The US Supreme Court repeatedly has cautioned that juveniles may lack the ability to exercise Miranda rights or make voluntary statements, but it has not required special procedures to protect them (Haley v. Ohio 1948; Gallegos v. Colorado 1962; In re Gault 1967; Fare v. Michael C. 1979; Yarborough v. Alvarado 2004; J.D.B. v. N. Carolina 2011). Instead, it uses the adult legal standard—knowing, intelligent, and voluntary—to gauge juveniles’ Miranda waivers (Fare v. Michael C. 1979). On the other hand, developmental psychologists question whether juveniles understand Miranda or are competent to exercise the rights. Younger adolescents may not understand Miranda’s words or the rights it conveys and may not be as able as adults to exercise the rights (Grisso 1980; Grisso et al. 2003). Most youths, 16 years of age and older, appear to understand Miranda on par with adults, although they lack mature judgment and may be more susceptible to adult authority influence (Grisso 1980, 1981; Feld 2013).

This chapter examines what happens when police interrogate older youths charged with felony offenses. Part I analyzes the law governing interrogation of juveniles and research on their competence to exercise the rights. Part II examines how police interrogate suspects and research on interrogation practices. Part III describes the study’s data and methodology. Part IV reports how police questioned 307 delinquents 16 years of age or older whom prosecutors charged with felonies. It describes how police secure Miranda waivers, how they interrogated youths, and how juveniles responded.

Interrogating Juveniles: Legal Expectations and Developmental Psychology

The Supreme Court recognized that youthfulness, lengthy questioning, and absence of counsel or parents could render juveniles’ statements involuntary (Haley v. Ohio 1948; Gallegos v. Colorado 1962). The Court in In re Gault (1967) granted delinquents the privilege against self-incrimination, and reiterated concern about youths’ vulnerability during questioning. Fare v. Michael C. (1979) used the adult “totality of the circumstances” test to evaluate juveniles’ Miranda waivers. Fare denied that youths’ developmental differences required additional procedural safeguards and demanded children to assert the rights clearly. J.D.B. v. N. Carolina (2011) ruled that age was an objective factor to consider in Miranda’s custody framework and concluded that police could gauge how a youth’s age would affect feelings of liberty restraint.

Most states use the same Miranda framework for juveniles and adults and require only an understanding of the rights and not collateral consequences (Feld 2006a, b, 2013). Trial judges consider how offenders’ age, education, I.Q., and prior police contacts, and the location, methods, and length of interrogation affected their Miranda waivers. About ten states require a parent to assist juveniles to invoke or waive Miranda (Larson 2003; Farber 2004; Drizin and Colgan 2004; Woolard et al. 2008).

Developmental psychologists distinguish between youths’ cognitive ability and maturity of judgment. By mid-adolescence, most youths can distinguish right from wrong and possess cognitive abilities comparable with adults (Steinberg and Cauffman 1999; Scott and Steinberg 2008; Steinberg et al. 2009). However, the ability to make good choices under laboratory conditions differs from the ability to make adult-like decisions under stress and with incomplete information (Steinberg and Cauffman 1996; Spear 2000).

Since the mid-1990s, the MacArthur Network on Adolescent Development and Juvenile Justice has studied decision-making and adjudicative competence (Scott and Steinberg 2008). The research distinguishes between cognitive ability and mature judgment, self-control and ability to resist external influences (Scott and Steinberg 2008). Although most youths 16 years of age or older exhibit cognitive abilities comparable with adults, they do not develop adult-like self-control and mature judgment for several more years.

Differences in knowledge, attitude toward risk, impulse control, and appreciation of consequences contribute to youths’ poorer decisions (Scott and Grisso 1997; Steinberg 2005; Scott and Steinberg 2008). Adolescents tend to underestimate risks, use a shorter frame, and focus on gains rather than losses (Furby and Beyth-Marom 1992; Grisso 2000). The widest divergence between juveniles’ and adults’ perceptions of and preference for risk occurs during mid-teens (Scott and Steinberg 2008). Neuroscientists attribute these differences to brain maturation and the increased ability of the prefrontal cortex (PFC) to perform executive functions and control impulses (Baird et al. 1999; Spear 2000; Dahl 2001; Gruber and Yurgelun-Todd 2006; Maroney 2009).

Despite these developmental differences, most states do not provide additional procedural safeguards, but use adult standards to gauge juveniles’ Miranda waivers. Some youths may not understand the words of the Miranda warning (Rogers et al. 2007; Rogers et al. 2008a, b). Some concepts—the meaning of a right, the term appointed to obtain counsel, and waive—render Miranda incomprehensible to many juveniles (Goldstein and Goldstein 2010).

Grisso (1980, 1981; Grisso et al. 2003) reported that many youths do not adequately understand and cannot exercise Miranda rights. Although juveniles 16 years or older understood Miranda about as well as did adults, substantial minorities of both groups misunderstood some components (Grisso 1980). Other studies report similar age-related improvements in cognitive ability and Miranda comprehension (Viljoen and Roesch 2005; Viljoen et al. 2007; Kassin et al. 2010). Even youths who understand Miranda may not be able, as adults to exercise their rights. Juveniles do not fully appreciate the function or importance of their rights (Grisso 1980, 1981), and perceive them as a privilege that authorities allow, but may unilaterally withdraw (Grisso 2002).

To be competent to stand trial, a person must understand proceedings, make rational decisions, and assist counsel (Dusky v. United States 1960; Drope v. Missouri 1975). Development limitations impair youths in the same ways that mental illness leaves adults incompetent (Grisso et al. 2003; Scott and Grisso 2005; Redlich 2004). Many juveniles 14 years of age and younger may be as severely impaired as adults found incompetent to stand trial (Bonnie and Grisso 2000). Differences in maturity and judgment cause even nominally competent adolescents to make poorer decisions than young adults (Grisso et al. 2003; Scott and Grisso 2005). Compromised competence affects youths’ ability to exercise Miranda rights.

Miranda characterized custodial interrogation as inherently compelling because a police-dominated environment created psychological pressures to comply. Children questioned by authority figures yield more easily to negative pressure (Gudjonsson 2003; Billings et al. 2007), and acquiesce more readily to suggestions during questioning than do adults (Ainsworth 1993; Bull and Corran 2003; Drizin and Leo 2004). Even older youths who understand Miranda may be more susceptible to power differentials and more prone to relinquish their rights.

Interrogation Practices and Empirical Assessments

Most police interrogators in the USA are trained in the Reid Method (Leo 2008). It teaches isolation and psychological manipulations—maximization and minimization techniques—to elicit confessions (Inbau et al. 2004). Police use negative incentives—confrontational tactics to scare or intimidate a suspect—and positive incentives—themes, scenarios, or sympathetic alternatives—to make it easier to confess (Kassin and McNall 1991; Ofshe and Leo 1997; Leo 2008). Maximization tactics “convey the interrogator’s rock-solid belief that the suspect is guilty and that all denials will fail. Such tactics include making an accusation, overriding objections, and citing evidence, real or manufactured, to shift the suspect’s mental statement from confident to hopeless” (Kassin et al. 2010, p. 12). Minimization techniques “provide the suspect with moral justification and face-saving excuses for having committed the crime in question. Using this approach, the interrogator offers sympathy and understanding; normalizes and minimizes the crime” (Kassin et al. 2010, p. 12). Reid training does not distinguish between youths and adults (Owen-Kostelnik et al. 2006; Meyer and Reppucci 2007), and teaches police to interrogate both similarly—“principles discussed with respect to adult suspects are just as applicable for use with younger ones” (Inbau et al. 2004, p. 298).

Interrogation in the UK is less confrontational and designed to elicit information rather than to secure a confession (Milne and Bull 1999; Bull and Milne 2004; Pearse and Gudjonsson 2003). In England and Wales, the Police and Criminal Evidence Act (PACE 1984) has required police to record interrogations for nearly two decades (Milne and Bull 1999; Bull and Soukara 2010). Police, psychologists, and lawyers developed an information-gathering method of interviewing that is less confrontational than the Reid approach (Gudjonsson 2003; Milne and Bull 1999). The mnemonic PEACE describes the five components of this interview approach—“Planning and Preparation,” “Engage and Explain,” “Account,” “Closure,” and “Evaluate” (Milne and Bull 1999). Minnesota interrogation practices reflect both Reid and PEACE elements (Nelson 2006).

In the decades since Miranda, we have remarkably few studies of how police question people (Leo 2008). Post-Miranda research in the late-1960s evaluated whether police warned suspects and how warnings affected rates of confessions (Wald et al. 1967;Neubauer 1974; Leo 1996a, c; Feld 2006a). In the mid-1990s, Richard Leo (1996b, c) conducted the only field study of interrogation in the USA. Legal scholars and criminologists have used indirect methods and studied tapes and transcripts of interrogations (Feld 2006a, b, 2013; King and Snook 2009), or attended prosecutors’ charging sessions and interviewed police about interrogations (Cassell and Hayman 1996). In England and Wales, analyses of PACE recordings have generated a substantial body of empirical research (Gudjonsson 2003; Milne and Bull 1999; Bull and Soukara 2010; Pearse and Gudjonsson 2003). Psychologist Saul Kassin and associates have conducted laboratory research on interrogation for decades (Kassin and Gudjonsson 2004; Kassin 2005; Kassin et al. 2010). Studies of false confessions provide another glimpse into how police interrogate suspects and highlight the vulnerability of youths (Drizin and Leo 2004; Garrett 2011).

Methodology and Data

The Minnesota Supreme Court in State v. Scales (1994) required police to record custodial interrogations of criminal suspects, including juveniles. County attorneys in Minnesota’s four largest counties—Anoka (suburban), Dakota (suburban), Hennepin (Minneapolis), and Ramsey (St. Paul)—allowed me to search closed files of 16- and 17-year-old youths charged with a felony and to copy those in which police interrogated or juveniles invoked Miranda. Delinquency trials of 16- and 17-year-old youths charged with felony offenses are public proceedings (Minn. Stat. Ann. 2005). Police conducted these interviews between 2003 and 2006. The four most populous counties account for almost half (47.6 %) of the state’s population and nearly half (45.6 %) of the delinquency petitions filed. The two urban counties accounted for somewhat more than half (53.4 %) the files. These files contained Scales interrogation tapes—i.e., recorded interrogation required by the Court in State v. Scales—or transcripts, police reports, juvenile court records, and sentences. Court Orders authorized access to juvenile courts files, but included confidentiality stipulations to protect juveniles’ identity and imposed methodological limitations.Footnote 1

I adapted and expanded codebooks used in prior interrogation research (Wald et al. 1967; Leo 1996b; Pearse and Gudjonsson 2003).Footnote 2 I reviewed police reports to learn about the crime, the context of interrogation, and evidence police possessed prior to questioning. I coded each file to analyze where, when, and who was present at an interrogation, how police administered the Miranda warning, whether juveniles invoked or waived, whether officers used Reid Method maximization and minimization techniques, and how juveniles responded. There may be some sample selection bias because these are cases in which prosecutors charged serious delinquents and perhaps include a larger proportion of juveniles who waived Miranda.Footnote 3 Despite these caveats, the data comprise the largest number of felony interrogations in the USA. More than 150 police officers from more than 50 agencies conducted these interrogations. I did saturation interviews with police, prosecutors, defense lawyers, and juvenile court judges to elicit their views, learn from their experience, and buttress my findings.Footnote 4

Police Interrogation

This section examines what happens in the interrogation room. It examines characteristics of youths who waived or invoked Miranda. It analyzes how police secured Miranda waivers and questioned those youths who waived. It focuses on the length and outcomes of interrogations.

Sample Characteristics

Table 3.1 reports that males comprised the vast majority (89.3 %) of youths whom police questioned. Prosecutors charged about half (55.0 %) with property offenses—burglary, larceny, and auto-theft—and nearly one-third (31.6 %) with crimes against person—murder, armed robbery, aggravated assault, and criminal sexual conduct. They charged the remaining youths with drug crimes (6.2 %), firearm offenses (5.5 %), and other felonies (1.6 %). They charged about half (56.4 %) with one felony, an additional quarter (25.1 %) with two crimes, and the remainder with three or more. Prosecutors filed certification motions and juvenile court judges transferred some of the most serious offenders to criminal court.

Table 3.1 Characteristics of juveniles interrogated
Full size table

Nearly one-third (30.6 %) of juveniles had no prior arrests. Police previously had taken more than one-third of these youths into custody for noncriminal status offenses (15.3 %) or misdemeanors (22.8 %). About one-third of these youths (35.1 %) had one or more prior felony arrests and more than half (57 %) had prior juvenile court referrals. Nearly one-third (29.9 %) were under court supervision—probation, placement, or parole status—when police interrogated them. About half of the youths were white (52.1 %) and the others (47.9 %) members of ethnic and racial minority groups—Black, Hispanic, Native American, and Asian. Black juveniles accounted for more than one-third (34.9 %) of the sample. The youths whom police interrogated included a larger proportion of males, more youths charged with property and violent crimes and with prior court referrals, and fewer charged with drug offenses than these counties’ 16- and 17-year-old felony dockets (Feld and Schaefer 2010a, b).

Securing Miranda Waivers

Miranda requires officers to warn suspects whom they take into custody and interrogate to dispel the inherent coercion of isolation and questioning. Police had formally arrested the vast majority (86.6 %) of these juveniles prior to questioning, and made a Scales recording whether they initially arrested or later released a youth. Police detained nearly two-thirds (61.7 %) of those whom they questioned and released the others to parents. About half (55.7 %) of interrogations took place in police stations and another quarter (23.1 %) in juvenile detention centers. Thus, police questioned more than three-quarters (78.8 %) of youths in interrogation rooms. Nearly one-tenth (8.1 %) of interrogations took place in a police car at the place of arrest. Police also interrogations juveniles at their homes (6.2 %) or at schools (6.2 %). Every juvenile in the sample received a proper Miranda warning.

When police give suspects their Miranda warnings, they have no incentive to encourage them to invoke their rights. Police used several tactics to predispose suspects to waive Miranda without alerting them to its significance—admonishing them to tell the truth, minimizing the warning, or advising that it is the only opportunity to tell their story (Leo 1996b; Leo and White 1999). They may use routine booking questions to establish rapport and predispose youths to waive before they give a warning, (Weisselberg 2008; Rhode Island v. Innis 1980; Pennsylvania v. Muniz 1990). In about half of cases (52.8 %), police gave the Miranda warning immediately after identifying the suspect. In the remaining cases (47.2 %), police asked juveniles booking questions—name, age and date of birth, address and telephone number, grade in school, and the like—and sometimes used juveniles’ responses to engage in casual conversations and to accustom them to answering questions.

Police sometimes framed a Miranda waiver as a prerequisite to a juvenile’s opportunity to tell his side of the story and emphasized the importance of telling the truth before they gave a Miranda warning. Officers characterized warnings as an administrative formality to complete before the suspect can talk. They sometimes referred to it as “paper work” to emphasize its bureaucratic or ritual quality. Officers regularly referred to youths’ familiarity with Miranda from television and movies. Miranda’s cultural ubiquity may detract from youths’ understanding, as the warning becomes background noise at an interrogation.

Suspects must either waive or clearly invoke their Miranda rights to silence and to counsel (Fare v. Michael C. 1979; Davis v. United States 1994; Berghuis v. Thompkins 2010). Police established that a juvenile understands her rights by reading the warning and then eliciting an affirmative response. In this study, officers read each right to the youth followed by the question “Do you understand that?” Juveniles acknowledged receiving each warning on the record—the Scales tape—and in about one-fifth (19.5 %) of cases initialed and signed a Miranda form. Police in this study consistently obtained express waivers. After they asked juveniles if they understood the warning, they concluded the waiver process, “Bearing in mind that I’m a police officer and I’ve just read your rights, are you willing to talk to me about this matter?” Another waiver formula ended, “Having these rights in mind, do you wish to talk to us now?”

Miranda reasoned that police must warn a suspect to dispel the compulsive pressures of custodial interrogation. A dissent in Miranda asked why those pressures do not coerce a waiver as readily as an unwarned statement (Miranda v. Arizona 1966). Legal analysts and criminologists concur that after police isolate suspects in a police-dominated environment, a warning cannot adequately empower them to invoke their rights (White 1997; Weisselberg 2008). Post-Miranda studies consistently report about 80 % or more of adults waive Miranda (Wald et al. 1967; Leo 1996b, 2008; Cassell and Hayman 1996; Kassin et al. 2007).

Three decades of research report that juveniles waive Miranda rights at somewhat higher rates than do adults—typically 90 % or more (Grisso 1980; Grisso and Pomiciter 1977; Feld 2006a; Goldstein and Goldstein 2010). Juveniles’ higher waiver rates may reflect their lack of understanding or inability to invoke Miranda effectively. Equally plausible, waivers may reflect prior justice system involvement and juveniles will have had less experience than adults will (Viljoen and Roesch 2005). Table 3.2 reports that the vast majority of youths (92.8 %) waived Miranda—a rate consistent with other juvenile studies and 10 % higher than that reported for adults. Justice system personnel confirmed the accuracy of these findings—almost all delinquents waived Miranda.Footnote 5

Table 3.2 Juveniles who waive or invoke Miranda rights by offense and prior record
Full size table

Analysts report a relationship between prior police experience and Miranda invocations (Leo 1996b; Kassin 2005). Post-Miranda research reported that defendants with prior arrests and felony convictions gave fewer confessions than did those with less experience. Older youths and those with prior felony referrals invoked more frequently than did younger juveniles and those without prior contacts (Grisso 1980; Grisso and Pomiciter 1977).

About one-third (35.1 %) of these youths had one or more felony arrests prior to the offense for which police questioned them. Juveniles with one or more prior felony arrests waived their rights at significantly lower rates (86.9 %) than did those with fewer or less serious police contacts (94.9 %). Youths who waived at prior interrogations may have learned that they derive no benefit from confessing. Youths with prior arrests would have spent more time with lawyers and learned from those experiences. They also may have learned to cope with and resist the pressures of interrogation.

Police Interrogation Tactics

In the UK, Norway, New Zealand, and Australia investigative interviews tend to be less confrontational, designed to obtain a free-narrative account, and aimed to elicit facts rather than to secure a confessions (Milne and Bull 1999; Bull and Milne 2004). By contrast, police in the USA traditionally use more accusatorial tactics to obtain incriminating admissions or leads to other evidence—physical evidence, co-offenders, witnesses, or stolen property—to strengthen prosecutors’ cases, and to produce guilty pleas. Police seek suspects’ statements—true or false—to control changes the accused may make in their stories and to impeach their credibility. Minnesota training advises officers to portray themselves as report writers who want to learn what happened to put in a statement for prosecutors and judges to evaluate (Nelson 2006). Police often described their roles as natural fact finders. They frequently advise suspects that the interview provides their opportunity to “tell their story.” Depending on how forthcoming a youth is initially, they may use maximization and minimization tactics to elicit a statement.

Police reported that they used maximization techniques regularly. They initially encouraged juveniles to commit to a story—true or false—and then used more confrontational tactics to challenge their version thereafter. Table 3.3 summarizes maximization strategies police used: confronted juveniles with evidence (54.4 %); accused them of lying (32.6 %); exhorted them to tell the truth (29.5 %); asked BAI questions (28.8 %); challenged inconsistencies (20.0 %); emphasized the seriousness of the offense (14.4 %); and accused them of other crimes (8.4 %).

Table 3.3 Maximization and minimization questions: Type and frequency
Full size table

In about one-third (30.9 %) of cases, police did not use any maximization techniques and in another quarter (23.1 %), they used only one. The paucity of maximization tactics suggests that most juveniles did not require a lot of prompting or intimidation to cooperate. Police used three or more maximization tactics in fewer than one-third (31.6 %) of cases.

In about half (54.4 %) the interrogations, police confronted juveniles with statements from witnesses or co-offenders, or referred to physical evidence. In most cases, police will not have analyzed DNA or fingerprint evidence in the short time between a suspect’s arrest and interrogation. Sometimes, police described an investigation as if they already had obtained the evidence. In other instances, they questioned youths about potential evidence that later investigation would reveal. They asked how a juvenile would respond to hypothetical evidence—“what if I told you” that someone had identified him or police found his fingerprints? Officers might ask a juvenile “is there any reason why” his DNA might be on a gun or he would appear on surveillance video?

In about one-third (32.6 %) of cases, officers accused juveniles of lying. Police typically allowed juveniles to commit to a story and then confronted them. In nearly one-third of cases (29.5 %), officers urged juveniles to be honest and tell the truth. Police intimated that their recommendations could affect prosecutors’ charges and judges’ decisions. They cautioned that prosecutors and judges reacted negatively to an implausible story and responded more favorably to truthful defendants.

Inbau et al. (2004) posit that innocent and guilty people respond differently to emotionally provocative questions and their reactions enable investigators accurately to classify them. Leo (1996b) reported that officers asked such Behavioral Analysis Interview (BAI) questions in about 40 % of interrogations. Police in this study used BAI questions in more than one-quarter (28.8 %) of interviews, most commonly “Do you know why I have asked to talk to you here today?” However, few of these files indicated that police had any conversations prior to Scales recordings and none in which they interrogated these youths.

Police began many interviews with an invitation to a youth to tell his story, but they warned that it was a time-limited opportunity. They cautioned that if youth did not take advantage of the chance to explain their involvement, then they might regret it later. Police sometimes withheld information from juveniles about the investigation to increase uncertainty and anxiety and cautioned a reluctant youth that without his version, other co-offenders might shift responsibility to him or make a deal at his expense.

Minimization tactics offer face-saving excuses or moral justifications that reduce a crime’s seriousness, provide a less odious motivation, or shift blame to a victim or accomplice (Kassin et al. 2010). The Reid Method teaches police to develop a theme or scenario to neutralize guilt, minimize responsibility, and make it easier to confess (Inbau et al. 2004; Leo 2008). Criminologists have used techniques of neutralization to understand how youths rationalize delinquent behavior (Sykes and Matza 1957). Many themes are extensions of criminal law defenses—provocation, intoxication, or insanity—that provide rationales to reduce moral constraints (Matza 1964). While delinquents may reject mental illness—insanity—as an excuse, they readily embrace “going crazy” or “being mad” to rationalize criminal conduct. Police sometimes suggested that getting mad, losing control, or excitement accounted for youths’ criminal misconduct. Intoxication provides an explanation for bad behavior, and juveniles readily invoked drinking alcohol or using drugs to excuse criminal conduct. Similarly, police diffused juveniles’ responsibility by suggesting that they succumbed to negative peer influences. Juveniles often commit their crimes in groups (Snyder and Sickmund 2006), and police can blame others and allow juveniles to shift blame as well. Parents regularly refer to errant children’s behavior as a mistake and youths learn that mistakes can mitigate responsibility. In the present study, police regularly encouraged juveniles to attribute their delinquency to a mistake.

Police used minimization tactics in about one-fifth (17.3 %) of these interrogations, far less often than they used maximization tactics (69.1 %). Although prosecutors charged all these youths with felonies, one officer explained that “most of these are fairly minor, so you don’t have to do a whole lot of minimizing.” Officers used scenarios or themes to reduce suspects’ guilt or culpability in 15.4 % of cases. Police described benefits juveniles might derive and appealed to self-interest in one-tenth (11.9 %) of cases. They expressed empathy in one-tenth of cases (10.5 %), and in fewer offered to investigate further or assist juveniles to receive help. Officers minimized a crime’s seriousness by describing its triviality in comparison with other delinquents’ offenses. Even a serious crime—a drive-by shooting—could have been worse if the shooter had hit the intended target. The rationale of juvenile courts—treatment rather than punishment—provided officers with another theme with which to offer help and to minimize seriousness. The infrequency of minimization tactics is consistent with research in the UK, and Minnesota training protocols that discourages their use (Nelson 2006; Soukura et al. 2009).

Juveniles’ Responses

How did the 285 youths who waived Miranda respond to police and how did their attitudes affect the information they provided? I coded the outcome of interrogations into three categories—confess, admit, or deny.Footnote 6 (Cassell and Hayman 1996; Wald et al. 1967). Table 3.4 reports that a majority (58.6 %) of juveniles confessed within a few minutes of waiving Miranda and did not require prompting by police. British research confirms that the majority of suspects confessed and “almost all did so near the beginning of the interviews” (Soukara et al. 2009, p. 495). The UK analysts concluded that “suspects enter a police interview having already decided whether to admit or deny the allegations against them” and interrogation tactics had little impact on whether they admit (Milne and Bull 1999, p. 81). (But see, Bull and Soukara (2010) for analyses of 40 interviews in which suspects “shifted from denying to admitting.”) An additional one-third (29.8 %) of juveniles provided statements of some evidentiary value, for example, admitting that they served as a lookout during a robbery or participated in a burglary even if they did not personally steal property. Justice personnel agreed that most juveniles made some incrimination admissions.

Table 3.4 Outcome of interrogation and youths’ attitudes
Full size table

Other studies corroborate similar high rates of admissions and confessions. Leo (1996b) found such outcomes in three-quarters (76 %) of cases in which adults waived Miranda. About two-thirds (64 %) of interrogations in the Yale-New Haven study yielded incriminating evidence (Wald et al. 1967). A survey of police estimated that two-thirds (68 %) of suspects incriminated themselves (Kassin et al. 2007). Other UK research reports a rate of 77 %, ranging from 64 to 97 % among various police stations (Evans 1993; Bull and Milne 2004). More than half (55 %) of delinquents held in detention reported they had confessed (Viljoen et al. 2005).

Only a small proportion (11.6 %) of juveniles made no incriminating admissions. Forms of resistance included noncooperation, denial of knowledge and culpability, lying, evasion, silence, or blame shifting. When confronted with resistance, police used more maximization techniques than they did with cooperative youths, but did not question them for longer periods. Once they recognized a youth was resistant, they concluded the interrogation with the comment that prosecutors and judges would not view them favorably.

Criminologists have studied the interplay between police discretion and juveniles’ attitudes (Black and Reiss 1970; Clark and Sykes 1974), and reported that for less serious crimes, deferential youths reduce likelihood of arrest and contumacious ones increase it (Piliavin and Briar 1964; LaFave 1965; Skolnick 1967; Bittner 1976). Studies of police and probation officers report that a youth’s attitude affected how officials perceived, imputed moral character, and responded to him or her (Emerson 1974; Cicourel 1995).

Police reported that juveniles’ attitudes ranged the gamut—“some are scared to death, and others, it’s almost a joke.” Many officers described youths as scared, especially “the kids that are new to the process.” Although police described some youths as confrontational, justice system personnel viewed most youths as cooperative or submissive. “I would say that 90 % or more would probably be cooperative and the other percentage would be the frequent-fliers so to speak.” Several officers used the expression “deer in the headlights” to describe youths’ demeanors in the interrogation room. Public defenders described juveniles as humbled or defeated when they confessed.

Ethnographers emphasize the variability of attitudes—“rude or impolite, aggressive or passive, laughter or tears, and the like”—and their impact on justice system processing (Cicourel 1995, p. xv). Juveniles exhibited many attitudes during interrogation—polite, cooperative, distressed, remorseful, frightened, cocky, resistant, aggressive, and confrontational—which could fluctuate from one minute to the next. Police reports frequently described juveniles’ demeanors, emotional responses, and behavior during interrogation. They documented whether they believed suspects told the truth or lied and indicated whether they cooperated or resisted. Based on the characterizations in officers’ reports and my impressions, I dichomized juveniles’ attitudes as cooperative or resistant. Other research used similar categories and described 80 % of suspects as cooperative (Baldwin 1993). Juveniles cooperate for many reasons—human decency in social interactions, fear and anxiety, dependency on authority figures, or the coercive pressures of isolation—but most exhibited positive attitudes.

As Table 3.4 indicates, the vast majority of juveniles (79.6 %) exhibited a cooperative demeanor and only one-fifth (20.4 %) appeared resistant. Not surprisingly, the vast majority (96.5 %) of cooperative juveniles confessed or made incriminating admissions. By contrast, fewer than one in ten (8.5 %) resistant juveniles confessed and almost half (43.1 %) provided no useful admissions (8.6 %). Only one-tenth (11.6 %) of youths denied involvement, but those who exhibited resistant attitudes accounted for more than three-quarters of them (75.8 %).

Police in the USA typically question suspects to elicit admissions or obtain statements that prosecutors can use to impeach testimony. Suspects’ answers may lead to other evidence—witnesses, co-offenders, or property. Table 3.4 reports how often interrogations yielded corroborating evidence. I defined corroborating evidence as evidence which police did not possess prior to questioning—leads to physical evidence, a crime scene diagram, identity of a co-offender, or unknown witness. By this conservative standard, about one-fifth (18.2 %) of interviews yielded information that police did not already have. Thus, interrogation did not produce much collateral evidence and gathering it appears to be a secondary goal.

Some police attributed the low-yield of corroborating evidence to time pressures produced by the volume of cases they investigated. Once police obtained an admission—which they did quickly—they did not press youths for additional evidence. Prosecutors confirmed that interrogations did not often lead to corroborating evidence, but attributed it to good preliminary investigations. Police questioned more than two-thirds (69.7 %) of juveniles within less than 24 hours of their crimes—i.e., police either caught them committing the crime or very shortly thereafter. As a result, police and prosecutors had strong enough evidence with which to obtain a conviction in about two-thirds (63.2 %) of cases even without conducting an interview. Prosecutors said that juveniles’ statements sometimes provided bases to obtain search warrants which produced additional evidence not disclosed directly in the interview.

Length of Interrogation

Table 3.5 reports the length of interrogations by type of offense and by whether the offense involved a firearm.Footnote 7 Police completed three-quarters (77.2 %) of interviews in less than 15 minutes and concluded nine in ten (90.5 %) in less than 30 minutes. In the longest interviews, police questioned three youths (1.1 %) for 1.5 hours. Although prosecutors charged youths with one or more felonies, brief interrogations are unlikely to cause false confessions (White 2001).

Table 3.5 Length of interrogation by type of offense* and weapon**
Full size table

Although these short interviews initially seemed surprising, other research confirms that interrogations of even 2 or 3 hours are exceptional and frequently problematic (Drizin and Leo 2004; Kassin et al. 2007). Police questioned suspects for more than an hour in only 15 % of cases in the Yale-New Haven study (Wald et al. 1967). Leo (1996c) reported that police questioned only one-quarter (28.7 %) of suspects for more than 1 hour. Cassell and Hayman (1996) reported that only 13 % of interrogations took more than 30 minutes and only one lasted longer than an hour. Research on British interrogations of juveniles reported that “[i]nterviews tended to be very brief with the majority taking less than fifteen minutes (71.4 %). Although the average length of interviews was around 14 min, the most frequent length was around 7 minutes” (Evans 1993, p. 26). Analyses of taped UK interrogations reported that “most were short and surprisingly amiable discussions” in which more than one-third of suspects confessed at the outset (Baldwin 1993, p. 331). Kassin and Gudjonsson (2004) reported that most interviews were short—80 % lasted less than 30 minutes; 95 % were completed within 1 hour—suspects confessed in 58 % of cases, police applied little interrogative pressure, and very few suspects who initially denied guilt subsequently confessed. Inbau et al. (2004) warn against interrogations longer than 4 hours—a length of time not observed in any published research. (But see Bull and Leahy–Harland (2012) for analyses of 56 interviews with an average duration of several hours).

I asked justice professionals to estimate the lengths of interviews and they universally agreed, “They’re actually very short.”Footnote 8 When asked why police concluded felony interrogations so quickly, justice system personnel attributed brevity to several factors. Many professionals referred to police workload pressures. Police conducted a form of triage, questioned suspects longer in more serious cases, but did not regard most juvenile felonies as serious crimes. Several officers attributed brief interrogations to the relative simplicity of most juveniles’ crimes and their ability to elicit admissions quickly.

A statistically significant relationship appeared between length of interrogation and type of offense. Police questioned a larger proportion of youths charged with property and drug crimes for 15 minutes or less than they did youths charged with other types of offenses. Crimes that involved some physical evidence—drugs, stolen property or automobiles—may have provided police with more evidence with which to confront these juveniles, resulting in shorter interviews.

Cases involving firearms resulted in longer interrogations. Although police questioned only 9.5 % of delinquents for longer than 30 minutes, they interrogated twice as many (20 %) juveniles charged with firearms offenses for longer than 30 minutes. I compared the lengths of interrogation in all cases that involved guns—armed robbery, assault with a gun, firearms possession, or burglary in which youths stole guns—with cases in which juveniles used other weapons—knives, blunt instruments, or automobiles—or did not use a weapon.

Guns provide an indicator of offense seriousness (Podkopacz and Feld 1996), and police questioned these juveniles longer and more aggressively. Police wanted to recover guns used or stolen by youths, and they used more maximization and minimization tactics to retrieve them. Officers referred to the benefits that would accrue to a youth who helped them to recover a gun and described the dangers guns posed to people who held them and those around them. Only two interrogations raised constitutional issues of voluntariness and both involved questioning to recover guns. In each case, police questioned juveniles for the longest time (1 and 1.5 hours), used the most maximization techniques, and made explicit quid-pro-quo promises of leniency to recover guns used or stolen by juveniles.

Police and justice system personnel confirmed the relationship between guns and length of interrogation and suggested that guns provide a reasonable proxy for crime seriousness. Police associated guns with youths’ involvement with gangs—another indicator of seriousness. Police also questioned youths to learn about other youths who had contact with the weapon. Juveniles knew that gun crimes garnered serious consequences for them, which gave them greater incentive to resist interrogators. Serious crimes are more likely to go to trial and police invested more energy to strengthen prosecutors’ cases.

Policy Implications

Although Miranda purported to bolster the adversary system and protect citizens, warnings failed to achieve those goals. In the decades since Miranda, the Court’s decisions have limited its scope, applicability, and the adverse consequences when police fail to comply (Slobogin 2007; Weisselberg 2008). Miranda’s assumption that a warning would empower suspects to resist the compulsive pressures of interrogation is demonstrably wrong. Post-Miranda research reports that 80 %of adults and 90 % of juveniles waive their sole protection in the interrogation room and only some sophisticated suspects invoke their rights. Although Miranda recognized that interrogation threatened the adversarial balance between the individual and the state, waivers provide police with a window of opportunity to conduct an inquisitorial examination. Perversely, Miranda allows judges to focus on ritualistic compliance with a procedural formality rather than to assess the voluntariness or reliability of a statement (Weisselberg 2008; Godsey 2005; Leo 2008). Judicial review of a Miranda waiver is the beginning and end of regulating interrogation (Missouri v. Seibert 2004).

Protecting Young Offenders in the Interrogation Room

Miranda is especially problematic for younger juveniles who may not understand its words or concepts. The Court has recognized youths’ vulnerabilities and distinguished between younger and older youths. Developmental psychologists corroborate their differing abilities. Younger juveniles’ incomplete understanding, impaired judgment, and heightened vulnerability warrant greater assistance—a non-waivable right to counsel—to assure voluntariness of a Miranda waiver and statement. Psychologists distinguish between youths’ cognitive abilities—their capacity to understand—and ability to make mature decisions and exercise self-control. Miranda requires only the ability to understand words, which developmental psychologists conclude that most 16- and 17-year-old youths can do.

The present study corroborates that 16- and 17-year-old juveniles appear to understand and exercise Miranda similarly to adults. This consistency inferentially bolsters research that younger juveniles lack understanding and competence to exercise the rights. Many, if not most, children 15 or younger do not understand Miranda or possess competence to make legal decisions (Grisso 1980; Grisso et al. 2003). Research on false confessions underscores the unique vulnerability of younger juveniles (Drizin and Leo 2004; Gross et al. 2005; Garrett 2011). In a prior study, police obtained more than one-third (35 %) of proven false confessions from suspects younger than 18 (Drizin and Leo 2004), and younger adolescents are at greater risk to confess falsely than older ones (Tepfer et al. 2010).

Developmental psychologists attribute their over-representation among false confessors to reduced cognitive ability, developmental immaturity, and increased susceptibility to manipulation (Bonnie and Grisso 2000; Tobey et al. 2000; Redlich et al. 2004). They have fewer life experiences or psychological resources with which to resist the pressures of interrogation (Drizin and Luloff 2007; Redlich et al. 2004). Juveniles’ lower social status and societal expectations of obedience to authority create pressures to waive (Gudjonsson 2003; Leo 2009). Juveniles are more likely than are adults to comply with authority figures, tell police what they think they want to hear, and respond to negative feedback (Gudjonsson 2003; Lyon 1999). The stress of interrogation intensifies their desire to escape in the short run by waiving and confessing (Owen-Kostelnik et al. 2006; Goldstein and Goldstein 2010). Limited ability to consider long-term consequences heightens their risk (Redlich 2010). The immature adolescent brain contributes to impulsive behavior and heightened vulnerability (Birckhead 2008; Maroney 2009).

Despite youths’ heightened susceptibility, police do not seem to incorporate developmental differences into the tactics they employ (Owen-Kostelnik et al. 2006). Techniques designed to manipulate adults—aggressive questioning, presenting false evidence, and leading questions—may create unique dangers when employed with youths (Kaban and Tobey 1999; Tanenhaus and Drizin 2002; Redlich and Drizin 2007). Police in this study did not report receiving special training to question juveniles and used the same tactics employed with adults.

The Supreme Court excluded statements taken from youths 15 years of age or younger—Haley, Gallegos, Gault, and J. D. B.—and admitted those obtained from 16- and 17-year-olds—Fare and Alvarado. The Court’s de facto line—15 and younger versus 16 and older—closely tracks what psychologists have found about youths’ ability to understand the warning and concepts. State courts and legislatures should formally adopt the functional line that the Court and psychologists discern between youths 16 and older and those 15 and younger.

Analysts advocate that juveniles younger than 16 years of age “should be accompanied and advised by a professional advocate, preferably an attorney, trained to serve in this role” (Kassin et al. 2010, p. 28). Juveniles should consult with an attorney, rather than rely on parents, before they exercise or waive their constitutional rights (Farber 2004; Bishop and Farber 2007). More than three decades ago, the American Bar Association endorsed mandatory, non-waivable counsel because it recognized that “Few juveniles have the experience and understanding to decide meaningfully that the assistance of counsel would not be helpful” (American Bar Association 1980, p. 92).

Requiring a child to consult an attorney assures an informed and voluntary waiver (Farber 2004; Drizin and Luloff 2007). If youths 15 years of age or younger consult with counsel prior to waiver, this may limit somewhat police’s ability to secure confessions. However, if younger juveniles cannot understand or exercise the rights without legal assistance, then to treat them as if they do denies fundamental fairness and enables the state to exploit their vulnerability. Fare emphasized lawyers’ unique roles in the justice system, and Haley, Gallegos, and Gault recognized younger juveniles’ exceptional needs for assistance.

Limiting the Length of Interrogations

Most false confessions emerge only after lengthy interrogations and youthfulness exacerbates those dangers (Drizin and Leo 2004; Kassin and Gudjonsson 2004; Gross et al. 2005). The Court recognizes that lengthy interrogations can produce involuntary confessions (Ashcraft v. Tennessee 1944), and found that questioning juveniles for 5 or 6 hours could produce an involuntary statement (Haley v. Ohio 1948; Gallegos v. Colorado 1962). Policy-makers should create a sliding-scale presumption of involuntariness based on length of interrogation and examine a confession’s reliability more closely as length of questioning increases (Leo et al. 2006).

The vast majority of the interrogations analyzed were very brief. Police concluded 90 % of these felony interrogations in less than 30 minutes. Every study reviewed reports that police completed nearly all interrogations in less than an hour and few take as long as 2 hours. By contrast, interrogations that elicit false confessions are usually long inquiries that wear down an innocent person’s resistance—85 %took at least 6 hours (Drizin and Leo 2004).

I cannot prescribe outer time limits because I did not encounter either lengthy or factually problematic interrogations. However, states should create a sliding-scale presumption that police elicited involuntary confession as the length of questioning increases. Police complete nearly all felony interrogations in less than 1 hour, but extract most false confessions only after grilling suspects for 6 hours or longer. These times provide a framework to limit interrogations and strengthen the presumption of coercion (White 2001; Kassin et al. 2010). Four hours would provide ample opportunity for police to obtain true confessions from guilty suspects willing to talk without increasing the risk of eliciting false confessions from innocent people.

On the Record

Within the past decade, legal scholars, psychologists, law enforcement, and justice system personnel have reached consensus that recording interrogations reduces coercion, diminishes dangers of false confessions, and increases reliability (Cassell 1998; Milne and Bull 1999; Gudjonsson 2003; Drizin and Reich 2004; Sullivan 2004, 2006, 2010; Garrett 2010, 2011). About a dozen states require police to record interrogations, albeit some under limited circumstances—homicide or young suspects (Leo 2008; Sullivan 2010; Garrett 2011). Many police departments have policies to record interrogations for some crimes (Sullivan 2006, 2010).

Recording creates an objective record and an independent basis to resolve disputes between police and defendants about Miranda warnings, waivers, or statements (Slobogin 2003). A complete record enables fact finders to decide whether a statement contains facts known to a guilty perpetrator or police supplied them to an innocent suspect during questioning (White 1997; Garrett 2010, 2011). Recording protects police from false claims of abuse (Cassell 1998; White 1997). It enables police to focus on suspects’ responses, to review details of an interview not captured in written notes, and to test them against subsequently discovered facts (Drizin and Reich 2004). It reduces the need for an officer to take notes or a second person to witness a statement, which may chill a suspect’s willingness to talk. Recording avoids distortions that occur when interviewers rely on memory or notes to summarize a statement (Milne and Bull 1999).

A recorded confession greatly strengthens prosecutors’ plea bargain advantage. It enables them to avoid suppression hearings, negotiate better pleas, and obtain convictions (White 1997; Sullivan 2006). Defense lawyers can review recordings rather than rely on clients’ imperfect recollections of a stressful event. Scales recordings have virtually eliminated motions to suppress confessions because tapes provide unimpeachable evidence. This generates substantial savings because police, prosecutors, and defense counsel do not have to prepare for and judges do not have to conduct suppression hearings about Miranda warnings, coercive tactics, or the accuracy of a statement. Interviews with prosecutors, defense lawyers, and judges confirmed that defenders filed few motions to suppress evidence for Miranda violations.

Justice system personnel attributed Scales’ reduction of suppression motions to several factors. First, police acted professionally and complied with Miranda’s protocol—there is no ambiguity about warnings and waivers. In addition, most juveniles confess and tapes provide unimpeachable evidence of their statements. Juveniles’ statements limited defense attorneys’ options and reinforce a system of plea bargains, rather than trials. Scales enhanced police professionalism, documented Miranda compliance, obviated suppression hearings, led quickly to guilty pleas, and focused lawyers’ attention on appropriate sentences rather than guilt or innocence. Scales enables professionals to administer an inquisitorial model of justice “on the record,” expedites processing of routine cases, and reserves court resources for complex cases.

For recording to be an effective safeguard, police must record all conversations—preliminary interviews and interrogations—rather than just a final statement—a “post-admission narrative” (Gudjonsson 2003; Garrett 2011). Otherwise, police may conduct a pre-interrogation interview, elicit incriminating information, and then record only a final confession after the “cat is out of the bag”—a variation of the two-step practice condemned in Missouri v. Siebert. Only a complete record of every interaction can protect against a final statement that ratifies an earlier coerced one or against a false confession contaminated by nonpublic facts that police supplied a suspect (Kassin 1997; Garrett 2011).

The Court repeatedly insists that American criminal and juvenile justice is an adversary system. Repeated assertions do not alter the reality that states establish most defendants’ guilt through an inquisitorial system in which suspects seal their fate in the interrogation room and render trial procedures a nullity—interrogation elicits confessions and confessions produce guilty pleas (Packer 1968). Because states do not and need not provide full adversarial testing in every case, we need stronger mechanisms to assure factual reliability of inquisitorial justice. Recording imposes no great burden on police, illuminates the inner-workings of the interrogation room, and provides an objective record on which a defendant may appeal to a judge. Because the vast majority of defendants do not receive a trial, judicial review of the record provides an alternative check to assure the reliability of routine felony justice.

This study is only the second naturalistic empirical study of police interrogation in the USA in the past three decades (Leo 1996b), and the first involving juveniles. We need far more empirical research on interrogations practices in general, in a number of different settings, and with more knowledge about characteristics of suspects. As more jurisdictions adopt taping and recording requirements, we will have further opportunity to conduct this type of constructive research. Recordings provide opportunities for psychologists, criminologists, police, and others to study systematically what actually occurs in the interrogation room. This will increase our fund of knowledge, enable us to develop more effective techniques to elicit true confessions from guilty defendants, reduce the likelihood of extracting false confessions from innocent suspects, and provide a stronger basis for systemic policy prescriptions.