The Allegory Of The Cave! (Statistical Studies of the Impact of the Miranda Decision on Law Enforcement)
Criminal Law & Procedure Practice Group Newsletter - Volume 3, Issue 2, Summer 1999
August 1, 1999Gary D. Beatty
THIRTY YEARS after the United States Supreme Court decision in Miranda v Arizona1, the debate about its impact on law enforcement continues. That decision, intended to curtail police misconduct in securing confessions, recognized two principles: First, that obtaining confessions (from guilty suspects) is a police practice that should be encouraged. Second, that how police obtain a confession is more important than whether a guilty perpetrator is convicted.
Opponents of the decision say Miranda exemplifies judicial activism which hampers police efforts to fight crime. Supporters respond that any restrictions on law enforcement are negligible, and are constitutionally mandated. Advocates for both sides of this philosophical debate now claim there is statistical proof of their respective positions, however an examination of one such statistical study illustrates why philosophical differences can rarely be resolved by numbers.
In 1996 Professors Paul Cassell2, and Stephen Schulhofer3 participated in a symposium sponsored by the Harvard University Student Federalist Society.4 Cassell argues that comparison of police case arrest/clearance rates before, and after, Miranda proved the decision had an adverse impact on law enforcement. Schulhofer says those same rates proved there was no significant impact. Both exhibit considerable expertise in statistical analysis of the data. Each challenges the others' methodology. Both erroneously assume that any impact of the Miranda decision, negative or otherwise, will show up at the arrest stage of the criminal justice process.
The impact, if any, of Miranda on law enforcement won't appear in arrest/clearance rates. Miranda effects the admissibility of a confession at trial, not the use of a confession to clear a case by arrest. To accept either Cassell's, or Schulhofer's, thesis to be proven by their statistics, requires two assumptions:
- Police never arrest anyone based on a confession which is later suppressed because of a Miranda violation and,
- That confessions are always used as evidence, unless suppressed because of a Miranda violation; and that suppression thwarts the prosecution, thus causing police to alter their future conduct.
Ample empirical evidence invalidates the first assumption. The circumstances of individual cases disprove the second . By the time Miranda can have any true impact on the process, too many other factors are involved to be able to measure any trickle-down effect on police conduct. Those factors include decisions which limit Miranda, and how confessions are actually used in criminal trials.
I: Supreme Court Decisions Subsequent To Miranda
Courts have been reluctant to let confessed criminals hide behind Miranda "technicalities"5. For instance, New York v. Quarles6, held that public safety prevails over Court-created prophylactic niceties, even where a violation is flagrant.7 "Inevitable discovery"8, and use of suppressed confessions for impeachment9, are other instances. The scope of what can be suppressed because of a Miranda violation is not even clear.10
In this legal environment, police will continue to use confessions to clear cases, without trying to predict how some court may ultimately rule on the use of the confession as evidence. Courts can change the rules at any time, and the attitude of most police officers is that a suspect may "beat the rap," but he won't "beat the ride." Even though the confession that lead to the arrest may be eventually suppressed, the suspect still gets the ride to the jailhouse after the arrest.11
II: How Confessions Are Used
In order to measure if Miranda impacts arrest/clearance rates, it is necessary to determine if police altered their investigative practices out of concern for how Miranda, standing alone, will effect use of confessions as evidence in trials. The realities of criminal prosecutions skew any attempts to isolate Miranda's effect. Following are four examples. Each involved a challenge to the use of the defendant's confession.
The Mock Cases
Gary Allen Mock, and a co-defendant, were both arrested for two murders. Each confessed to some involvement in events of each death.12
In the first case, based on his confession, Mock was arrested then alternatively charged as a principal to First Degree Murder, and with being an Accessory After the Fact to that Murder.13 Mock's challenge to the use of his confession, based on Miranda, was denied. The jury heard his confession, and convicted him as an Accessory, but acquitted him of the homicide. The conviction was affirmed.14
In the second case15, police used the same confession, as the basis for his arrest. Mock was indicted for First Degree Murder. The State sought to use the confession in trial16. Again Mock moved to suppress his confession, claiming the same Miranda violation he previously raised unsuccessfully in the first case. The legal issues were the same, and the law had not changed. But the trial Judge in the first case retired, and a new Judge presided over the second case. The suppression motion was granted.17 The State proceeded to trial without the confession, and Mock was convicted of First Degree Murder.
Both murders were cleared by arrest, using the same confession. Two Judges made diametrically opposed rulings on the use of the confession at trial, each of which was affirmed. In the case where the State used the confession, Mock was acquitted of murder. Where the confession was suppressed because of Miranda, and the jury never heard it, he was convicted.18 The effect of Miranda can't be determined in either case, so how can it influence future police conduct?
* * *
Another obstacle to isolating the impact of Miranda occurs when prosecutors don't use a defendant's confession at trial. Before appellate courts can render opinions that induce police to modify their actions, the prosecution must seek to use a confession as evidence. If a confession is not used, police conduct in obtaining the confession is irrelevant to the outcome, and can not become a judicial vehicle to make law which effects future police conduct.
The Hoskins Case
Johnny Hoskins19 was arrested for a traffic violation while driving a stolen car. On each of the following four days he gave police a different custodial confession regarding the fate of the car's owner, including leading police to the owner's body. Using the confessions, the case was cleared by Hoskins' arrest for murder.
Prior to trial, Hoskins raised Miranda challenges to each of his four confessions. The trial Judge suppressed two, including the one given in conjunction with location of the body, but determined that evidence taken from the body was admissible under "inevitable discovery".
The State did not use any of Hoskins' confessions at trial, even the two the Judge refused to suppress. There was overwhelming physical, and circumstantial, evidence against Hoskins. His confessions, though incriminating, were attempts to minimize his own involvement, and shift guilt to others. Prosecutors decided that if the Jury was going to hear Hoskins' version of the crime, he would have to tell it from the witness stand, where he could be cross-examined. Hoskins elected not to testify. He was convicted, and is now on death row. His confessions, and the police actions that secured them, were not even considered by the Court that affirmed the conviction.20
The Woodward Case
In some cases, the defendant does not want to have a confession suppressed. Johnny Woodward was arrested for attempting to murder his girlfriend's unarmed husband.21 After his arrest, Woodward confessed to shooting the victim, but said he acted in self defense.
Woodward never moved to suppress his confession. He wanted the State to use it at trial, so the jury could hear how it was self-defense. The victim testified he was unarmed, standing on his own porch, when Woodward shot him from the sidewalk. An eye-witness corroborated the victim's testimony. The State rested without even mentioning Woodward's confession to the jury. Having no other way to tell his self-defense story, Woodward testified. The prosecutor's cross-examination revealed glaring discrepancies between Woodward's trial testimony, and his confession to the police.
Although Woodward never sought to suppress his confession before trial, when the State offered the confession to impeach him, he tried to have it suppressed. The Judge declined to hear the suppression motion, ruling that even if there was a Miranda violation the jury could still hear the confession to impeach the trial testimony. Woodward was convicted of attempted murder.
* * *
Neither Cassell's nor Schulhofer's analysis would account for the impact of Miranda, if there is one, on any of these four cases. They illustrate the difficulty of even isolating Miranda as a factor in the process, let alone determining any impact. Perhaps it may be possible to formulate a statistical model isolating the impact, but even the most obvious - counting how many times a guilty perpetrator "beats the rap" because of a suppressed confession - may not prove anything. In how many such cases did the police stop the investigation, after obtaining the confession, instead of continuing to gather other incriminating evidence?
The debate over the impact of Miranda will continue. Studies purporting to settle the question are fodder for political demagoguery, and sustain life in academia. Whether the Miranda decision is good, or bad, for law enforcement depends on one's political/philosophical perspective. Trying to settle a philosophical debate with statistics is like counting shadows on a wall.22
* Gary Beatty received his law degree from Florida State University, is Board Certified in Criminal Trials by the Florida Bar.
1. 384 US 436 (1966).
2. Professor of Law, University of Utah College of Law.
3. Professor of Law, University of Chicago.
4. Both appear in the Harvard Journal of Law and Public Policy Vol. 20, Number 2 (Winter, 1997).
5. In Duckworth v Eagan 492 US 195 (1989) the Court refused to require strict, formalistic, recitation of the Miranda warnings.
6. 467 US 649 (1984).
7. See also US v Corina 34 F3rd 876 (9 CCA, 1994); State v Kunkel 404 NW2d 69 (Wis. Ct. App. 1987); People v Riddle 184 Cal. Rptr. 170 (Cal. Dist. Ct. App., 1978).
8. Nix v Williams 467 US 431 (1984), held that if incriminating evidence found in violation of Miranda would have inevitably been found independently of the improperly obtained confession, then the violation doesn't matter and the evidence can be used.
9. See Harris v New York 401 US 222 (1971)
10. For example Michigan v Tucker 417 US 433 (1974); US v Scalf 708 F2d 1540 (10 CCA, 1983); and Wilson v Zant 290 SE2d 442 (Ga. 1982), all question the extent of the "fruit of the poison tree" doctrine to Miranda violations.
11. The arresting officer, and the law enforcement agency also get an arrest stat. Since officer performance ratings, and agency budgets, may be based on arrest stats, why should police care if some "liberal judge" eventually throws out the confession.
12. Mock, and the co-defendant, each confessed involvement but (no surprise) fingered the other as the actual killer. The co-defendant hung himself in his jail cell while awaiting trial, thus precluding any measurable Miranda impact on that arrest/clearance.
13. Consolidated cases 91-11766 CFB & 92-8228 CFA (Brevard County, Florida).
14. Mock v State 625 So2d 1335 (Fla. 5th Ct. App., 1993).
15. Case 92-11647 CFA (Brevard County, Florida).
16. Mock's admissions about both murders were included within the same statement given to police.
17. The State took an interlocutory appeal of the suppression order to the same appellate Court which had affirmed Mock's conviction in the first case. The first trial judge's denial of Mock's suppression motion had been an issue in the appeal of the first conviction. Prosecutors assumed that since the suppression issues were the same in both cases, that the results on appeal would the same. They were wrong. Suppression of the confession was affirmed. State v Mock 626 So2d 704 (Fla. 5th Ct. App., 1993).
18. Mock's admissions supported both prosecution theories in the first case. The Jury was instructed that if they voted to convict they would have to choose one or the other, but could not convict for both. In the second case, his admissions supported his conviction as either the actual perpetrator, or as a principal to the murder.
19. Case 92-17795 CFA (Brevard County, Florida).
20. Conviction Affirmed, but remanded for further sentencing proceedings. Hoskins v State 702 So2d 202 (Fla. 1997).
21. Case #95-28792 CFA (Brevard County, Florida).
22. "...they were in the habit of conferring honors among themselves on those who were quickest to observe the passing shadows, ... and who were thereafter best able to draw conclusions..." Plato's Republic, Book VII, "The Cave Allegory", Annotated by W. Gordon Snow.