of eyewitness identification. Indeed, one point well established in the psychology literature is that most people's intuitions on the subject of identification are wrong. See Christopher Chabris & Daniel Simons, The Invisible Gorilla: How Our Intuitions Deceive Us (2010). We held in United States v. Williams, 522 F.3d 809 (7th Cir.2008), that someone who contends that a particular kind of procedure led to an unreliable identification needs evidence—if not from an expert's affidavit, then from published work such as Elizabeth F. Loftus, et al., Eyewitness Testimony: Civil and Criminal (4th ed. 2007), the standard text in this field. Phillips has not referred us to such evidence; he has only a lawyer's confidence that what Allen did would have produced a worthless identification. Lawyers' talk is no substitute for data.

Indeed, the approach of Biggers itself has been questioned by social scientists. The American Psychological Association filed a brief as amicus curiae in Perry v. New Hampshire, ––– U.S. ––––, 132 S.Ct. 716, 181 L.Ed.2d 694 (2012), in which the Court declined to extend Biggers to acts by private parties that may influence eyewitnesses. The Association told the Justices:

In Biggers and Manson [ v. Brathwaite, 432 U.S. 98, 97 S.Ct. 2243, 53 L.Ed.2d 140 (1977) ], this Court enumerated five factors relevant to the probable accuracy of an eyewitness identification: “the opportunity of the witness to view the criminal at the time of the crime, the witness' degree of attention, the accuracy of his prior description of the criminal, the level of certainty demonstrated at the confrontation, and the time between the crime and the confrontation.” Manson, 432 U.S. at 114 (citing Biggers, 409 U.S. at 199–200 ). As shown by the discussion in the text, most of these factors are indeed relevant to probable accuracy—with the notable exception of witness certainty, see infra n.14. But given that notable exception, and given the plethora of other accuracy-related factors that researchers have identified since Biggers and Manson, APA urges the Court, in an appropriate case, to revisit the Manson framework so as to bring it in line with current scientific knowledge. APA Brief at 13 n.8. Footnote 14, to which note 8 refers, adds:

Jurors' evident belief that eyewitness confidence correlates with accurate identifications was once shared by many in the judiciary. Indeed, in Biggers this Court stated, albeit without citing any scientific authorities, that confidence is an indication of accuracy. See 409 U.S. at 199–200 . Subsequent research, however, has called this notion into very serious question. As one report concluded, “[t]he outcomes of empirical studies, reviews, and meta-analyses have converged on the conclusion that the confidence-accuracy relationship for eyewitness identification is weak, with average confidence-accuracy correlations generally estimated between little more than 0 and .29.” Brewer et al., The Confidence–Accuracy Relationship in Eyewitness Identification, 8 J. Experimental Psychol. Applied 44, 44–45 (2002). Even these various correlation figures are likely overestimates, moreover, because the confidence of eyewitnesses in actual cases, unlike in controlled experiments, may be infected by positive feedback received in the investigative process (for example, an officer stating during a photo array or line-up, “good, you identified the suspect”). [Citations omitted.] The Court bypassed this topic in Perry, leaving to the future any inquiry into the Biggers framework. Perry holds that in the main the validity of an eyewitness identification is for the jury—which implies that it is not tortious to obtain that identification in order to have evidence to present at trial. Neither the Supreme Court nor this court has held that, as a matter of law, mentioning a suspect's name spoils an identification. Without a solid basis in the social science of eyewitness identification, a court could not appropriately create such a rule. Phillips has not offered any basis, solid or otherwise, for us to do so.

We mentioned earlier the need for a proviso. It is this. Suppose a particular technique that officers may use to trick a person into making an unreliable identification has already been forbidden by an authoritative judicial decision (that is, by the Supreme Court or the court of appeals with territorial jurisdiction) and the officer uses it anyway. An officer who employs the forbidden technique in order to manufacture an identification can't complain when a court provides a remedy. Several courts of appeals have concluded that, when a state actor deliberately uses a forbidden technique to generate a false identification, an award of damages is permissible under 42 U.S.C. § 1983. See, e.g., Good v. Curtis, 601 F.3d 393, 398–99 (5th Cir.2010) (intentionally manipulating a lineup); Brodnicki v. Omaha, 75 F.3d 1261, 1265–66 (8th Cir.1996) (dictum; officer held not liable). This approach does not assist Phillips, however, because no judicial decision has held that mentioning a suspect's name in the hearing of a potential witness is improper. What Allen did was not remotely similar to showing a witness just one photo and asking her to “confirm” that the photo depicts the culprit.

And if, despite the lack of scientific support, we were to create the sort of rule that Phillips proposes, it could not do him any good. A rule of law devised after the events does not support an award of damages; officer Allen would be entitled to qualified immunity. See al–Kidd, 131 S.Ct. at 2083–85. It was not “clearly established” in February 2005, when Allen arrived in Elizabeth's hospital room, that mentioning a suspect's name in the presence of an eyewitness, and then showing that witness a photo spread containing the suspect's picture, violates the Constitution.

If Allen manipulated Ruby into selecting Phillips, he may have a remedy under state law. The district court dismissed all state-law claims without prejudice. But Phillips does not have a damages remedy for wrongful arrest under § 1983 and the fourth amendment. None of the parties' other arguments requires discussion.

Affirmed.