Criminal psychology studies why people commit acts that their societies have defined as crimes, how those acts can be predicted, investigated, adjudicated, and in some cases prevented, and what the psychological consequences of crime are for offenders, victims, and the communities that contain them. It sits at the intersection of clinical psychology, developmental science, social psychology, law, and criminology, and it inherits the difficulties of each. It cannot proceed purely as a natural science, because what counts as a crime is partly a political fact about a particular society at a particular time. Nor can it proceed purely as a social science, because individual differences in temperament, neurodevelopment, and cognitive functioning genuinely matter for who becomes involved in criminal behaviour and how. The field’s productive tension is that it has to hold both frames simultaneously.

The first reason criminal psychology is a psychological problem — rather than only a legal or sociological one — is that the mental life of the offender is not optional to understanding the act. A homicide committed in the grip of a first-break psychotic episode, a homicide executed coldly for contract payment, a homicide that erupts out of an escalating domestic argument, and a homicide committed by a sixteen-year-old following an older peer into a robbery gone wrong are legally similar events and psychologically very different ones. They involve different developmental histories, different proximal emotional states, different decision processes, and different implications for how one would intervene. To treat all four as interchangeable is to misunderstand what has happened and to misdesign the response.

The second reason is that the victim’s experience, the witness’s memory, the interrogating officer’s assumptions, the juror’s attributions, and the sentencing judge’s intuitions about dangerousness are all psychological phenomena, and each has been shown to be systematically fallible in ways that affect outcomes. Eyewitness misidentification contributes to a large fraction of the wrongful convictions later overturned by DNA evidence. Confessions obtained under certain interrogation conditions are, counter-intuitively, sometimes false. Jurors’ intuitions about who looks credible track face shape and accent more than truth. Risk-assessment instruments developed in one population generalise imperfectly to another. The criminal-justice process is shot through with judgement, and judgement is the subject matter of psychology.

The third reason is cultural. The modern discipline of forensic psychology took shape largely in North America, the United Kingdom, and continental Europe, and its assumptions are partly local. Adversarial trials, jury decision-making, insanity defences, and structured risk-assessment tools are institutional features that a student reading the Anglo-American textbook tradition can easily mistake for universal facts about how crime is handled. China has developed its own criminal-psychology tradition — 犯罪心理学 — rooted in different institutions: an inquisitorial rather than adversarial procedure, professional judges rather than lay juries, research concentrated in public-security universities and Ministry of Justice institutes, and a more centralised state role in defining both crime and rehabilitation. This course treats mainland Chinese criminal psychology as a co-equal case alongside the Anglophone mainstream, not as an exotic appendix. The comparison is analytically productive: it lets us see which findings are about human beings and which are about particular legal cultures.

This is an explanatory course, not a true-crime entertainment course. It will discuss violent offences, sexual offences, and the psychology of extreme antisocial behaviour, but it will do so without the sensationalism that the genre encourages. Students who enter expecting to be thrilled by serial-killer biographies will find the material disappointing. Students who enter wanting to understand why ordinary developmental trajectories occasionally end in catastrophic outcomes, and why some of the intuitions people hold about criminals are empirically wrong, will find the material demanding but rewarding. The field’s purpose is not to entertain; it is to reduce the harm that crime produces, and that purpose is served by analysis, not by fascination.

Sources and References

Anglophone Textbooks and Reference Works

• Bartol, C. R., & Bartol, A. M. (2021). Criminal Behavior: A Psychological Approach (12th ed.). Pearson.
• Howitt, D. (2018). Introduction to Forensic and Criminal Psychology (6th ed.). Pearson.
• Blackburn, R. (1993). The Psychology of Criminal Conduct: Theory, Research and Practice. Wiley.
• Andrews, D. A., & Bonta, J. (2010). The Psychology of Criminal Conduct (5th ed.). Anderson.
• Hollin, C. R. (2013). Psychology and Crime: An Introduction to Criminological Psychology (2nd ed.). Routledge.
• Canter, D., & Youngs, D. (2009). Investigative Psychology: Offender Profiling and the Analysis of Criminal Action. Wiley.
• Raine, A. (2013). The Anatomy of Violence: The Biological Roots of Crime. Pantheon.
• Hare, R. D. (1999). Without Conscience: The Disturbing World of the Psychopaths Among Us. Guilford Press.
• Cleckley, H. (1941/1976). The Mask of Sanity (5th ed.). Mosby.
• Loftus, E. F. (1996). Eyewitness Testimony. Harvard University Press.
• Maruna, S. (2001). Making Good: How Ex-Convicts Reform and Rebuild Their Lives. American Psychological Association.
• Ward, T., & Maruna, S. (2007). Rehabilitation: Beyond the Risk Paradigm. Routledge.
• Sampson, R. J., & Laub, J. H. (1993). Crime in the Making: Pathways and Turning Points Through Life. Harvard University Press.

Key Research Programmes and Articles

• Moffitt, T. E. (1993). Adolescence-limited and life-course-persistent antisocial behavior: A developmental taxonomy. Psychological Review, 100, 674–701.
• Farrington, D. P., and the Cambridge Study in Delinquent Development — a longitudinal cohort followed from age 8 into middle age.
• Kassin, S. M., and colleagues on interrogation and false confessions.
• Gudjonsson, G. H., on interrogative suggestibility and the psychology of confessions.
• Hare, R. D., on the development of the Psychopathy Checklist–Revised (PCL-R).
• Patrick, C. J., on the triarchic model of psychopathy (boldness, meanness, disinhibition).
• Raine, A., Yang, Y., and colleagues on prefrontal and amygdala abnormalities in antisocial populations.
• Caspi, A., Moffitt, T. E., and colleagues on the MAOA-by-maltreatment gene–environment interaction (2002, Science).
• Akers, R. L., on social-learning theory of crime.
• Agnew, R., on general strain theory.
• Hirschi, T., and Gottfredson, M., on social control and self-control theories.
• Selected articles from Law and Human Behavior, Criminal Justice and Behavior, Psychology, Crime & Law, Journal of Interpersonal Violence, Aggression and Violent Behavior, and Journal of Research in Crime and Delinquency.

Chinese Criminal Psychology Scholarship

• 罗大华 (Luo Dahua), 《犯罪心理学》 — the foundational Chinese textbook, multiple editions, long used at 中国政法大学 and across the public-security university system.
• 刘健清 (Liu Jianqing), 中国政法大学, 犯罪心理学 public lecture series (33 讲) — a widely circulated introduction to the field’s Chinese articulation.
• 李玫瑾 (Li Meijin), 中国人民公安大学, works on adolescent crime, criminal profiling, and family upbringing; public lectures on 《犯罪心理解析》.
• 梅传强, 《犯罪心理学》, legal-psychology tradition at 西南政法大学.
• 林秉贤, textbook contributions on criminal psychology and offender classification.
• 张保平, 中国人民公安大学, on interrogation psychology and criminal investigative psychology.
• 中国人民公安大学《犯罪心理解析》public lecture series and associated research output.
• Publications of the Ministry of Public Security research institutes and of 中国政法大学犯罪心理学研究中心.

Comparable University Courses

• John Jay College of Criminal Justice, CUNY: PSY 242 Psychology of Criminal Behavior.
• Simon Fraser University: PSYC 355 Forensic Psychology.
• University of Cambridge, Institute of Criminology: Psychological Criminology and associated MPhil seminars.
• University of Portsmouth: MSc Forensic Psychology (British Psychological Society accredited).
• UCLA: PSYCH 137G Psychology and Law.
• Stanford University: PSYCH 155 Psychology and the Law (and related offerings through Stanford Law School).
• 中国政法大学: 《犯罪心理学》 undergraduate and graduate courses.
• 中国人民公安大学: 《犯罪心理学》, 《犯罪心理画像》, 《讯问心理学》.

Chapter 1: Defining the Field

Criminal psychology is not a single discipline with a single subject matter. It is a cluster of overlapping research programmes and professional practices that share a common object — the psychology of criminal behaviour and of the criminal-justice process — but diverge in method, in institutional home, and in what they take the most important questions to be. Before we can build a theory of why people commit crimes, we need to be clear about what the field claims to do and what it does not.

Criminal Psychology, Forensic Psychology, and Criminology

Three terms get used in overlapping ways, and the distinctions are worth keeping straight. Criminology is the broad interdisciplinary study of crime as a social phenomenon; it draws on sociology, economics, law, and statistics as much as on psychology, and much of its classical tradition — Durkheim on anomie, Sutherland on differential association, Merton on strain — is sociological. Forensic psychology is the application of psychological science to legal questions: competency to stand trial, risk assessment, child-custody evaluation, jury consultation, expert testimony. Criminal psychology sits between the two. It asks specifically psychological questions about criminal behaviour — how it develops, what traits predict it, how offenders think and feel before, during, and after their offences — and it draws on both criminological theory and forensic practice.

In the Anglophone world these distinctions are sometimes treated as significant and sometimes collapsed. A department of psychology may call the course Psychology and Law or Forensic Psychology and teach roughly the same material. In the Chinese academic system, the distinctions map onto institutional homes: 犯罪学 (criminology) tends to live in law schools and policy institutes; 司法精神病学 (forensic psychiatry) is a medical specialty; 犯罪心理学 (criminal psychology) is taught both in law schools (notably 中国政法大学) and in the public-security university system (notably 中国人民公安大学), where it is closely connected to police training and investigation.

Crime as a Moving Target

A second preliminary point is that the category of “crime” is not fixed. It is defined by legislatures and adjudicated by courts, and both change. Homosexuality was a serious crime in most Anglophone jurisdictions within living memory and is no longer. Corporate misconduct that would have produced no legal consequence fifty years ago can now produce billion-dollar settlements. Domestic violence was historically treated as a private matter in most jurisdictions and is now routinely prosecuted. In China, the 1979 criminal code and its 1997 revision reorganised the boundary of the criminal in significant ways, and continuing amendments — on cybercrime, on sexual offences against minors, on corruption — have reshaped it further.

This means that a psychological theory of “what makes people commit crimes” is always a theory about people who violate a particular legal code at a particular historical moment. The question of why someone steals is not quite the same question as why someone violates the specific property laws of the Criminal Code of Canada as amended in 2024. For most ordinary offences — violence against persons, theft, sexual assault — there is enough cross-jurisdictional and cross-historical consistency in what counts as wrong that the psychological analysis travels reasonably well. For other offences — drug possession, vice offences, political crimes, many regulatory offences — the law does more of the heavy lifting, and the psychological category is correspondingly less stable.

Four Levels of Explanation

A useful habit in criminal psychology, borrowed loosely from Tinbergen’s four questions in ethology, is to keep four explanatory levels distinct.

Level 1: Biological and neurodevelopmental factors. Genetic variation, perinatal insult, head injury, prefrontal and amygdala abnormalities, autonomic under-arousal, and neurochemical differences all have documented associations with antisocial outcomes. None of these is a “crime gene” or a “criminal brain,” but they contribute to the distribution of dispositions from which criminal behaviour sometimes emerges.

Level 2: Individual psychology. Personality traits (especially callous-unemotional traits, impulsivity, and low constraint), cognitive style (hostile attribution bias, poor perspective-taking), emotion regulation, moral reasoning, and self-concept shape how a person navigates temptation, provocation, and opportunity.

Level 3: Social environment. Family structure, parenting practices, peer networks, school environment, neighbourhood disadvantage, and local cultures of violence determine what provocations are encountered, what responses are modelled, and what sanctions are visible.

Level 4: Institutional and macro context. Legal codes, policing practices, economic conditions, gun availability, media environment, and state capacity set the stage on which the lower-level dynamics play out. The difference between a drunken argument ending in a fatal shooting and a drunken argument ending in a hospital visit often depends less on the psychology of the participants than on whether a firearm is within reach.

A serious explanation of an offence usually touches all four levels. A weakness in much public discussion of crime — and of much early criminological theory — is the reduction of a four-level problem to one level: “it’s genetics,” “it’s bad parenting,” “it’s poverty,” “it’s guns.” Each of these is sometimes partially right and never fully right by itself.

Chapter 2: Historical Foundations

Modern criminal psychology carries the marks of its nineteenth- and twentieth-century origins, and some of its recurring disputes — nature versus nurture, the born criminal versus the produced criminal, the clinic versus the statistical actuarial table — recapitulate debates that began long before the field had its present name.

Lombroso and the Italian Positivist School

Cesare Lombroso, a nineteenth-century Italian physician, is the usual starting point. His 1876 book L’uomo delinquente (The Criminal Man) proposed that a substantial portion of criminals were born criminals — atavistic throwbacks to an earlier evolutionary type, identifiable by physical stigmata such as asymmetrical faces, unusual skull shapes, and long arms. Lombroso collected skull measurements in Italian prisons and reported that they differed systematically from those of non-criminals. His student Enrico Ferri and his colleague Raffaele Garofalo rounded out what became known as the Italian positivist school of criminology, which held that crime could be studied scientifically by measuring the characteristics of criminals.

Almost every specific claim Lombroso made turned out to be wrong. His measurements did not replicate when Charles Goring, in a 1913 English study of three thousand convicts, used proper comparison groups. His atavism theory had no evolutionary plausibility. His physical stigmata confused the effects of poverty, nutrition, and occupational injury with innate criminal type. The school is mostly remembered now as a cautionary example of what happens when a theoretically confident scientist collects data without adequate controls.

But two elements of Lombroso’s programme survived. The first is the idea that criminal behaviour has partly biological roots — an idea that was discredited for most of the twentieth century and has returned, in a far more careful form, through the neuroscience of antisocial behaviour. The second is the commitment to studying offenders as individuals with measurable characteristics, rather than as abstract legal cases. Both are genuinely part of the field today.

Early Twentieth Century: From Born Criminal to Psychosocial

The early twentieth century saw the rejection of strict biological determinism and the rise of psychosocial theories. Psychoanalytic accounts — Freud’s, and later August Aichhorn’s and Kate Friedlander’s work on delinquency — treated antisocial behaviour as the expression of unresolved unconscious conflict, often rooted in disturbed early relationships. These accounts are now mostly of historical interest, but they pioneered the idea that criminal behaviour is developmentally produced rather than inborn, and they fed the mid-century clinical interest in offender treatment.

At the same time, the Chicago School of sociology, led by figures such as Robert Park, Ernest Burgess, Clifford Shaw, and Henry McKay, was mapping crime onto urban geography. Their finding that delinquency rates concentrated in specific zones of the city regardless of which ethnic groups occupied those zones at a given time was an early and powerful demonstration that crime is produced at least partly by environmental conditions, not by the characteristics of particular populations. This insight eventually became part of the social-disorganisation tradition that still informs contemporary urban criminology.

By mid-century, behaviourist and social-learning accounts had displaced psychoanalysis as the dominant psychological framework. Albert Bandura’s work on observational learning, especially the Bobo-doll studies on the imitation of aggression, suggested that antisocial behaviour was acquired through the same mechanisms as any other learned behaviour. Ronald Akers would later integrate social learning explicitly into criminology through his differential-association-reinforcement theory.

The Chinese Tradition

Chinese criminal psychology did not develop as a straightforward import of the Western field. It drew on Western sources, particularly Soviet and later American scholarship, but it was shaped by its own institutional context. A distinctive Chinese academic discipline of 犯罪心理学 took clearer form in the reform era after 1978, when legal and psychological research both revived after the Cultural Revolution. Earlier work had existed — Republican-era scholars had translated European sources, and some legal-psychology writing appeared in the 1950s — but the continuous modern discipline traces to the late 1970s and 1980s.

罗大华 (Luo Dahua), based at 中国政法大学, produced the first widely used contemporary Chinese textbook on criminal psychology, 犯罪心理学, which went through multiple editions and became the standard reference across Chinese law schools and public-security universities. His framework was broadly integrative — drawing on personality theory, developmental psychology, and social-environmental analysis — and it established the vocabulary in which subsequent Chinese work would be conducted. 刘健清 (Liu Jianqing), also at 中国政法大学, developed a widely circulated 33-lecture public course that served as the teaching template for a generation.

The institutional geography of Chinese criminal psychology is worth noting. Research and teaching are concentrated in four types of institution: (1) the 政法大学 system (notably 中国政法大学, 西南政法大学, 华东政法大学), which houses legal-psychology research linked to the courts and prosecution; (2) the 公安大学 system (notably 中国人民公安大学), which trains police officers and emphasises investigative applications such as 犯罪心理画像 (criminal psychological profiling) and 讯问心理学 (interrogation psychology); (3) Ministry of Justice and Ministry of Public Security research institutes; and (4) psychology departments in comprehensive universities (such as 中国人民大学, 北京大学, 北京师范大学), which often produce the more experimentally oriented work.

The result is a field with a different centre of gravity from the Anglophone one. Anglo-American criminal psychology is closely tied to clinical forensic practice and to adversarial expert-witness work; Chinese criminal psychology is more closely tied to police training and to the inquisitorial administration of criminal justice. Neither is more or less scientific; they are shaped by the institutional demands of the systems within which they operate.

Chapter 3: Theories of Criminal Behaviour

No single theory explains crime. The phenomenon is too heterogeneous — the processes that produce a habitual shoplifter, a serial sex offender, a white-collar embezzler, and an adolescent gang member are not the same — and the causal chain runs through too many levels. What the field has instead is a toolkit of theories, each capturing some part of the phenomenon, and the skill of applying each to the offences where it fits best.

Biological and Neurodevelopmental Theories

The modern biological tradition, unlike Lombroso’s, does not claim that there is a criminal type. It claims, more modestly, that variation in genes, brain structure, and physiological functioning contributes to variation in the dispositions — impulsivity, emotional reactivity, callousness, executive control — that in turn contribute to antisocial outcomes.

Behavioural genetic studies, particularly twin and adoption studies, consistently estimate moderate heritability (typically 40–60%) for antisocial behaviour, with higher heritability for persistent, severe, and early-onset forms. The heritable component is not for “crime” as such; it is for the trait distributions from which criminal behaviour can emerge. Environmental factors, both shared (family-level) and non-shared (unique to the individual), account for the remainder and interact pervasively with the genetic factors.

One of the most influential gene–environment interaction findings came from Avshalom Caspi, Terrie Moffitt, and colleagues in a 2002 Science paper. Using a New Zealand birth cohort (the Dunedin study), they showed that a common polymorphism in the MAOA gene (which codes for monoamine oxidase A, an enzyme that degrades neurotransmitters) moderated the effect of childhood maltreatment on adult antisocial behaviour. Boys with the low-activity MAOA variant who were also maltreated showed substantially elevated rates of later antisocial behaviour; boys with the same genetic variant who were not maltreated did not. Neither factor alone was sufficient; the combination was. This finding, much replicated with varying results, captures the modern biological framing: genes matter because of how they shape responsiveness to environments, not because they operate in isolation.

Neuroimaging work, notably the programme of Adrian Raine and colleagues, has identified structural and functional differences in antisocial populations. Reduced prefrontal grey matter volume, amygdala abnormalities, and autonomic under-arousal (lower resting heart rate, reduced skin-conductance response) have been found in samples of violent offenders and in community samples with high antisocial scores. Raine’s The Anatomy of Violence (2013) synthesises this literature and makes the reasonable case that a proportion of serious persistent violence has neurobiological correlates that can, in principle, inform prevention. The reasonable reservations — that brain correlates are not deterministic, that imaging studies are often underpowered, that the direction of causation is sometimes unclear — apply and must be kept in mind.

Personality and Psychopathy

A second psychological tradition focuses on stable individual differences that predict criminal behaviour. Impulsivity, low conscientiousness, low agreeableness, sensation-seeking, and callous-unemotional traits have all been studied as predictors. The construct with the strongest explanatory track record for serious and persistent offending is psychopathy, which the course takes up in detail in Chapter 4.

Social-Learning Theory

Albert Bandura’s social-learning theory, as extended to crime by Ronald Akers, holds that criminal behaviour is learned by the same mechanisms as any other behaviour: observation of models, reinforcement of imitations, acquisition of attitudes and justifications through interaction with others. Akers’s differential association–reinforcement theory integrates Edwin Sutherland’s earlier claim that criminal behaviour is learned in intimate personal groups with operant-conditioning principles. The implication — that the people one spends time with substantially shape the probability of offending — is robust; peer effects are among the strongest predictors of adolescent delinquency.

Strain Theory

Robert Merton’s 1938 strain theory argued that societies produce crime when they hold up cultural goals (in the American case, economic success) while restricting legitimate means to those goals for large segments of the population. Those under such strain adopt one of several adaptations, including innovation (pursuing the goals through illegitimate means). Robert Agnew’s general strain theory (1992) broadens the framework, identifying multiple sources of strain — not only thwarted positive goals but loss of valued outcomes and exposure to negative stimuli — and locating the criminogenic effect in the negative emotions (anger, frustration, despair) that strain produces. General strain theory has generated useful empirical work, particularly on adolescent offending.

Control Theory

Travis Hirschi’s 1969 social control theory inverted the usual question. Rather than asking why some people offend, it asked why most people do not. Hirschi’s answer was that most people are restrained by strong bonds to conventional society: attachment to parents and peers, commitment to conventional pursuits (school, work), involvement in legitimate activities, and belief in the moral validity of law. When these bonds are weak, offending becomes more likely. In 1990 Hirschi and Michael Gottfredson extended the theory in A General Theory of Crime, arguing that the central individual difference predicting offending is self-control, a trait they locate in early parenting. The Gottfredson-Hirschi theory has been much criticised for its claim of generality and for its measurement choices, but its core insight — that some people are more capable of inhibiting immediate impulses in service of longer-term interests, and that capability predicts offending across offence types — has empirical support.

Developmental and Life-Course Criminology

Developmental criminology, most influentially expressed in Terrie Moffitt’s 1993 dual taxonomy, distinguishes two main trajectories of antisocial behaviour. Adolescence-limited offenders begin antisocial behaviour around puberty, peak in mid-adolescence, and desist in early adulthood; their behaviour is typically normative for their age cohort, a response to the gap between biological maturity and social adult status, and does not indicate serious underlying pathology. Life-course-persistent offenders show behavioural problems from early childhood, often linked to neurodevelopmental and family-environmental risk factors, and continue antisocial behaviour throughout the life course. The two groups differ not only in trajectory but in aetiology: life-course-persistent offending is more heritable, more neurodevelopmentally rooted, and more likely to involve serious violence; adolescence-limited offending is more social-learning based and typically lower in severity.

David Farrington’s Cambridge Study in Delinquent Development, a longitudinal study of 411 South London boys followed from age 8 into their fifties, has produced much of the best empirical evidence on the development of offending. Farrington’s Integrated Cognitive Antisocial Potential (ICAP) theory synthesises findings from the study into a framework in which long-term antisocial potential (shaped by early family and individual risk) interacts with short-term situational antisocial potential (shaped by opportunities, provocations, and substance use) to produce specific offences.

Robert Sampson and John Laub, reanalysing the Glueck longitudinal data in Crime in the Making (1993) and subsequent work, added a life-course perspective emphasising turning points — marriage, stable employment, military service — that redirect antisocial trajectories. Their work, complementing Moffitt’s more typological view, argues that change is possible throughout adulthood and that the social bonds that produce desistance are at least as important as the dispositions that produce initial offending.

Routine Activities and Rational Choice

The routine activities theory of Lawrence Cohen and Marcus Felson (1979) argues that crime occurs when three elements converge in space and time: a motivated offender, a suitable target, and the absence of a capable guardian. The theory is deliberately silent about why some people are motivated offenders, focusing instead on the opportunity structure within which offending occurs. Related rational choice perspectives (Derek Cornish and Ronald Clarke) model offenders as making bounded rational calculations about costs and benefits of specific offences. These frameworks, sometimes dismissed as shallow, have been remarkably productive in crime-prevention research: designing environments to reduce the convergence of the three elements (street lighting, CCTV, transit design, product-level anti-theft features) has produced measurable reductions in specific offence types.

Chapter 4: Psychopathy and Antisocial Personality

Of all the constructs in criminal psychology, psychopathy is the one with the longest continuous history and the most clinical weight. It is also one of the most contested, because its status — between a personality disorder, a developmental condition, and a descriptive label for a cluster of socially destructive traits — has never fully stabilised.

Cleckley’s Clinical Tradition

The modern construct traces to Hervey Cleckley’s 1941 monograph The Mask of Sanity. Cleckley, working with patients in an American psychiatric hospital, described a type of individual who appeared on surface examination to be normal, charming, and articulate but who on closer acquaintance displayed a consistent pattern of callousness, shallow emotion, unreliability, absence of guilt, and an inability to learn from punishment. The “mask” in the title referred to the deceptive surface normality; the “sanity” was a comment on the patients’ lack of the overt symptoms — delusions, hallucinations, disorganisation — that usually produced psychiatric diagnosis. Cleckley’s prototype was not the violent offender but the slick, charming, socially destructive individual who moved through institutions causing damage without triggering the usual protective responses.

Cleckley’s sixteen clinical criteria — including superficial charm, absence of delusions or nervousness, unreliability, untruthfulness, lack of remorse, poor judgement, pathological egocentricity, shallow affect, specific loss of insight, unresponsiveness in interpersonal relations, and failure to follow any life plan — were descriptive and clinical rather than operationalised for research. They provided the starting point for Robert Hare’s systematic measurement programme.

Hare and the PCL-R

Robert Hare, beginning his work in Canadian prisons in the 1970s, developed the Psychopathy Checklist and its 1991 revision, the Psychopathy Checklist–Revised (PCL-R). The PCL-R is a structured clinical assessment, scored by a trained rater on the basis of a semi-structured interview plus file review, yielding a score from 0 to 40 on twenty items. The items load onto two factors: Factor 1 (interpersonal-affective traits — glibness, grandiosity, pathological lying, manipulation, lack of remorse, shallow affect, callousness, failure to accept responsibility) and Factor 2 (social deviance — need for stimulation, parasitic lifestyle, poor behavioural control, early behavioural problems, impulsivity, irresponsibility, juvenile delinquency, revocation of conditional release, criminal versatility). More recent models unpack these into three or four facets.

A PCL-R cutoff of 30 in North American correctional samples identifies roughly the top 15–25% of inmates as psychopathic; in community samples the prevalence is substantially lower. The PCL-R has accumulated strong evidence for reliability when administered by trained raters, for predictive validity with respect to violent recidivism, and for association with a range of laboratory markers (deficient fear conditioning, reduced startle modulation, atypical moral-judgement patterns). It is one of the most widely used instruments in forensic psychology and has been translated and normed for Chinese correctional populations, though the cross-cultural stability of the construct is a continuing research question.

Psychopathy and Antisocial Personality Disorder

The DSM-5 diagnosis of Antisocial Personality Disorder (ASPD) overlaps with psychopathy but is not identical. ASPD is defined primarily by behavioural criteria — repeated criminal acts, deceitfulness, impulsivity, irritability, disregard for safety, irresponsibility, lack of remorse — and a history of conduct disorder before age 15. A high proportion of incarcerated offenders meet ASPD criteria; a smaller proportion meet PCL-R psychopathy criteria. Psychopathy, as Hare operationalises it, requires the interpersonal-affective Factor 1 traits that ASPD does not emphasise. The two constructs identify overlapping but distinguishable populations, and the research literature has generally found that psychopathy predicts violent recidivism more strongly than does ASPD alone.

Neurobiology

Psychopathy has been associated with reduced amygdala responsiveness to fearful stimuli, impaired fear conditioning, reduced autonomic reactivity, and altered connectivity between prefrontal and limbic regions. James Blair’s developmental model argues that psychopathic individuals have a specific deficit in the processing of distress cues — fear and sadness in others — that in typical development supports empathy and moral socialisation. The deficit, on this model, does not prevent learning in general; it prevents the specific kind of learning that normally restrains aggression by linking it to the distress it causes.

Christopher Patrick’s triarchic model unpacks psychopathy into three partly distinct dimensions: boldness (social dominance, fearlessness, stress resilience), meanness (callousness, lack of empathy, predatory exploitation), and disinhibition (impulsivity, poor regulation). The model helps explain why psychopathy-spectrum individuals are heterogeneous and why some combinations (high boldness, low meanness, low disinhibition) may be relatively adaptive in some social roles while others (high meanness, high disinhibition) are strongly associated with violent offending.

Treatment Pessimism and Its Reconsideration

Psychopathy has long carried a reputation for treatment resistance. Early studies, particularly a controversial Canadian evaluation of a therapeutic community programme, reported that treatment was associated with higher rates of violent recidivism in psychopathic participants, apparently because the treatment gave them additional interpersonal skills they then used manipulatively. These findings, now understood as partly artefactual, reinforced a clinical culture of pessimism about treating adult psychopaths.

More recent research has tempered the pessimism in two directions. First, interventions with children and adolescents displaying callous-unemotional traits show more promise than interventions with adults, consistent with general developmental principles about the malleability of personality. Second, structured, cognitive-behavioural, risk-responsivity-matched programmes for adults have shown modest positive effects even with high-PCL-R offenders, particularly on Factor 2 (social-deviance) behaviours, even when Factor 1 traits remain stable. The revised consensus is that adult psychopathy is difficult but not impossible to treat, and that the claim of complete untreatability is an overstatement.

Chapter 5: Violent Offending

Violence is not a single phenomenon. The psychology behind a planned armed robbery is different from the psychology behind a drunken brawl, which is different from the psychology behind a fatal domestic argument, which is different from the psychology behind sexual violence. The field’s most useful distinctions cut across rather than within legal categories.

Instrumental Versus Reactive Violence

A long-standing distinction separates instrumental (or proactive, predatory) violence from reactive (or affective, hostile) violence. Instrumental violence is goal-directed, relatively cool, planned, and used as a means to some other end — money, status, elimination of a witness, enforcement of a criminal agreement. Reactive violence is defensive, hot, unplanned, and emerges in response to perceived provocation; it is accompanied by strong arousal, often by anger or fear, and it is typically regretted afterwards.

These are poles of a continuum rather than discrete types. Many offences mix features of both. But the distinction matters: instrumental violence is more strongly associated with psychopathy and with persistent criminal careers, while reactive violence is more strongly associated with emotional dysregulation, hostile attribution bias, and substance use. The interventions that plausibly reduce them differ: instrumental violence is reduced mainly by altering incentive structures and by incapacitation of persistent offenders, while reactive violence is reduced by teaching emotion regulation, altering attention to provocation cues, and reducing access to lethal means.

Homicide

Homicide is rare relative to other violent offences, but it is the offence around which public imagination of crime is often built. Cross-nationally, homicide rates vary by orders of magnitude. The United States has a homicide rate several times that of most comparable high-income countries; Canada’s rate is roughly one-third of the US rate; China’s reported rate is substantially lower still, though measurement comparability across jurisdictions is imperfect.

Much of the cross-national variation is explained by three factors: the density and lethality of interpersonal violence in general, the availability of firearms, and the organisation of the drug economy. The United States’ exceptional homicide rate among developed countries is driven largely by firearm homicides, particularly in settings where illegal drug markets and disadvantaged-neighbourhood concentrations interact. Where firearms are not readily available to offenders, the equivalent disputes more often end in serious assault rather than death.

Within any jurisdiction, homicide is heterogeneous. Intimate-partner homicides (disproportionately female victims), altercation-escalation homicides (disproportionately young-male participants), felony-circumstance homicides (occurring during another crime such as robbery), and rarer patterns (contract killing, serial killing) have different psychological and situational dynamics. Serial killing, despite its cultural prominence, accounts for a tiny fraction of homicide deaths and is massively overrepresented in public discussion relative to its actual frequency.

Intimate Partner Violence

Intimate partner violence (IPV) is one of the most common forms of violence in both Anglophone and Chinese societies, and it is one of the most underreported. The psychology is different from stranger violence. IPV typically involves repeat victimisation of the same target, a pattern of coercive control that may or may not involve physical violence, and a complex mixture of attachment, jealousy, substance use, and gender-role enforcement. Michael Johnson’s distinction between intimate terrorism (a pattern of coercive control, usually male-perpetrated, in which violence is one of several tools used to subordinate a partner) and situational couple violence (bidirectional, less severe, emerging from unresolved conflict) has become standard in the field.

IPV is one of the clearest examples of how legal definitions and cultural understandings of crime shift. It was treated, across most jurisdictions including both Canada and China, as a private matter well into the twentieth century. The legal reframing as a crime subject to police and prosecutorial response is relatively recent (in China, the Anti-Domestic Violence Law of 2015 was a significant formal shift, though implementation remains uneven). Psychological research on IPV has tracked — and in some cases driven — this shift, producing evidence on victim psychology, perpetrator typologies (generally-violent/antisocial, borderline/dysphoric, family-only), and intervention effectiveness.

Sexual Offending

Sexual offending is a heterogeneous category that resists a single psychological explanation. Richard Laws and William O’Donohue’s work, Karl Hanson’s meta-analyses of risk factors, and the development of instruments such as the Static-99R for risk assessment have produced a reasonably well-characterised empirical landscape. Recidivism rates, contrary to popular belief, are generally lower than for many other offence types when measured by reconviction over standard follow-up periods; the popular perception of uniformly high sexual-offence recidivism is not supported by the evidence.

Typologies distinguish, for example, between rapists (motivated by anger, power, sadism, or opportunism) and child-sex offenders (with further subdivisions by age preference, gender preference, and degree of fixation versus regression), and recognise that these subtypes differ in risk factors, criminogenic needs, and responsiveness to intervention. The RNR (risk-needs-responsivity) framework, discussed in Chapter 12, has been applied to sexual offending with modest but real positive effects.

Juvenile Violence

Juvenile violence, mostly committed by adolescent males, is a significant component of the total violence picture, and much of it fits Moffitt’s adolescence-limited trajectory — peaking in mid-adolescence, diminishing by the early twenties, carried out in groups rather than alone, often under the influence of peers and substances. A smaller subgroup, fitting the life-course-persistent pattern, commits disproportionately severe and persistent violence; identification and intervention with this subgroup is a focus of developmental prevention research.

Cross-culturally, juvenile-violence patterns differ in ways that reflect institutional context. In many Chinese urban settings, youth violence involving firearms is extremely rare because firearms are heavily restricted; school violence and brawls involving knives are more characteristic. In parts of the United States, youth homicide involving handguns is a major component of the total. The underlying psychology — status competition, peer dynamics, identity formation, family stress — is similar; the outputs differ because the available means differ.

Cross-Cultural Differences in Rates and Forms

The violent-crime picture differs substantially between the United States, Canada, and mainland China, and understanding why is one of the most instructive exercises in criminal psychology.

The United States has a homicide rate of roughly 5–7 per 100,000 per year, high among developed countries, concentrated in specific urban and demographic segments, and dominated by firearm homicide. Canada has a homicide rate of roughly 1.7–2 per 100,000, with a smaller firearm share. Mainland China’s official homicide rate is reported in the range of 0.5–1 per 100,000, with very low firearm involvement due to strict firearm controls; the rate has declined over the past two decades. No single factor fully explains these differences, but firearm availability, economic inequality, drug-market organisation, and state capacity for social control all contribute.

One important interpretive point: lower rates of lethal violence in a jurisdiction do not necessarily reflect lower rates of underlying interpersonal conflict. They can reflect differences in the lethality of the means available when conflict does erupt. The criminological literature on firearm availability and homicide — by Franklin Zimring, Philip Cook, and others — repeatedly finds that where similarly aggressive interpersonal encounters occur with versus without ready firearm access, the death rate differs substantially. Chinese criminal psychologists working within the domestic context observe, correspondingly, that the ecology of violence they study is shaped by a set of institutional choices about means-access that differ sharply from the American case.

Chapter 6: Mental Disorder and Crime

The relationship between mental disorder and crime is one of the areas where public intuition and empirical evidence diverge most sharply. The popular image — that the mentally ill are disproportionately dangerous — is largely wrong in the strong form in which it is usually stated, but there are specific, narrower relationships that are real and important.

Base Rates and the Myth of Dangerousness

The great majority of people with mental disorders, including people with severe disorders such as schizophrenia and bipolar disorder, do not commit violent acts. The great majority of violence is committed by people without diagnosable mental disorder. Meta-analyses of the mental-disorder–violence relationship, most influentially by Seena Fazel and colleagues, find that while certain disorders elevate risk modestly, the absolute risk of violent offending among people with severe mental illness remains low in absolute terms. Untreated substance use, by contrast, has a much stronger association with violence than any non-substance psychiatric disorder, and the combination of psychiatric disorder with substance use has a substantially higher associated risk than either alone.

Public discourse routinely confuses base rate with relative risk. If schizophrenia doubles the probability of violent offending (a reasonable estimate for some populations), the absolute probability is still low — low single-digit percentages over several years for the large majority of patients — and the population of people without schizophrenia produces far more total violence than the population with it. A mental-health-centric theory of violence is wrong on arithmetic grounds.

Specific Clinical Conditions

Within these limits, certain clinical conditions have identifiable relationships with offending.

Schizophrenia and related psychotic disorders: the modest elevation in violence risk is concentrated in specific circumstances — acute psychotic episodes with persecutory delusions or command hallucinations, untreated substance co-use, and a small subgroup of patients with comorbid antisocial traits. Adequate treatment substantially reduces the elevation.

Mood disorders: depression is associated with some suicidal and some aggressive behaviour but is not a major driver of violent offending. Bipolar disorder in manic phases is associated with impulsive aggression and risky behaviour. Post-partum depression and psychosis are associated with rare but serious offences including filicide.

Substance use disorders: across populations, these are the largest psychiatric contributor to crime, partly through acute intoxication (disinhibition, impaired judgement), partly through long-term effects on impulse control and affect regulation, and partly through the involvement in illegal markets that drug use often entails.

Personality disorders: beyond antisocial and psychopathic personality (discussed in Chapter 4), borderline personality disorder has associations with self-directed aggression and with reactive interpersonal violence; narcissistic personality disorder with certain patterns of interpersonal exploitation. The associations are real but heterogeneous.

Neurodevelopmental conditions: attention-deficit/hyperactivity disorder and some autism-spectrum presentations are associated with modest elevations in some offence types, usually through impulsivity and social-cognitive differences rather than through any direct “criminogenic” mechanism. Intellectual disability is associated with vulnerability to exploitation and, in some cases, to inappropriate behaviour that may be criminalised despite reduced culpability.

The Causal Structure

The causal relationship between mental disorder and crime is bidirectional and confounded. Some mental disorders elevate the probability of offending through specific mechanisms (delusions, disinhibition, emotional dysregulation). Involvement in crime, conversely, elevates the probability of later mental disorder through exposure to violence, incarceration, substance use, and social disruption. And both mental disorder and offending share common upstream causes — childhood adversity, poverty, social dislocation, substance use — so a simple causal arrow from one to the other misrepresents the field.

For criminal-psychology practice, the upshot is that most offenders are not primarily driven by mental disorder, and that most mentally ill people are not offenders, but that the specific intersection — particularly of severe mental illness with untreated substance use, or of acute psychosis with persecutory content — does produce a disproportionate number of the most tragic and least preventable cases. Criminal-justice systems that fail to distinguish the large majority of non-dangerous mentally ill people from the small, genuinely high-risk subgroup produce both unnecessary stigma and unnecessary risk.

Chapter 7: Juvenile Delinquency and Developmental Pathways

Juvenile delinquency is where criminal psychology meets developmental psychology most directly. The overwhelming majority of offenders begin offending in adolescence, and most stop within a few years. The question is less “why do some adolescents offend?” — a lot of them do — and more “why do some persist, and what differentiates them from those who desist?”

The Age–Crime Curve

Across populations, eras, and jurisdictions, the age–crime curve has a remarkably consistent shape: offending rises sharply in early adolescence, peaks in mid- to late adolescence, and declines steadily thereafter. This pattern has been documented in Western societies, in Japan, and, where data permit, in China. Its robustness is one of the central puzzles of criminology, and different theories explain it differently: Hirschi and Gottfredson locate it in maturation of self-control; Moffitt in the closing of the maturity gap; Sampson and Laub in the accumulation of adult social bonds.

Moffitt’s Taxonomy Revisited

As introduced in Chapter 3, Moffitt’s dual taxonomy distinguishes adolescence-limited and life-course-persistent offenders. The former begin offending around puberty as a social-mimicry response to the gap between biological maturity and adult social status, engage in relatively low-severity offences alongside peers, and desist as adult roles open up. The latter show behavioural problems from early childhood, often in interaction with neurocognitive deficits and adverse family environments, and accumulate a persistent criminal trajectory. The two groups differ in prevalence (adolescence-limited is far more common), in severity (life-course-persistent disproportionately accounts for the serious and violent offences), and in response to intervention.

Subsequent work has refined the taxonomy. A third group — abstainers, who do not offend at all in adolescence — appears to have distinctive characteristics (unusually high self-control, strong parental relationships, sometimes social withdrawal). A fourth group — late-onset offenders, who begin in early adulthood — is less well-characterised and may represent a heterogeneous mix.

Risk and Protective Factors

Longitudinal studies — particularly Farrington’s Cambridge Study, the Pittsburgh Youth Study, the Dunedin study, and the Rochester Youth Development Study — have identified a fairly stable set of risk factors for persistent juvenile offending. These cluster at four levels.

At the individual level: early conduct problems, low self-control, hyperactivity, low IQ, early substance use. At the family level: harsh, inconsistent, or neglectful parenting; parental criminality or substance use; family conflict; lack of supervision. At the peer level: association with delinquent peers, gang involvement, peer rejection in childhood followed by association with similarly rejected peers in adolescence. At the neighbourhood/school level: school failure, neighbourhood disorganisation, concentrated disadvantage, exposure to community violence.

Protective factors — those associated with lower risk of offending — include high IQ, easy temperament, strong attachment to at least one prosocial adult, school engagement, religious involvement in some contexts, and what Werner and Smith’s Kauai longitudinal study called the presence of “turn-around” relationships in adolescence.

Chinese Perspectives

Chinese research on juvenile delinquency, notably the work of 李玫瑾 (Li Meijin) at 中国人民公安大学 and a substantial textbook literature on 青少年犯罪, has emphasised family upbringing and early moral education as foundational. Li’s public lectures, widely circulated in mainland media, argue that many later patterns of serious antisocial behaviour have roots in the first three to six years of life, in patterns of attachment, discipline, and emotional responsiveness within the family. The framework shares much with international developmental criminology but puts heavier emphasis on family-level moral formation than does the Anglophone mainstream, reflecting both Chinese cultural assumptions about family responsibility and the institutional absence of the large residential-care and foster-care systems that shape some Western juvenile pathways.

The cross-cultural picture of juvenile offending also differs in its forms. Western juvenile offending involves a larger proportion of firearm-related offences, drug-market-related offences, and violence connected to urban-gang structures. Chinese juvenile offending, as documented in the mainland literature, more often involves theft, group fighting, cyber-offences, and, in a contested recent literature, bullying (校园霸凌) that has received substantial public attention. The underlying developmental processes are similar; the forms express the local ecology.

Chapter 8: Criminal Profiling and Investigative Psychology

Few topics in criminal psychology have had a larger gap between popular perception and scientific reality than criminal profiling. The cultural image, shaped by films and television, is of a semi-mystical analyst who infers the killer’s psychology from the crime scene with uncanny accuracy. The reality is more modest, more uneven, and considerably more interesting.

The FBI BAU Tradition

The modern Anglophone tradition of profiling originated in the Behavioral Science Unit (later the Behavioral Analysis Unit, BAU) of the FBI in the 1970s. Agents such as Robert Ressler, John Douglas, and Roy Hazelwood interviewed a sample of convicted serial murderers and sexual offenders and used the interviews to develop typologies — most famously the organised/disorganised offender distinction — that they applied to unsolved cases. The BAU developed a practice of reviewing crime-scene evidence and producing inferences about likely offender characteristics (age, race, occupation, relationship history, likelihood of repeat offending) that investigators could use to narrow suspect pools.

The BAU’s work entered popular culture through books and films (Mindhunter, The Silence of the Lambs) and was widely imitated by police agencies in other countries. It also attracted substantial academic criticism. The original interview sample was small and unrepresentative, the typologies were based on clinical impression rather than systematic analysis, and the validity of profiles produced by the method was not subjected to rigorous evaluation until relatively late. When evaluations were done, they generally found that experienced profilers were somewhat better than novices at some profile tasks but not consistently better than structured statistical approaches, and that the organised/disorganised dichotomy did not hold up in empirical analyses of crime-scene data.

Canter’s Investigative Psychology

David Canter, working in the United Kingdom from the 1980s, developed an alternative approach that he called investigative psychology. Rather than relying on clinical intuition, Canter’s programme used statistical analysis of large samples of solved cases to identify reliable associations between crime-scene behaviours and offender characteristics. The approach — described in Canter and Youngs’s Investigative Psychology (2009) — treats profiling as an empirical research problem rather than an interpretive art, and it aims to produce falsifiable claims about the correlations investigators can actually rely on.

Investigative psychology’s findings have been mixed. Some offender-characteristic inferences (for example, about the relationship between the offender’s home base and the geographic distribution of offence locations) have held up reasonably well. Other inferences (for example, about personality traits inferred from specific crime-scene behaviours) have proved harder to validate. The value of the approach is less in producing dramatic profile breakthroughs and more in disciplining the inferences that investigators would make anyway, making them explicit, and testing them against data.

Geographic Profiling

Geographic profiling, developed most systematically by Kim Rossmo (a former Canadian police officer turned academic), uses the spatial pattern of a series of offences to estimate the likely home base of the offender. The technique rests on well-documented regularities in offender spatial behaviour: offenders disproportionately commit offences in areas they know, typically within a few kilometres of home, following a distance-decay function, with a buffer zone immediately around the home where offences are less likely (to reduce recognition risk). Geographic profiling, applied to cases with multiple offences, can narrow the search area substantially and has been used operationally by several police agencies.

Chinese Criminal Psychological Profiling

Chinese criminal psychology has developed its own tradition of 犯罪心理画像 (criminal psychological portrait), concentrated in the public-security university system and applied in major-case investigations. The tradition draws on both translated Western sources and indigenous casework, and it tends to be more integrated with interrogation and interview practice than the Anglophone BAU tradition. 李玫瑾’s work, widely circulated in both professional and popular channels, applies developmental and family-history analysis to offender-characteristic inference. 张保平 and colleagues at 中国人民公安大学 have produced training materials that combine profile construction with investigative and interrogation strategies.

The Chinese tradition operates within an inquisitorial criminal-procedure system where profiling is used primarily to guide investigation and interrogation, not to generate expert testimony for adversarial trial. This institutional difference shapes what profiling is used for and what standards of justification it operates under. It is not subjected to the same Daubert-style admissibility challenges that have, in Anglo-American jurisdictions, constrained the use of profile testimony at trial.

Limits of Profiling

The honest assessment of profiling, after four decades of research, is that it is useful but oversold. It can help prioritise leads in complex cases with multiple offences; it cannot reliably produce individually identifying inferences from a single crime scene. It works best as an adjunct to conventional investigation, not as a substitute. Agencies that have treated it as a primary tool have sometimes produced notorious failures — inflating confidence in specific suspects, directing investigative resources away from productive lines of inquiry. The field’s maturation has involved, on the whole, a recognition of these limits.

Chapter 9: Eyewitness Memory and Interrogation

Two of the most reliable findings in the psychology of the criminal-justice process are that eyewitness memory is less accurate than witnesses themselves believe, and that confessions obtained under certain conditions are less reliable than investigators assume. Both findings have substantial operational consequences for how criminal cases are built and evaluated.

Eyewitness Memory

Memory is not a recording. It is a reconstructive process in which perceptual encoding, storage, and retrieval each introduce distortion. Elizabeth Loftus’s research programme, beginning in the 1970s, has documented in dozens of experiments how post-event information can alter the content of memories without the witness’s awareness. In one classic paradigm, participants who viewed a staged car accident and were then asked either “How fast were the cars going when they hit each other?” or “How fast were the cars going when they smashed into each other?” produced different speed estimates and differed in their later reports of having seen broken glass (which was not present). The misinformation effect — the integration of post-event information into the reported memory — is one of the most replicated findings in cognitive psychology.

Loftus’s Eyewitness Testimony (1996) and her subsequent work document several related findings: confident witnesses are not reliably more accurate than less confident ones, particularly after post-event feedback has altered their confidence; cross-race identification is less accurate than same-race identification (the own-race bias); weapons focus reduces the encoding of perpetrator features; line-up procedures that allow for relative rather than absolute judgements produce higher false-identification rates; and system variables (how the police conduct the identification) and estimator variables (the conditions of the original witnessing) both matter, with system variables being the ones investigators can actually control.

The Innocence Project, a legal-reform initiative that has used post-conviction DNA testing to re-examine closed cases, has consistently found eyewitness misidentification implicated in a large fraction of the wrongful convictions it has overturned — typically cited at around 70% of the DNA exoneration cases. This is not because eyewitnesses are commonly malicious; they are not. It is because ordinary eyewitness memory, under the conditions of real crime witnessing, is less accurate than the justice system has traditionally treated it as being.

Interrogation

The second major process-psychology finding concerns interrogation. The dominant Anglo-American interrogation method for much of the twentieth century was the Reid technique, developed by John Reid and associates from the 1940s onward. The Reid technique is a structured nine-step process designed to obtain confessions from suspects believed to be guilty. It relies on confrontation (telling the suspect the evidence is overwhelming), theme development (offering the suspect moral excuses for the behaviour), minimisation (downplaying the severity), and sustained pressure under conditions of suspect isolation. The training emphasises behaviour-symptom analysis — the claim that trained interrogators can reliably distinguish truthful from deceptive suspects through verbal and non-verbal cues.

The research on the Reid technique is damning in specific respects. Behaviour-symptom analysis has not held up in controlled tests; trained interrogators are not much more accurate than untrained ones at detecting deception, and both groups perform modestly above chance. More seriously, the technique produces false confessions in some suspects — particularly juveniles, intellectually disabled individuals, and those with certain personality characteristics — at rates that are not trivial. Saul Kassin and colleagues have documented the dynamics of false confessions, distinguishing voluntary (offered without interrogation pressure), compliant (offered to escape the interrogation even while the suspect knows she is innocent), and internalised (offered after the suspect has come to believe her own guilt) false confessions. The Innocence Project cases include a substantial proportion in which DNA-exonerated individuals had, before trial, confessed.

The PEACE Model

In the United Kingdom, concerns about Reid-style interrogation — particularly after several miscarriages of justice in the 1970s and 1980s — led to the development and adoption of the PEACE model: Planning and preparation, Engage and explain, Account (letting the suspect give a full account), Closure, and Evaluation. The PEACE approach is non-confrontational, oriented toward information gathering rather than confession production, and is used by police in England, Wales, New Zealand, Norway, and elsewhere. Evaluations have found that PEACE produces comparable or higher rates of useful information than confrontational approaches and lower rates of false confession. It has not been universally adopted; North American interrogation practice has been slower to reform.

Chinese Interrogation Psychology

讯问心理学 (interrogation psychology) is a well-developed branch of Chinese criminal psychology, taught systematically at 中国人民公安大学 and in Ministry of Public Security training institutes. The Chinese tradition has its own literature on the psychology of confession, the techniques of structured interview, and the dynamics of suspect cognition. The institutional context is different from the Anglo-American: confessions carry different weight under Chinese criminal procedure, and the training and oversight structures that govern interrogation differ. The Chinese academic literature on interrogation has, like its Anglophone counterpart, grappled with the problem of false confession, particularly following a series of high-profile wrongful-conviction cases (the 呼格吉勒图案 and similar cases) that, beginning in the 2010s, prompted significant reform discussion and procedural changes including the expansion of audio-video recording of interrogations.

Chapter 10: Victimology and the Psychology of Victims

Victimology — the study of crime victims — was, for much of the history of criminology, a peripheral concern. The field focused on offenders. Over the last several decades, the psychology of victimisation has become a much more central topic, driven both by research interest and by the advocacy movements that insisted on the victim’s centrality to the justice process.

Psychological Consequences of Victimisation

Crime victimisation produces a range of psychological sequelae. Post-traumatic stress disorder (PTSD) — re-experiencing symptoms, avoidance, negative alterations in cognition and mood, hyperarousal — is elevated after violent victimisation, particularly sexual assault and severe physical assault. Depression, anxiety disorders, substance use disorders, and difficulties in intimate relationships are more common among victims of serious crime than in demographically matched non-victims. The effects are not uniform; resilience is common, and many victims recover without clinical intervention. But the prevalence of clinically significant symptoms in the months following serious victimisation is substantial.

The psychological picture differs by offence type. Sexual-assault victims face specific patterns including shame, self-blame, and disclosure difficulty. Intimate-partner violence victims, because the perpetrator is an intimate, face ongoing risk assessment and complex attachment dynamics that make the trauma difficult to resolve. Victims of property crime show less severe but still real effects, particularly when the crime involves home invasion, which violates the sense of the home as safe.

The Just-World Belief and Victim Blaming

Melvin Lerner’s just-world hypothesis — the tendency to believe that the world is fundamentally just, that people get what they deserve — has been invoked to explain a durable pattern in which observers attribute responsibility to victims of crime. A just-world believer, confronted with a victimisation that appears random and undeserved, is motivated to reduce the cognitive threat by finding something the victim did wrong. This psychological tendency is one root of victim blaming: the attribution to victims of behaviours or characteristics that are taken to explain or mitigate the offender’s culpability.

Victim blaming is documented across offence types but is particularly pronounced in sexual assault cases, where victims are often questioned about their clothing, their alcohol use, their prior sexual history, and their relationship to the perpetrator — lines of inquiry that would be recognised as absurd if applied to victims of robbery or assault. The research on jury decision-making and on survey attitudes finds that victim-blaming attributions are reduced when participants are prompted to consider the offender’s responsibility explicitly and when the victim’s behaviour is not framed in terms that invite just-world reasoning.

Secondary Victimisation

Secondary victimisation — the additional harm done to victims by the criminal-justice process itself — has been a focus of victimological reform. Victims of sexual assault who report to police may encounter disbelief, intrusive questioning, and long processing delays; victims may face cross-examination at trial that replicates the trauma of the original offence. Reforms including specialised sexual-assault units, victim advocates, restrictions on rape-shield-protected cross-examination, and trauma-informed interviewing have attempted to reduce secondary victimisation without compromising defendants’ due-process rights. The balance is often contested and varies across jurisdictions.

Chinese reform discourse on victimology has focused somewhat differently, with greater emphasis on 恢复性司法 (restorative justice) practices that attempt to restore relationships between offenders and victims within local communities, and with substantial public attention to the treatment of victims in high-profile cases. The institutional context — a system in which prosecutors represent public interest and victims’ procedural role is more limited than in some Western models — shapes how victim-centred reforms are constructed.

Chapter 11: Psychology in the Courtroom

The application of psychology to the courtroom covers a range of practical questions: whether a defendant is competent to stand trial, whether a defendant is criminally responsible given mental illness at the time of the offence, how juries reach decisions, and how risk is assessed at sentencing and release.

Competency to Stand Trial

A defendant must be able to understand the proceedings and to assist in her own defence. In both Canadian and American law, a defendant found not competent cannot be tried; she is typically committed for restoration of competence, which in practice means treatment for the psychiatric or cognitive condition underlying the incompetence. Competency evaluations are among the most common forensic-psychology assessments. The standards — grounded in Dusky v. United States (1960) in American law, and in the Canadian Criminal Code’s fitness to stand trial provisions — assess factual understanding of the charges and process, rational understanding, and ability to consult with counsel.

The Insanity Defence

Criminal responsibility is distinct from competence. It concerns the defendant’s mental state at the time of the offence, not at the time of trial. The legal tests have varied historically.

The M’Naghten rule (1843, England) holds that a defendant is not responsible if, at the time of the offence, she was labouring under such a defect of reason from disease of the mind as not to know the nature and quality of the act, or, if she did know it, that she did not know she was doing what was wrong. M’Naghten remains the basis of insanity law in England, Canada (in modified form), and many American jurisdictions.

The Durham rule (1954, American) briefly held that a defendant was not responsible if the act was the product of mental disease or defect, but it was rejected as too vague and too expansive. The American Law Institute (ALI) test, adopted by many American jurisdictions following the 1962 Model Penal Code, combined cognitive and volitional prongs: the defendant is not responsible if, as a result of mental disease or defect, she lacked substantial capacity either to appreciate the criminality of the conduct or to conform to the requirements of law. Following the 1982 acquittal of John Hinckley for the attempted assassination of President Reagan, the volitional prong was narrowed or removed in many jurisdictions.

Canadian law uses the Not Criminally Responsible on account of Mental Disorder (NCRMD) standard, codified in Section 16 of the Criminal Code. A defendant found NCRMD is not acquitted in the usual sense; she is dealt with under a specialised review-board process that assesses ongoing risk and decides on hospitalisation, conditional discharge, or absolute discharge.

Chinese criminal law addresses mental-disorder-based reduction of responsibility in Article 18 of the Criminal Code, distinguishing full incapacity (no responsibility), diminished capacity (reduced responsibility), and full capacity. Forensic psychiatric assessment — 司法精神病鉴定 — is carried out by designated institutions, and the institutional arrangement differs from the adversarial-expert model familiar in Anglo-American practice.

Insanity acquittals are rare in all these systems. Public perception consistently overestimates both the frequency with which the defence is raised and the frequency with which it succeeds. In the United States, the defence is raised in well under 1% of felony cases and succeeds in a minority of those.

Jury Decision-Making

In jurisdictions with jury trials, the psychology of jury decision-making is a large research area. Findings include: juries are somewhat influenced by pre-trial publicity despite instructions to disregard it; individual jurors’ verdicts are shaped by first-ballot positions and group deliberation dynamics; certain kinds of evidence (confessions, eyewitness identifications) carry disproportionate weight relative to their actual reliability; and demographic characteristics of jurors, defendants, and victims interact in predictable ways with verdict outcomes.

The Chinese system does not use juries in the Anglo-American sense. Trials are conducted by professional judges, sometimes sitting with 人民陪审员 (people’s assessors) who play an advisory role. Research on Chinese judicial decision-making is a different enterprise, concerned with the psychology of professional adjudicators under institutional pressures, and it has produced its own findings on factors such as case load, precedent (in a system without binding common-law precedent), and institutional review.

Risk Assessment

Forensic practice is increasingly organised around structured risk assessment instruments that combine static (historical) and dynamic (changeable) risk factors to estimate the probability of future offending. The HCR-20 (Historical, Clinical, Risk-management) is a widely used structured professional judgement tool for violence risk. Static-99R and Stable-2007 are used for sexual-offence risk. The Level of Service/Case Management Inventory (LS/CMI), developed by Andrews and Bonta, is used for general criminogenic-risk assessment in Canadian corrections and elsewhere.

Structured instruments outperform unstructured clinical judgement in most controlled comparisons. They are not perfect: base rates of some serious outcomes (severe violence over short time horizons) are low, which limits predictive accuracy; instruments developed in one population generalise imperfectly; and the use of risk instruments raises serious ethical questions about group-level characteristics driving individual-level decisions. These debates are not closed.

Chapter 12: Punishment, Rehabilitation, and Reintegration

What to do with convicted offenders is a question with psychological, ethical, and political dimensions. The empirical psychology of punishment and rehabilitation is clear enough to rule out some positions, but the field has not converged on a single answer.

Deterrence Theory and Its Evidence

Deterrence theory, descended from Beccaria and Bentham, holds that punishment reduces crime by making the expected cost of offending exceed the expected benefit. Modern deterrence research distinguishes specific deterrence (effect on the individual punished) from general deterrence (effect on observers who learn of the punishment), and identifies three components of deterrent strength: certainty, severity, and celerity (speed) of punishment.

The evidence is substantially more supportive of certainty than of severity. Across multiple research traditions — including natural-experiment evidence, panel-data studies, and surveys of offender perception — the probability of being caught has much larger deterrent effects than the length of the sentence imposed when caught. The policy implication, which is politically uncomfortable in many jurisdictions, is that marginal increases in sentence length produce smaller deterrence gains than marginal improvements in detection and prosecution. The intuition that longer sentences deter proportionally more is not well-supported.

Specific deterrence through incarceration produces mixed results. Incarceration does not, on average, reduce subsequent offending among those incarcerated; for many, it is associated with neutral or slightly negative effects on desistance. Incapacitation — preventing offending during the period of incarceration — is real but subject to diminishing returns as the population of those most likely to offend is progressively incarcerated.

The RNR Model

The Risk–Need–Responsivity (RNR) framework, developed by Donald Andrews and James Bonta, is the dominant evidence-based framework for correctional programming. Its three principles:

Risk: match intensity of intervention to level of risk. High-risk offenders should receive more intensive intervention; low-risk offenders should receive less, because intensive intervention with low-risk offenders can actually worsen outcomes by exposing them to high-risk peers and disrupting protective ties.

Need: target criminogenic needs — the dynamic risk factors empirically linked to reoffending (antisocial cognition, antisocial peers, substance use, family/marital problems, school/work difficulties, absence of prosocial leisure). Non-criminogenic needs (such as self-esteem improvement, anxiety reduction, vague personal growth) have not been shown to reduce reoffending.

Responsivity: match intervention style to the offender’s learning style, motivation, and capacities. Cognitive-behavioural approaches have the strongest empirical support.

Meta-analyses consistently show that RNR-adherent programmes produce meaningful reductions in reoffending — typically in the range of 10–25% relative reductions — and that non-adherent programmes produce much smaller or no effects. RNR has been widely adopted in Canadian corrections and increasingly in other jurisdictions.

The Good Lives Model

Tony Ward and colleagues have argued, as a partial alternative and complement to RNR, for the Good Lives Model (GLM), which frames rehabilitation not only as risk reduction but as the positive construction of a life in which the offender can achieve normal human goods — autonomy, relatedness, competence, purpose — through non-criminal means. The GLM does not discard the empirical findings of RNR; it argues that motivation for change is better sustained when framed in terms of what offenders are building rather than only what they are avoiding. The evidence comparing GLM-informed programmes to RNR-only programmes is still developing, but the GLM has shifted clinical practice in many correctional settings toward a more growth-oriented framing without abandoning the evidence-based structure of RNR.

Restorative Justice

Restorative justice approaches — victim-offender mediation, family group conferencing, sentencing circles — aim to repair the harm of the offence through direct engagement between offender and victim, often with community participation. The evidence shows modest positive effects on victim satisfaction, modest positive effects on offender desistance for some offence types, and considerable variability in implementation quality. Restorative justice is not suited to all cases (serious violence where the victim does not consent; cases with significant power asymmetry) but has proven useful as one element of a broader response.

恢复性司法 has been a growing research and practice area in mainland China, with pilot programmes and academic engagement from multiple law schools. The Chinese application has drawn on both Anglophone restorative-justice sources and indigenous traditions of community mediation — the 人民调解 (people’s mediation) tradition has roots in pre-modern Chinese dispute resolution and was institutionalised in the socialist legal system — producing a distinctive synthesis.

Desistance Research

Sampson and Laub’s work on life-course criminology, mentioned earlier, has generated a specific research programme on desistance — how and why people stop offending. The findings, drawn from longitudinal studies and from qualitative work such as Shadd Maruna’s Making Good (2001), emphasise that desistance is typically not a single decision but a gradual process involving: accumulation of conventional social bonds (stable employment, stable relationship, parenthood); cognitive transformation in which the offender constructs a non-criminal identity narrative; reduced situational exposure to offending opportunities and antisocial peers; and often a period of “knifing off” of past associations.

Maruna’s interview work with desisting ex-offenders identified a recurring narrative structure he called the redemption script: a story in which the person retrospectively makes sense of her past offending and her present change, constructing an identity that includes the past without being defined by it. Maruna argued that correctional systems that frustrate this narrative work — by treating ex-offenders primarily as risks to be managed rather than as people in the process of change — inadvertently undermine the psychological scaffolding of desistance. Ward and Maruna’s Rehabilitation: Beyond the Risk Paradigm (2007) develops this argument systematically.

Chapter 13: Chinese Criminal Psychology as a Distinct Tradition

Chinese criminal psychology is not simply the Anglophone field practised in Chinese. It has its own institutional structure, its own intellectual lineages, and its own characteristic emphases. Understanding it as a distinct tradition, rather than as an imperfect reflection of the Western field, is necessary both to read its literature accurately and to understand why cross-cultural comparison is productive.

Institutional Geography

The institutional base of Chinese criminal psychology has three main pillars. The first is the 政法大学 (politics and law university) system, led by 中国政法大学 (China University of Political Science and Law), with substantial programmes at 西南政法大学, 华东政法大学, and 中南财经政法大学. These are integrated law-and-politics institutions in which criminal psychology is taught both in the undergraduate law curriculum and in dedicated graduate programmes, and where research connects closely to the courts, the procuracy, and legislative reform.

The second is the 公安大学 (public-security university) system, led by 中国人民公安大学 (People’s Public Security University of China). This system trains police officers and public-security officials, and its criminal-psychology research is strongly oriented toward investigative applications: profiling, interrogation, witness interviewing, hostage negotiation, and analysis of major-case behavioural evidence. The emphasis on 犯罪心理画像 (criminal psychological profiling) and 讯问心理学 (interrogation psychology) is particularly developed here.

The third is the general university psychology and criminology departments — in 中国人民大学, 北京大学, 北京师范大学, 华东师范大学, and others — where more experimentally oriented research on antisocial behaviour, juvenile development, and social cognition is conducted. Ministry of Justice and Ministry of Public Security research institutes also contribute.

Key Figures and Works

罗大华 (Luo Dahua) is the foundational figure of contemporary Chinese criminal psychology. His textbook 犯罪心理学, issued in multiple editions beginning in the 1980s, established the vocabulary and framework of the discipline. The textbook is organised integratively, covering individual psychological factors, family and social factors, developmental trajectories, specific offence types, and applications to investigation and prevention. It remains the standard reference at most Chinese law schools.

刘健清 (Liu Jianqing), also at 中国政法大学, produced a 33-lecture public-course series on 犯罪心理学 that has been widely circulated online and has served as a teaching template for subsequent courses. His presentation integrates theoretical exposition with concrete case material and emphasises the practical utility of the field for legal practitioners.

李玫瑾 (Li Meijin), of 中国人民公安大学, has become the most publicly visible figure in Chinese criminal psychology. Her work spans academic research on juvenile delinquency and family-development factors, applied research on criminal profiling, and extensive public lecturing — both within the public-security system and through mass media channels, particularly her 犯罪心理解析 public-lecture series. Her insistence on the importance of the first three to six years of life for later antisocial trajectories has been influential both in professional circles and in popular Chinese parenting discourse, though some of her specific claims have generated academic debate.

梅传强, of 西南政法大学, has produced textbook contributions that emphasise the connection between criminal psychology and criminal law doctrine. 林秉贤 has written on offender classification and correctional psychology. 张保平, at 中国人民公安大学, has worked on interrogation psychology and case-consultation methodology.

Characteristic Emphases

Reading across this literature, several characteristic emphases distinguish Chinese criminal psychology from the Anglophone mainstream.

First, a heavier emphasis on family-level developmental factors and on moral formation in early childhood. This is partly a reflection of Chinese cultural assumptions about family responsibility for child development, partly a reflection of the institutional absence of the large out-of-home care systems that loom in Western child-welfare and juvenile-justice research, and partly an empirical judgement about what the evidence supports. 李玫瑾’s work is the most visible expression, but the emphasis runs through the mainstream textbooks.

Second, a closer integration of research with applied investigative practice. The Anglophone field, particularly as practised in American academic psychology, has kept a certain distance from police work; the Chinese field, housed substantially in public-security universities, has not. The practical orientation produces both strengths (direct engagement with operational problems) and limitations (fewer institutional incentives for purely theoretical or critical work).

Third, an institutional framing that is state-organised rather than adversarial. Chinese criminal procedure does not feature adversarial expert testimony in the same way American or Canadian trials do, and criminal-psychology expertise enters the system differently: through investigative consultation, through state-conducted forensic-psychiatric examination, and through judicial reference rather than through defence-versus-prosecution expert battles. This shapes what research questions are most salient and what standards of justification operate.

Fourth, a more prominent role for 恢复性司法 (restorative justice) informed by indigenous mediation traditions alongside imported restorative-justice literature. The continuity between pre-modern Chinese community-dispute practices, socialist 人民调解 institutions, and contemporary restorative-justice programmes produces a distinctive Chinese articulation of the restorative framework.

李玫瑾 and Popular Criminal Psychology

One feature of the Chinese landscape worth noting is the prominent popular presence of criminal psychology — more prominent, arguably, than in most Anglophone settings. 李玫瑾’s public lectures, widely disseminated through online video and television, have introduced a Chinese mass audience to concepts of developmental criminology, family attachment, and profiling in a way that has no direct parallel in most Western countries (where popular criminology is dominated by true-crime entertainment rather than by academic-popular synthesis). This has had mixed effects: it has raised public understanding of some findings but has also exposed the field to debates over the accuracy of particular popular claims, the ethics of commenting publicly on ongoing cases, and the boundary between scholarly and entertainment roles. The phenomenon is itself an object of interest for the cultural study of how criminal-psychology knowledge circulates.

Chapter 14: Cross-Cultural Comparison

The comparison between Anglophone (particularly American, Canadian, and British) and mainland Chinese criminal psychology is analytically productive because it holds the underlying psychological processes approximately constant while varying the institutional, legal, and cultural conditions under which those processes operate. The differences that remain are informative about what is machinery and what is incentive.

Rates and Forms of Violence

As noted in Chapter 5, homicide rates differ substantially: roughly 5–7 per 100,000 in the United States, 1.7–2 in Canada, and 0.5–1 in mainland China. Firearm availability is a large part of the American exceptionalism. The Chinese low rate reflects both strict firearm controls and a more developed apparatus of state control, surveillance, and informal social monitoring; it is also consistent with a long historical trend of relatively low lethal-violence rates in Han Chinese society, a trend that predates the current political system. Canadian rates fall between the two, with firearm regulation stricter than the US but looser than China, and with cultural patterns that have historically produced lower interpersonal-violence rates than the US.

The forms of violence differ correspondingly. US juvenile violence disproportionately involves handguns; Chinese juvenile violence disproportionately involves knives, blunt instruments, or unarmed assault. US prison populations are disproportionately large relative to population; Chinese prison populations are substantial but proportionally smaller. US criminal-justice involvement is racially stratified in ways specific to the US history of race; Chinese criminal-justice involvement has its own stratifications — by rural/urban origin, by 户籍 (household registration) status, by migrant-worker status — that do not map onto the American pattern.

Informal Social Control

Both the Chinese and the Western traditions recognise that most social control is not formal legal sanction but informal reputational, familial, and community pressure. They differ in the density and structure of that informal control.

Chinese society, particularly in the rural and small-city settings that remain important even in a heavily urbanised country, continues to feature relatively dense family and lineage networks that monitor and constrain behaviour. The concept of 面子 (face) — the public standing that one maintains within one’s social world, and that one can gain or lose through behaviour — organises a substantial amount of social regulation that in a Western setting might require formal institutions. A young man whose offending would embarrass his parents and damage the family’s standing in the neighbourhood faces a form of restraint that is psychologically real and operationally effective.

Western societies retain informal social control, but its density has generally declined with urbanisation, geographic mobility, and the individualisation of life trajectories. The result is a heavier reliance on formal legal and institutional sanction to produce the social order that informal control produces less reliably. Anglophone criminologists — Sampson on “collective efficacy,” Putnam on “social capital” — have been concerned precisely with the consequences of this erosion for crime rates and community well-being.

The Psychology of Face and Responsibility

The cultural framing of individual responsibility for offending differs in ways that matter for both the psychology of the offender and the psychology of the response. A familiar Anglophone assumption is that an adult offender is primarily individually responsible for her acts; family and community may share in the grief and the social consequences, but the locus of responsibility is the individual. A Chinese cultural framing — not universal but common — extends responsibility more diffusely across the family and sometimes the community. The offender’s parents may experience the offending as a failure of upbringing; the offender may experience the offending as a betrayal of the family; and the community may treat both the offender and the family as implicated.

This has consequences for several processes. It shapes the motivation to offend: a Chinese young man considering a shameful act may face a stronger anticipatory inhibition from the family-level cost than an Anglophone young man considering the same act. It shapes the response to offending: family-level shame may generate a stronger effort to achieve desistance and reintegration but may also generate pressure to conceal. It shapes rehabilitation: reintegration into family and community can provide a powerful scaffold for desistance, but the same shame that motivated initial inhibition can inhibit the honest acknowledgement on which desistance partly depends.

None of this is to essentialise Chinese or Anglophone psychology. Both populations contain enormous variation; both cultures have shifted substantially over the past half-century; and the generalisations should be read as modal tendencies rather than universal traits. But the tendencies are real, and they matter for any serious cross-cultural engagement with criminal-psychology material.

Firearms and the Ecology of Lethal Outcomes

The firearms question deserves one further comment because it illustrates how institutional choices shape the phenomena that criminal psychology studies. In the United States, civilian firearm ownership is approximately one firearm per resident. In Canada, it is substantially lower and more tightly regulated, though still significant. In mainland China, civilian firearm ownership is essentially zero for ordinary residents, enforced through stringent controls.

The psychological dispositions that produce interpersonal violence — status competition, jealousy, anger, impulsive aggression, alcohol-fuelled escalation — exist in all three populations. What differs is what happens when those dispositions erupt. An angry encounter between two young men in an American urban neighbourhood can terminate in a fatal shooting; the same encounter in a Canadian city is more likely to terminate in a non-lethal assault or nothing at all; the same encounter in a Chinese city is even less likely to involve a lethal outcome. The criminal psychology of the participants is, on average, quite similar. The ecology of consequences differs.

This is an uncomfortable finding for theories that locate the explanation of homicide rates primarily in the psychology or culture of the offending population. The explanation is substantially institutional. Criminal psychology, as an empirical field, cannot ignore this; any theory of lethal violence that does not account for the means-access factor is incomplete.

The Meaning of Prevention

The Anglophone tradition of crime prevention — situational crime prevention, environmental design, developmental prevention, early-intervention programmes — is well developed and empirically grounded. The Chinese tradition has its own extensive literature on 犯罪预防 (crime prevention), often framed in terms of 综合治理 (comprehensive governance), that integrates police work, community organisations, schools, families, and social services into a denser intervention network than the typical Anglophone framework.

Both traditions recognise that prevention is more effective than punishment after the fact, and both are informed by the same developmental-criminology findings about the risk factors for persistent offending. They differ in the density of the intervention network they presuppose, in the role of the state, and in the balance between informal and formal social control. The Chinese model achieves some of its prevention outcomes through institutional density that would be difficult to replicate in a looser institutional context; the Anglophone model has developed evidence-based practices that are portable across contexts in ways the dense institutional model is not.

Chapter 15: Closing Reflections

Criminal psychology, at its best, is a field of careful empirical work on phenomena that are emotionally charged, politically loaded, and resistant to simple analysis. It sits between the clinic and the courtroom, between individual pathology and institutional structure, between the psychology of one person and the psychology of a whole system for handling that person. A student finishing this course should be able to do several things.

First, she should be able to think about a criminal event at four levels simultaneously — biological, individual-psychological, social, and institutional — without collapsing any level into the others. The collapse is the characteristic error of public discourse on crime, and resisting it is one of the main intellectual contributions of the field.

Second, she should be appropriately sceptical of confident claims about offenders. Profilers who claim near-certain insight from crime scenes, commentators who claim to know from a news report whether a defendant is lying, politicians who claim that longer sentences will solve crime, advocates who claim that any specific intervention will transform criminal-justice outcomes — all of these should be read against the empirical literature, which tends to be more modest. The honest positions in criminal psychology are often the boring ones.

Third, she should hold the Anglophone and Chinese traditions of the field in mind together. The Anglophone tradition, despite its greater English-language visibility, is not the complete field. The Chinese tradition is not a minor variant. Each has made contributions the other has not, and each illuminates assumptions of the other. Students who read only one tradition absorb its parochialism unconsciously; students who read both are positioned to see where claims are about human beings and where they are about local institutional arrangements.

Fourth, she should be aware of the ethical weight of the field. Criminal psychology produces claims that are used in decisions about liberty, about blame, about who is treated and who is punished, about who is believed and who is disbelieved. The claims are never purely descriptive; they always have moral and political consequences. A criminal psychologist who ignores this is not doing neutral science; she is unconsciously participating in the politics of the institution that uses her work. The honest alternative is to do the empirical work as carefully as possible and to be transparent about its uses, its uncertainties, and its limits.

The field is not going to produce a final theory of crime. The phenomena are too heterogeneous and the conditions under which they unfold too variable. What the field can produce, and what this course has tried to introduce, is a set of analytic habits and a catalogue of robust findings that make the reader’s engagement with criminal-justice questions less confused, less credulous, and less vulnerable to the reassuring simplifications that public discourse about crime tends to offer.