The Police Coercion Requirement

The Police Coercion Requirement

The factors covered so far treat a suspect's age, schooling, emotional history, and mental condition as circumstances that help decide whether a confession was voluntary, and the totality test counts the suspect's mental and physical condition alongside the setting, duration, and manner of the interrogation.¹ A reader could take that list to mean that a fragile enough suspect makes any confession involuntary, whatever the police did. Colorado v. Connelly closes off that reading by holding that only official compulsion by the state makes a confession involuntary for Fourteenth Amendment purposes, so a confession is not involuntary merely because it results from the defendant's mental illness or lack of free will when the police did not coerce it.²

The limit refines the earlier factors without repudiating them. A suspect's characteristics still measure how much pressure it takes to overbear a will, but they cannot supply the pressure themselves. Two reasons sit behind that. The due process guarantee restrains the government, so a claim under it needs some act of the government to complain about. And the remedy of exclusion exists to discourage unconstitutional police conduct, which leaves it nothing to do where the police did nothing to bring the statement about. Reading the clause to reach confessions the police never induced would stretch both the guarantee and its remedy past their purposes. The three parts below take the rule itself, then the role it leaves for mental condition, then its corollary for pressure applied by private people.

Police Overreaching Required

The facts of Colorado v. Connelly show the rule at work. Connelly, a man with chronic schizophrenia, walked up to a uniformed officer and, without any prompting, said that he had killed someone. Psychiatric testimony later described a psychosis in which hallucinated voices commanded him to confess, and the Colorado courts suppressed his statements because the illness had impaired his capacity to make a free and rational choice.³ The Supreme Court reversed. It held that "coercive police activity is a necessary predicate to the finding that a confession is not 'voluntary'" under the Due Process Clause of the Fourteenth Amendment.⁴ Without police conduct causally related to the confession, no state actor has deprived the defendant of due process, and the inquiry ends there.⁵

That holding was drawn from a long line of decisions rather than announced fresh. Connelly traced the pattern across the roughly fifty years since Brown v. Mississippi, which held that confessions extracted by brutal torture violate due process, and found police overreaching in the cases that had held confessions involuntary.⁶ Spano v. New York belongs to that line. Spano's limited schooling and emotional instability counted against the confession because many officers questioned him through the night while a friend on the force played on his sympathy, so the condition and the police conduct worked together.⁷ Connelly's illness was serious, yet no police pressure acted on it, and that absence decided the case.

The requirement has a textual echo in the Fifth Amendment, which protects against being "compelled in any criminal case to be a witness against himself," and compulsion is something one party does to another.⁸ Connelly accordingly described the concern of the Fifth Amendment, the basis of the Miranda rule, as governmental coercion alone.⁹ Both sources of the voluntariness rule therefore begin with the same threshold question, which is whether the state pressed the suspect.

A fair objection is that a confession produced by psychosis may simply be false. The Court treated that worry as real but placed it elsewhere, holding that the unreliability of a statement is a matter for the evidence law of the forum and not for the Due Process Clause.¹⁰ The functional reason is that due process voluntariness polices what the government does to suspects, and a reliability test would turn the clause into a general rule of evidence. The incentive reason is subtler. Exclusion changes police behavior only when police behavior caused the statement, so suppressing a confession the police did nothing to induce would cost the prosecution its evidence while teaching officers nothing, since no different conduct on their part would have prevented it.

On an exam stem that describes a delusional or psychotic suspect, the common error is to treat mental illness alone as enough to make the confession involuntary. The illness belongs in the analysis only after something the police did has been identified for it to act upon.

Mental Condition's Limited Role

Connelly did not remove mental condition from the voluntariness analysis. The Court acknowledged that as interrogators have turned to subtler forms of psychological persuasion, courts have treated the defendant's mental condition as a more significant factor in judging voluntariness. It then held that mental condition, by itself and apart from its relation to official coercion, should never dispose of the constitutional inquiry.¹² The condition therefore keeps a real job, which is to show how easily a given amount of police pressure overcomes a given suspect.

The relationship works like multiplication. Susceptibility determines how much a unit of pressure accomplishes, and pressure determines whether there is anything for susceptibility to enlarge. A suspect who is young, poorly schooled, or emotionally unstable will yield to questioning that a hardier person would resist, which is why the totality test lists the suspect's mental and physical condition among the circumstances to be weighed.¹³ But a very large susceptibility multiplied by no police pressure is still nothing, and that is the situation Connelly decided. Because the weighing is cumulative, the scale slides in both directions: moderate pressure can become coercive against a severe condition, and intense pressure can overbear a suspect whose condition is modest.

A defendant whose illness makes a confession unreliable is not left without recourse, though the recourse lies outside this inquiry. The reliability of the statement is governed by the forum's evidence law and tested before the jury, not by the Due Process Clause.¹⁴

The pair differs in one variable, the presence of police pressure, and that variable moves the answer even though the diagnosis is identical. The tempting argument on the second set of facts is that the illness alone settles the matter, which fails for the same reason as in Connelly. The argument that succeeds ties the illness to the detectives' conduct and asks how that conduct acted on a suspect in that condition.

Private Pressure Falls Outside

The requirement of official compulsion has a corollary. If due process asks whether the state compelled the statement, then pressure from anyone other than the state cannot supply what the clause demands. Connelly put the point in unqualified terms: "The most outrageous behavior by a private party seeking to secure evidence against a defendant does not make that evidence inadmissible under the Due Process Clause."¹⁶ The hallucinated voices that drove Connelly to confess were a form of pressure that came from within, and the Court treated them the way it treats a threat from a private person: neither is the state's doing, so neither engages the clause.

The reason is the same one that supports the police-overreaching requirement. The guarantee restrains the government, and exclusion deters police misconduct, so when officers took no part in producing the statement there is neither a violation to remedy nor behavior to deter. The corollary does not make private coercion legally harmless. The private actor may face criminal or civil consequences, and the forum's evidence rules may still screen out an unreliable statement. The point is only that the Due Process Clause is not the tool for excluding it.

The twin shows what carries the result, which is whether the state stands behind the pressure. A private person who acts at police direction is the state for this purpose, since the rule asks who compelled the statement and not whose name is on the badge. Students lose this issue when they treat the brutality of the private conduct as the test, because how outrageous the pressure was matters only after the state is shown to be responsible for it.

Police Overreaching Required

The requirement of police overreaching is easiest to see in the case that announced it. In Colorado v. Connelly the defendant approached a Denver police officer, said he had killed someone, and said he wanted to talk about it. After being advised of his Miranda rights he waived them and described where the body of the victim, a young girl, could be found.¹⁸ A psychiatrist later testified that Connelly was a chronic schizophrenic in a psychotic state, that the psychosis motivated his confession, and that he could not make a free and intelligent decision to confess.¹⁹ On that record the Colorado courts held the statements involuntary and held that admitting them violated due process. The Supreme Court reversed. It held that only official compulsion by the state makes a confession involuntary for Fourteenth Amendment purposes, so a confession is not involuntary merely because it results from mental illness or a lack of free will when the police did not coerce it.²⁰

The fact that decided the case was a finding about the police. The Colorado trial court found that the police committed no wrongful acts, the defendant never challenged that finding, and the Colorado Supreme Court left it undisturbed.²¹ That court nonetheless found enough state action in the fact that a state court had admitted the confession into evidence. The Supreme Court rejected that reasoning because it overlooks the essential link between coercive activity of the State and the resulting confession.²² If admitting a statement at trial were itself the required state action, the State would be involved in every confession ever offered, and the inquiry would collapse into a search for the defendant's private motives. Courts would have to divine why a defendant chose to speak even where no one claims that government conduct coerced the choice.²³ The link therefore has to run from something the police did to the statement the defendant made.

Connelly presented that requirement as the product of a long line of decisions rather than a new departure. The line begins with Brown v. Mississippi, where police extracted confessions through brutal torture, and the Court condemned the conduct as "revolting to the sense of justice" even though the Fifth Amendment privilege did not yet bind the States.²⁴ The Court kept that due process focus after Malloy v. Hogan applied the privilege to the States, and the cases decided in the fifty years since Brown have, in Connelly's account, focused on "the crucial element of police overreaching."²⁵ Each confession case turned on its own set of factors, but every one contained a substantial element of coercive police conduct.²⁶ Brown therefore marks the pole where the requirement is plainly met, since beating suspects supplies exactly the coercion the cases demand. Connelly marks the pole where it is plainly missing, since a confession driven entirely by the defendant's own mental state leaves the due process claim without a state actor to point to.

The mental-condition cases the defendant relied on show the same pattern. In Blackburn v. Alabama the defendant was probably insane when he confessed, and the police learned during the interrogation that he had a history of mental problems. They then exploited the weakness with an eight- to nine-hour sustained interrogation in a tiny room sometimes filled with officers, without his friends, relatives, or counsel, and with a confession composed by a deputy sheriff rather than by him.²⁷ In Townsend v. Sain a police physician gave the suspect a drug with truth-serum properties, and the officers who then obtained the confession knew of the drug.²⁸ Both cases involved a mentally compromised suspect, as Connelly did. What separates them from Connelly is that in each one police conduct acted on the compromised condition. Mental condition is therefore relevant to how susceptible a suspect is to coercion, but an examination of the confessant's state of mind can never by itself conclude the due process inquiry.²⁹ Spano v. New York fits the same frame, because mental condition became a more significant factor as interrogators turned to subtler forms of psychological persuasion, and the persuasion was the police's own.³⁰

Police overreaching is a broad idea and does not require blows. In Arizona v. Fulminante a prisoner confessed to a fellow inmate who was actually a government agent, after the inmate suggested he could protect the prisoner from rough treatment by other inmates only if the prisoner told the truth. The Court held that coercion need not depend on actual violence by a government agent, that a credible threat of physical violence is enough, and that coercion can be mental as well as physical.³¹ The label also reaches beyond uniformed officers, since a government employee need not be a law enforcement official for questioning to implicate the Fifth Amendment.³² What matters is that the State, through someone acting for it, applied the pressure.

The dissent in Connelly would have decided the case the other way, and its reasoning is worth knowing because it is the competing theory of what makes a confession involuntary. The dissent argued that the absence of police wrongdoing should not by itself decide the voluntariness of a confession by a mentally ill person, because the voluntariness requirement reflects the importance of both free will and reliability and calls for an inquiry into the totality of the circumstances.³³ It read the majority as restricting "involuntary" to police-coerced confessions, which refuses to treat free will as a value of constitutional consequence. It relied on the older test of whether the confession is the product of an essentially free and unconstrained choice, and on the idea that the line falls where governing self-direction is lost and "compulsion, of whatever nature or however infused," propels the confession.³⁴ On the dissent's account, police overreaching had been present in every earlier case only because every earlier case happened to involve it, so the free-will concern was an independent one.³⁵

That view is not the law. The majority answered that sustaining the defendant's claim would require a brand new constitutional right, "the right of a criminal defendant to confess to his crime only when totally rational and properly motivated."³⁶ The two opinions disagree about what the constitutional wrong is. The dissent locates it in the confessor's lost freedom, whatever its source, while the majority locates it in the State's act of forcing a statement out of someone. The majority view controls, so an exam answer applies the police-overreaching requirement and mentions the free-will theory only to reject it.

Each half of the requirement has its own logic. The demand for coercive police conduct follows from the premise that the Due Process Clause restrains the government, a premise the Court said the voluntariness cases had always honored.³⁷ It also fits the purpose of exclusion, which is to deter future constitutional violations. Suppressing a statement the police did nothing wrong to obtain would serve no purpose in enforcing constitutional guarantees while still costing the prosecution relevant evidence.³⁸ The causal-connection half does subtler work. It ties the remedy to the misconduct that actually produced the statement, so that pressure the police applied cannot taint a confession it did not bring about, and a confession cannot be excluded merely because the police were somewhere in the picture. Courts applying Connelly accordingly treat coercive police conduct as the predicate and then weigh the pressure against the suspect's capacity to resist it.³⁹

The concern that a confession produced by psychosis may simply be false has a different home. The Court acknowledged that a statement from someone in the defendant's condition might prove quite unreliable, but it assigned that problem to the evidence law of the forum, because due process aims at fundamental fairness in the use of evidence, whether true or false.⁴⁰ States do supply such a route. Texas, for example, resolves claims that a confession was involuntary because of the defendant's state of mind under its Confession Statute rather than under the Due Process Clause.⁴¹

The twin isolates the variable that moves the answer. The suspects' internal states are the same, and only the presence of state pressure differs. On an exam stem the common error is to treat mental illness or a lack of free will standing alone as enough to suppress a confession. The illness or distress enters the analysis only after something the police did has been identified for it to act upon.

Mental Condition's Limited Role

Earlier in this chapter, the suspect's own characteristics (age, schooling, emotional history) were treated as a factor that sets how much pressure it takes to overbear a will. That factor survives Colorado v. Connelly, but the case attached a condition to it. The Court agreed that "mental condition is surely relevant to an individual's susceptibility to police coercion," and it added that examining the confessant's state of mind can never by itself conclude the due process inquiry.⁴³ Condition therefore operates as a multiplier on police pressure, and a multiplier applied to no pressure at all yields nothing. Before Connelly, any fact suggesting the defendant was more susceptible than the average person counted toward whether this defendant's free choice was impaired. Since Connelly, the suspect's characteristics cannot dispose of the inquiry in isolation, and their relevance depends on the police conduct they interact with.⁴⁴

Each half of that arrangement has its own reason. Condition matters because resistance to pressure differs from one person to the next: hours of questioning that a rested adult might withstand can leave a wounded or severely ill suspect unable to hold out, so one tactic may overbear one will and leave another untouched. Condition cannot decide the case alone because the constitutional wrong is state conduct, and suppressing a statement the police did nothing wrong to obtain would deter no future violation.⁴⁵ A contrary rule would also leave officers with nothing to correct, since an undetected illness would put any confession at risk however carefully the interrogation was run. The law ties the condition back to what officers can control by judging their techniques in light of what they knew or should have known about the suspect's ability to comprehend what was happening.⁴⁶ Where officers are unaware of an illness and conduct an interrogation that complies with Miranda, obtaining a confession does not violate due process.⁴⁷

Two decisions show the combination at work. In Blackburn v. Alabama the suspect was probably insane when he confessed, and the police learned during the interrogation that he had a history of mental problems. They then kept up an interrogation of eight to nine hours in a tiny room that was at times filled with officers, with none of his friends, relatives, or counsel present, and a deputy sheriff composed the confession rather than the suspect.⁴⁸ The impairment made the suspect easy to break and the tactics exploited that ease, which is why the confession was involuntary. In Townsend v. Sain a police physician gave the suspect a drug with truth-serum properties, and the officers who then obtained the confession knew he had been given drugs.⁴⁹ Connelly later described Townsend as police wrongdoing, even though the earlier opinion had proceeded on the assumption that neither the physician nor the interrogators knew the drug's truth-serum character, which indicates that an impairment produced by the state's own agents counts as state conduct.⁵⁰

Connelly supplies the opposite pole with a comparably impaired suspect. He was a chronic schizophrenic in a psychotic state, and his psychosis motivated the confession.⁵¹ The officers responded to his approach by reciting the Miranda warnings, and although he mentioned past stays in mental hospitals, he displayed no signs of illness during the interrogation.⁵² The trial court found that the police committed no wrongful acts, and that finding was never disturbed.⁵³ Blackburn and Connelly therefore differ in one variable. Each suspect was seriously impaired, but only in Blackburn did police conduct act on the impairment. Connelly's condition was real and it caused his confession, yet a cause located entirely inside the suspect has no state actor attached to it.⁵⁴

The twin changes one fact, what the detective did, and the outcome moves with it while the suspect's condition stays constant. That is the same variable that separated Blackburn from Connelly. Students lose this issue when they treat a sympathetic mental condition as the end of the analysis, because a condition can magnify pressure but cannot supply it. The court still has to find coercive conduct by the state before the condition adds anything.

The pairing also covers a weakened suspect whose weakness is physical. In Mincey v. Arizona the defendant had been seriously wounded a few hours before the questioning, was in intensive care, complained that the pain was unbearable, and gave some answers that were not coherent, all while lying on a hospital bed encumbered by tubes, needles, and breathing apparatus.⁵⁵ The detective did not beat or drug him. But Mincey said more than once that he wanted a lawyer and asked the detective to stop, and the detective paused only while Mincey lost consciousness or received treatment, returning to the questioning each time.⁵⁶ The Court held that none of the gross abuses seen in other cases, such as beatings or truth serums, was needed, because the circumstances as a whole showed that his will was overborne and that the statements could not be used in any way against him at trial.⁵⁷ Mincey is thus the weakened-suspect counterpart to Blackburn: a lowered capacity to resist met police conduct that took advantage of it.

Courts also look for evidence that the condition actually operated on the decision to speak, rather than assuming it did. In People v. Knapp a New York appellate court found a confession involuntary where the defendant had an IQ of 68, was highly suggestible, and was interrogated by a detective skilled in getting people to talk.⁵⁸ In Williams v. State a Texas appellate court reached the opposite result for a defendant who described delusions about fighting the devil, because the video showed that he understood the officers, answered coherently, and had himself asked to speak with investigators.⁵⁹ The trial court there found no coercive or improper police conduct, and the appellate court treated mental illness as only one characteristic among many.⁶⁰ The same attention to susceptibility appears in Berghuis v. Thompkins, where the voluntariness holding rested partly on the absence of any evidence that the suspect was particularly susceptible to coercion.⁶¹ Whether illness also defeats a Miranda waiver is a separate question that the later section on Miranda takes up.

Private Pressure Falls Outside

The police-overreaching requirement has a corollary that follows from who the Fourteenth Amendment restrains. Because the Due Process Clause limits the government, pressure that originates with someone other than the state cannot make a confession involuntary in the constitutional sense.⁶³ In Colorado v. Connelly the Court described its involuntary-confession cases as consistent with the settled rule that a due process claim requires some form of state action, meaning conduct attributable to the government. It added that the most outrageous behavior by a private party seeking to secure evidence against a defendant does not make that evidence inadmissible under the Due Process Clause.⁶⁴ The first question is therefore who applied the pressure, and the harshness of the pressure matters only after the answer is the state.

The reason for the rule lies in the cost and purpose of exclusion. Keeping a confession out of evidence removes relevant proof from the prosecution's case, and that cost is justified only when exclusion discourages future constitutional violations by the government.⁶⁵ Deterrence works on actors who can change their conduct in response to the penalty, and the police are such actors only when their own conduct produced the statement. Coercion by a private party gives officers nothing to correct, so suppressing the statement would cost the prosecution evidence and prevent no violation. That is why Connelly found that suppression in such a case would serve no purpose in enforcing constitutional guarantees.⁶⁶

A tempting alternative is to locate the state action in the courtroom rather than the interrogation room. The Colorado trial court had found that the police committed no wrongful acts, and that finding was never disturbed, so the Colorado Supreme Court reasoned instead that state action was present because the confession was admitted into evidence in a court of the State.⁶⁷ Connelly rejected that theory because it fails to recognize the essential link between coercive activity by the State and the confession that results.⁶⁸ Every confession a prosecutor offers is offered in a state courtroom, so under the Colorado court's view the state action element would be met automatically in every case and would screen out nothing. The claim would then become an inquiry into why the defendant chose to speak, with no showing that the government coerced the choice, and sustaining it would require a new constitutional right to confess only when totally rational and properly motivated.⁶⁹

The actor that must be identified is the state, which is a wider category than uniformed police. A government employee need not be a law enforcement officer for questioning to implicate the Fifth Amendment's strictures, so the phrase police overreaching in Connelly should not be read literally.⁷⁰ The rule screens out private persons; an official acting for the government stays inside the analysis even when that official is not a police officer.

The twin holds the threat constant and changes only the person who delivers it, which is the variable Connelly treats as decisive. In the first version the due process claim ends at the state-action requirement. In the second it proceeds to the totality analysis taught earlier in this chapter.