Serial Killers
in a Hurry

The Whole Thing in One Page

The serial killer in popular culture is a species apart: a white male genius with no conscience, a private ritual and a clue left for a detective clever enough to understand him. He hunts strangers, grows bolder, follows a signature and eventually wants to be caught. The picture is coherent, frightening and built from cases that made the best entertainment.

Serial murder is less tidy. Even the definition moves. One FBI formulation covers two or more victims killed unlawfully by the same offender or offenders in separate events. Older rules demanded three deaths and a cooling-off period, but neither threshold reveals a natural kind. Separation defines the category. Sequence explains it.

After a first killing, four conditions must align for another to follow. The offender must choose to kill again. Another reachable person must enter the offender's access. Nothing must interrupt the pathway first. The separate events must remain unrecognised long enough for the next attempt. These are gates, not stages of destiny. A witness, survivor, prison sentence, lost job, alert relative or linked record can close the route.

The mind matters at the first gate, but there is no single mind to find. Some offenders seek sexual domination or sadistic control. Others kill for money, grievance, status, ideology or concealment. Psychopathy can help explain callousness and manipulation in some cases, yet it is neither necessary nor sufficient. Psychosis is relevant to some individual cases, but evidence from narrower sexual-homicide samples cannot supply one prevalence rate for serial murder. Childhood cruelty, bed-wetting and fire-setting do not form a validated warning code. A risk factor is not a forecast.

Access turns willingness into possible repetition. A doctor has different access from a stranger; a carer from a mobile offender; someone targeting those at the edge of official attention from one attacking relatives. Serial murder therefore belongs partly to institutions. Death certification, missing-person systems, police boundaries, evidence preservation, data sharing and assumptions about whose disappearance is urgent can lengthen or shorten a sequence. None causes the offender's choice. Each can affect whether another choice remains possible.

The numbers record this process imperfectly. There is no census of serial killers. Counts are assembled from convictions, confessions, linked cases, unresolved suspicions and researchers' rules. They overrepresent offenders who were caught and victims who were identified. They miss deaths classified as accidents, overdoses, natural causes or unrelated murders. A 2008 FBI practitioner report put the category below one per cent of murders for a given year, an order-of-magnitude anchor rather than a current headcount.

Known United States series appear to have fallen markedly from their late twentieth-century peak. Better linkage, DNA, cameras, phones, altered routines, incarceration and changing homicide patterns may all have contributed, but no study has separated their shares. Recent decades also have less time to be solved and linked. The direction is credible. The exact curve and causes remain less secure.

The deepest myth is that the subject is chiefly about extraordinary offenders. It is about repeated choices passing through ordinary openings: access, weak guardianship, missed signals and records not yet joined. The offender supplies the violence. Circumstance supplies reach. Systems influence interruption and recognition. The final number is what remains after all four gates have done their work.

That is the book.

Why You Should Care

In October 2019 the FBI said that Samuel Little had confessed to ninety-three murders and that its analysts considered every confession credible. At that stage, law enforcement had verified fifty. The number was shocking. The route by which so many deaths stayed separate was worse. Some had been classified as overdoses, accidents or deaths of undetermined cause. Some bodies had never been found. Little told investigators that he had believed nobody was accounting for his victims.

That sentence explains more than another tour of his childhood or fantasies would. A person willing to kill was one condition. Women whose lives and deaths were easy for institutions to discount were another. Separate jurisdictions, incomplete records and plausible alternative explanations supplied the rest. The series was hidden in plain sight because it did not look like one series from any single desk.

You should care first because the public model gets risk badly wrong. The scale estimate used here comes from American FBI practitioners, who put the category below one per cent of murders in a year. Homicide more often arises from relationships, conflicts and circumstances unlike the stranger-genius story. A frightening exception has been promoted into the representative face of murder. That distortion can waste attention while making ordinary forms of violence look ordinary enough to ignore.

Second, this is a severe test of how you reason from memorable cases. The sample available to you is selected twice: offenders must be detected, then their stories must be retold. The most unusual, theatrical and prolific cases receive the widest circulation. From that collection, audiences infer the typical offender. It is the statistical equivalent of studying aviation through famous crashes and then describing a normal flight from the wreckage.

Third, the psychology matters beyond crime. Words such as psychopath, sadist, narcissist and mentally ill are used as if they name one descending scale of evil. They do not. Traits, diagnoses, motives and acts are different things. Confusing them encourages amateur diagnosis, stigma and false confidence about prediction. Understanding a risk factor is not the same as recognising a future murderer.

Fourth, the investigative lesson is transferable. Rare events are often distributed across institutions that each see too little. One hospital notices a death. One police force records a missing person. Another finds an unidentified body. A database can connect what no individual remembers, but only if the events are entered, described compatibly and treated as worth comparing. Information systems do not merely store reality. They help determine which pattern becomes visible to investigators.

There is an ethical reason as well. Serial-killer culture routinely grants the offender a full identity and compresses victims into a count. The killer gets a nickname, psychology and preferred brand of spectacles. The dead become numbers proving his importance. A truthful account must reverse that allocation of attention. Victim selection is central evidence, but a victim is not a type chosen to complete somebody else's profile.

Finally, serial murder shows why prevention can make a problem harder to measure. Better linkage may stop a willing offender after one known killing, so the person never enters a serial-murder database. Faster detection can prevent later offences when it enables arrest or another effective interruption, while removing evidence of what might otherwise have happened. Weak systems can permit a longer, clearer series. Stronger systems may produce interrupted paths that never qualify for the category. Counting outcomes alone can therefore make failure easier to see than successful prevention.

The subject becomes more disturbing when the mythology is removed. There is no hidden race of super-predators waiting to be decoded by intuition. There are rare offenders, varied motives, available victims and institutions with uneven powers of recognition. The useful question is not whether you could spot the monster in a room. It is what would allow separate harms to continue without anyone seeing that they belong together.

The Core Ideas

The Series Is a Sequence

A mass killing announces itself. Several people die in one event, one place or one connected burst, and nobody needs a theory to decide that the deaths belong together. Serial murder begins with the opposite problem. One person is killed. The event appears to end. Somewhere else, or much later, another person dies. The common offender may exist in the world before the link exists in any record.

This is why definitions matter. Older accounts often required at least three victims and a cooling-off period between killings. The three-victim rule gave agencies and researchers a threshold, but nature did not supply it. An offender arrested after two separate murders is not psychologically transformed by the missing third body. Cooling-off is weaker still. The interval may contain work, family life, imprisonment, illness, loss of access, hesitation, planning or no identifiable emotional change. Time separates the events. It does not reveal what the offender felt.

A symposium hosted by the FBI in 2005 proposed a practical definition later published in its monograph: the unlawful killing of two or more victims by the same offender or offenders in separate events. Specialists still debate motive, reliable linkage, accomplices and where rapid sequences belong. The formulation remains useful because it identifies the operational problem without pretending to settle the offender's inner state.

The boundary with mass and rapid-sequence killing concerns the structure of events, not a ranking of horror. Multiple deaths in one attack create an immediate collective scene. Killings across separate episodes require a connection to be discovered. The older term spree murder tried to name a middle category with little meaningful interval, but its reliance on inferred cooling produced unstable borders. Labels organise cases. They should not smuggle in claims about sanity, motive or danger that timing alone cannot support.

Once one killing has occurred, a series requires four gates to remain open. The offender must decide or remain willing to kill again. A new potential victim must come within reach. The route must survive interruption by witnesses, resistance, arrest, imprisonment, illness, changed work or lost opportunity. The separate events must remain unrecognised long enough for another attempt. The gates can open in different orders and close for reasons nobody later knows. They describe conditions, not a script every offender follows.

This model restores agency without making personality the entire explanation. The offender chooses the act. Access and circumstance shape which acts are feasible. Guardians, institutions and chance affect whether the offender remains able to repeat. An intended series stopped after one murder cannot be observed as a serial-murder case. A long series is therefore a selected outcome among many possible pathways, not proof that the offender possessed an exceptional essence from the beginning.

The interval permits adaptation on both sides. An offender may learn which account was accepted, where control failed or which victims drew little attention. Investigators may compare injuries, geography, contact patterns, DNA, missing-person reports and survivor accounts, provided somebody sees reason to compare them. Sometimes the offender adapts. Sometimes investigators do. Often chance intervenes before either process deserves a grand theory.

This reframes apparent skill. A long sequence can include planning and deception, but it can also record ordinary luck, reachable victims, evidence crossing police boundaries, deaths assigned plausible causes or institutions that did not share what they knew. Some offenders make repeated errors against systems that are not yet looking for one person.

The public story starts with a monster who cannot stop. The better model starts with a repeated choice and asks why the route to another victim remained open.

Minds, Not a Mind

Ask why serial killers kill and the tempting answer is to find the common wound, urge or defect. The problem is that the category is defined by repeated behaviour, not by one motive. People can arrive at the same outward pattern by different routes.

Some cases involve sexual arousal linked to domination, humiliation or suffering. Some centre on control without an obvious sexual motive. Others involve profit, inheritance, insurance, theft or relief from a burden. A caregiver may kill people already expected to die. A partner may take part through loyalty, dependency or shared hostility. An offender may target members of a group framed as contaminating or deserving. Another may kill witnesses, unwanted dependants or people who obstruct fraud. The word serial tells you that the killings occurred in separate events. It does not tell you what they meant to the offender.

Sex, partnership and setting alter what becomes visible. In 2017 the Canadian nurse Elizabeth Wettlaufer pleaded guilty to eight murders, four attempted murders and two aggravated assaults involving people receiving long-term or home care. Ontario's public inquiry later concluded that the offences would not have been discovered without her confession and identified systemic vulnerabilities rather than one failed gatekeeper. Her case does not establish a female type. It shows why a model built around the lone male sexual predator can miss medical, financial, domestic and team offending.

The contrast between Harold Shipman and Samuel Little makes the point without turning either man into a type. Shipman was a trusted British general practitioner whose position gave him private access to patients and authority over the records surrounding their deaths. The first report of the public inquiry concluded that he had killed at least 215 patients. Little moved through a different social world and targeted women whose deaths and disappearances were often weakly investigated. Both produced long series. The common feature was repeated killing plus delayed recognition, not a shared profession, victim relationship or public persona.

Even apparently similar offenders differ. Two men may both kill strangers and seek control, while one plans far ahead and the other acts opportunistically. One may feel intense fantasy and another describe little. One may preserve ordinary relationships through compartmentalisation; another lives chaotically. Interviews after arrest add another difficulty. Offenders lie, boast, minimise, absorb media accounts and learn the language investigators expect. A confession can establish an act without opening a transparent window into its meaning.

Mental illness requires the same separation. Psychosis is documented in some sexual-homicide cases, but reviews of that narrower literature find it uncommon in most samples and do not establish a prevalence rate for serial murder as a whole. Delusions can shape perceived threat or mission in a particular case. Many offenders conceal conduct, choose circumstances and change behaviour in response to risk, showing practical contact with reality. Calling every repeated murderer mad may express moral disbelief, but it is not a clinical account.

Childhood adversity is common among many offending populations and can affect attachment, emotion regulation, substance use, aggression and life chances. It is neither present in every serial murderer nor rare among people who never become violent. Retrospective biographies invite a search for the childhood event that made the ending inevitable. Millions of people survive abuse without harming anyone. A risk factor changes probability; it does not write a verdict.

The strongest account is plural and layered. Biology can affect temperament. Development can shape coping and attachment. Social learning can supply scripts. Fantasy can organise desire. Opportunity can make action feasible. None removes the offender's agency. Choice occurs within constraints, and no one combination of causes is mandatory.

There is no serial-killer mind. There are minds that reached repeated murder by different paths, then records that made the paths look more alike than they were.

Psychopathy, Sadism and Fantasy

Psychopathy is the word most often used to make the subject feel solved. It is a research and forensic construct describing a constellation of interpersonal, emotional and behavioural traits that can include manipulation, callousness, shallow affect, irresponsibility and persistent antisocial conduct. Structured measures require trained assessment and collateral evidence. Psychopathy is not a diagnosis listed under that name in the DSM-5-TR, a synonym for evil, or a medical scan that reveals a killer.

The distinction from antisocial personality disorder matters. Antisocial personality disorder is a formal clinical diagnosis centred heavily on a persistent pattern of violating rights and social norms. Psychopathy is a narrower construct with greater emphasis on affective and interpersonal features as well as behaviour. The categories overlap, but they are not interchangeable. Neither identifies serial murder. Many people with antisocial traits do not kill. A high score on a measure of psychopathic traits does not make violence inevitable. Some serial murderers do not fit the classic psychopathic picture.

Biological research does not rescue prediction. Group studies can report average associations between biological measures and aggression, impulse control or callous traits in selected populations. They do not provide a validated test that identifies a future serial murderer in one person. Biological contribution is compatible with responsibility and uncertainty. It is not a hidden mugshot.

Where psychopathy is present, it may explain part of the mechanism: reduced guilt, ease of deception, instrumental use of other people and the ability to return to ordinary life after harm. It does not supply the target, opportunity, motive or repeated sequence. A trait profile can help explain how somebody acts without explaining why these victims were killed at these times.

Sadism is another term that expands until it means enjoyment of cruelty. In clinical use, sexual sadism concerns recurrent sexual arousal from another person's suffering. A disorder requires clinically significant distress or impairment, or acting on the urges with a non-consenting person. Some sexually motivated serial murderers show sadistic interests and behaviour. Others do not. Serial murder also includes cases driven by profit, grievance, concealment, caregiving roles and other motives for which sexual sadism offers little.

Fantasy can matter without operating like a command. Most people have aggressive, sexual or transgressive thoughts they never act upon. In some offenders, recurrent fantasies become more elaborate, rehearsed or tied to arousal and control. They may guide victim choice or the imagined sequence of an offence. Yet the existence of fantasy cannot explain the crossing from thought to action. Opportunity, inhibition, moral judgement, fear of consequences, relationships and circumstance stand between them. A theory that begins with fantasy and ends with murder has skipped the decisive step.

Popular accounts often fill that step with escalation. The child harms an animal, becomes a voyeur, attacks an adult and then kills. Real lives are less obedient. Behaviour can persist, stop, change direction or appear after the act it is supposed to predict. The Macdonald triad, animal cruelty, fire-setting and persistent bed-wetting, survives because it gives childhood a memorable code. A critical review found that the three behaviours rarely appear together and that the evidence does not support the package as a predictor of later violence. Each behaviour may warrant care or safeguarding on its own terms. None is a serial-killer test.

Prediction fails partly because the outcome is so rare. Even a marker strongly associated with violence would produce many false alarms if applied to millions of children, because almost none will become serial murderers. Add loose definitions and retrospective reporting, and the apparent warning sign grows sharper only after the ending is known.

The mind should be examined without pretending it contains a barcode. Psychopathy can lower one barrier. Sadism can supply one reward. Fantasy can organise one script. None creates a series alone.

Opportunity Makes Repetition

A desire can remain a desire for life. Repeated murder requires a route from intention to another human being, enough privacy or authority to act, and enough failure around the event for the offender to retain that route.

Opportunity is often mistaken for convenience. It is broader. It includes employment, housing, transport, social status, trust, routines, knowledge of institutions and access to people whose circumstances reduce guardianship. A doctor meets patients in private and is expected to explain death. A carer controls medication and records. A landlord knows who lives alone. A mobile offender encounters people beyond one police district. An offender inside a family can exploit the assumption that repeated illness or loss is private misfortune. The role does not cause the crime. It shapes what can be done, to whom and under what cover.

Place changes the calculation. Dense cities create witnesses and cameras, but also anonymity and jurisdictional seams. Roads widen the pool of contact while scattering scenes. Institutions concentrate vulnerable people and can concentrate oversight, though oversight may be deferential to professional authority. The same environment can raise opportunity and detection at once. What matters is the particular meeting of target, access, privacy and guardianship, not a simple claim that one place breeds serial murder.

Shipman's case shows the difference between motive and series formation. Whatever internal reward he obtained, his medical position supplied repeated contact with older patients, clinical authority and a death-certification system that gave too much weight to the treating doctor's account. The public inquiry did not discover a special medical psychology. It exposed a structure in which one trusted professional could influence both the event and its official explanation. From 9 September 2024, every death in England and Wales must receive independent medical-examiner scrutiny or a coroner's investigation. Government guidance cites the Shipman Inquiry among several reports and inquiries that underlined the reform. His case helped change the architecture around ordinary deaths without being its sole cause.

Victim availability is part of opportunity, but the phrase can become an accusation if handled badly. A person is not responsible for being accessible. The relevant question is how daily life distributes exposure and protection. Street-based work, unstable housing, migration status, addiction, institutional care, estrangement and poverty can increase contact with predatory people while reducing the speed or confidence of official response. An offender can notice those asymmetries without possessing deep psychological insight. Society has already marked where attention is thin.

Opportunity also explains why demographic patterns vary across settings. Women identified as serial murderers have often been found in caregiving, family or financial contexts in which repeated deaths could be mistaken for illness or ordinary loss. Men dominate many datasets, especially those built around sexual stranger murder, but the sample partly reflects which forms investigators and researchers were trained to recognise. Change the definition, period, country or source, and the portrait changes.

This does not reduce agency. Opportunity theories sometimes sound as though environments commit crimes. They do not. Countless doctors, carers, drivers and landlords possess access and use it to help people. The offender supplies willingness and choice. The environment affects how often that choice can be exercised before resistance arrives.

Prevention therefore has more levers than personality prediction. Independent review, continuity of records, rapid missing-person assessment, cross-boundary comparison, safe reporting routes and attention to unusual clusters can remove opportunity or shorten it. None requires identifying a future serial killer from a face or childhood history. They alter the conditions under which any offender could repeat.

The mind may explain why somebody wants control. Opportunity explains why the want becomes a series rather than a private fantasy, one offence or a failed attempt.

Behaviour Is Evidence, Not Identity

Crime fiction offers a clean reversal. The offender leaves behaviour at a scene. The profiler reads it and reconstructs the person: intelligence, work, relationships, habits, perhaps the street where he lives. Real behaviour can be informative. It rarely carries that much identity.

Modus operandi means the methods used to commit an offence and manage practical problems. Those methods are learned and adaptive. An offender may change approach after resistance, witnesses, media coverage, injury, loss of access or a near arrest. A method that worked once may fail in a new setting. Consistency exists, but it competes with necessity, experience and chance.

Signature is meant to describe behaviour that expresses psychological needs beyond what was required to complete the crime. The concept can help investigators ask whether uncommon acts connect offences. It becomes dangerous when treated as a fixed personal seal. Behaviour depends on time, opportunity, victim response, intoxication, interruption and what evidence survives. Different offenders can perform similar acts; the same offender can omit or alter them. Meaning is inferred, not stamped into the scene.

The organised-disorganised typology promised more. Order, planning and concealment supposedly indicated a socially competent offender; disorder and impulsivity indicated an impaired one. David Canter and colleagues tested thirty-nine features from one hundred United States serial murders and found no distinct pair of types. Organised features were common, disorganised features rarer, and mixed scenes ordinary. A scene can become chaotic because a victim resists. It can appear orderly because the offence itself required preparation. Neither observation yields a full biography.

Two tasks are often confused. Case linkage asks whether separate crimes may share an offender. Offender profiling asks what characteristics that unknown person may possess. Behaviour can be more useful for the first task than the second. A rare combination of conduct, geography, contact route and physical evidence may justify comparing cases without supporting a confident claim about age, job or personality. Even linkage remains probabilistic: common methods can recur independently, and one offender can change.

A defensible analysis therefore works through alternatives. Similar cases may share an offender, a victim environment, an available weapon, copied publicity or coincidence. Dissimilar cases may connect because access changed or the offender adapted. Confidence should rise when several independent features converge and when the proposed link explains differences rather than hiding them. It should fall when every contradiction becomes proof that the offender is unusually clever. A theory able to absorb any result cannot be tested.

Databases support this work only when cases enter them. The FBI established the Violent Criminal Apprehension Program in 1985 to collect and compare information about serious violent crimes across jurisdictions. DNA later allowed biological material from separate scenes to be associated, and renewed cold-case methods sometimes restored links years afterwards. This book leaves laboratory practice to Forensics in a Hurry. The structural point is enough: records and traces can remember across time when individual investigators cannot.

Behavioural advice is most useful when it produces ranked questions, search priorities or testable hypotheses. It does not identify a suspect, establish guilt or replace physical, documentary and testimonial evidence. A pattern can guide attention. It cannot sign its own name.

The Victims the Story Deletes

A serial-murder story usually introduces the killer first and numbers the victims afterwards. This is narratively efficient and intellectually corrupting. It makes the offender the source of unity and turns separate lives into evidence of his scale.

Victimology is the study of a victim's life, routines, relationships and circumstances as they bear on an investigation. Used well, it can reveal contact routes, identify suspects, distinguish risk from coincidence and connect cases. Used badly, it becomes a posthumous audit of whether somebody lived respectably enough to deserve urgency. The same facts can explain exposure without assigning blame.

Offenders do not select from humanity at random. They select from people they can approach, control, deceive or make disappear with the least resistance. That produces patterned victimisation. People engaged in street-based sex work, living without secure housing, using drugs, estranged from family, moving across borders or subject to discrimination can face high exposure and low institutional protection at once. Elderly or disabled people in care may be vulnerable for the opposite reason: their deaths are expected and authority is concentrated around those providing help.

The phrase “the missing missing” describes people whose absence never becomes a usable missing-person record. Some are not reported. Some reports receive low priority. Some unidentified remains are not compared with reports elsewhere. Some deaths are assigned a cause that ends inquiry. Kenna Quinet's work shows why any estimate of serial-murder victimisation must consider these omissions, even though attempts to turn them into a grand total rest on assumptions too fragile for confidence.

Samuel Little's case gives the abstraction names, though the FBI's 2019 account still had many unidentified. His victims were often women living precariously. Investigators said that numerous deaths had first been treated as overdoses, accidents or undetermined causes. The failure was not one dramatic missed clue. It was the cumulative effect of lives receiving less documentation, deaths receiving plausible low-attention explanations and cases remaining local.

Media selection deepens the distortion. Victims who fit familiar ideals of innocence receive more sustained coverage. Offenders who provide photographs, interviews, courtroom spectacle or a nickname become cultural objects. The unequal exchange is obvious in bookshelves: one man's face repeated across covers, many victims represented by a number beside it. Publicity can help identify victims or generate leads, but fame also rewards the offender's claim to authorship.

Families and survivors carry a different record. They know the missed appointments, unanswered calls, interrupted plans and years spent pressing institutions to compare what looked isolated. Their knowledge can correct official assumptions, but it is unevenly preserved and often arrives in public only after a suspect becomes famous. The clean final victim count hides disputed identifications, people still unnamed and relatives who never received a settled answer.

Restoring victims does not mean writing sentimental mini-biographies that interrupt analysis. It means refusing to let the killer's categories become ours. “Prostitute”, “runaway”, “addict”, “patient” and “drifter” describe a circumstance or role, not the worth of the person. The investigative question is what that circumstance changed about contact, reporting and recognition. The moral fact is that nobody becomes less murdered because a system expected trouble around them.

A series is connected by one offender's acts. It can be lengthened when institutions repeatedly fail to treat each victim as a person whose disappearance or death requires an answer.

The Numbers Record Offending and Discovery

There is no turnstile counting serial killers as they enter the category. Every dataset is assembled backwards from cases that became knowable.

The first measurement choice is the definition. Researchers may require two victims or three, include or exclude accomplices, count attempts, separate rapid sequences, exclude political or contract killing, and demand different levels of case linkage. Change those rules and the total changes before any crime has changed. A number without its entrance policy is not a population fact.

Victim counts have their own ladder of certainty. An offender may be convicted of some murders, confess to others, be linked by investigators to further cases and remain suspected in still more. Databases use confirmed, linked, admitted, probable and suspected differently. Confessions can be truthful, mistaken, strategic or inflated. The largest number attached to a famous offender is often the least secure and the most repeated.

The denominator is no firmer. In 2008, FBI practitioners placed serial murder below one per cent of murders in a given year. That is a useful correction to its cultural prominence, not a precise annual rate for every country or later period. Unknown series, disputed links and different recording systems prevent a global headcount.

Samples can look authoritative while answering narrower questions. The FBI's 2014 Pathways for Investigations study used 480 cases attributed to 92 offenders. They were United States cases from 1960 to 2006, submitted or solicited through a specialist federal centre, and the design included single male offenders acting alone. The authors stated that the sample was not representative of the United States. It can illuminate patterns inside comparable cases. It cannot establish the universal sex, race, motive or behaviour of serial killers.

Demographic claims inherit the same selection. The claim that all serial killers are white men is false. Documented cases include women, teams and offenders from many racial groups. Yet replacing the myth with one confident pie chart repeats the mistake. Race categories vary across countries and time. Police attention, victim groups, clearance, definitions, historical archives and which offences researchers seek all shape who enters the table. Identified offenders are not a random sample of hidden offenders.

Trend claims add a time problem. A 2019 peer-reviewed study found a marked fall in known United States series from the 1980s peak into the 2010s. Its result supports a directional decline, but the study used assembled case records rather than a national census and could not assign causal shares. DNA, databases, digital traces, cameras, altered mobility, incarceration, changing victim environments, lower homicide and public vigilance are plausible contributors. They remain a set of hypotheses, not a solved decomposition.

Recent decades are also right-censored. A series beginning in the 1980s has had decades to be solved, confessed or reclassified. One beginning in 2025 has not. Genetic genealogy and renewed cold-case work can move victims into older decades long after the murders. A graph by decade therefore compares mature history with an unfinished present.

Return to the four gates. Detection affects the final gate directly by connecting separate events, and it can close the pathway before the next killing. A person arrested after one offence may never become observable as a serial murderer, even if another attack was intended. Better prevention can reduce the category while removing evidence of the cases it prevented. Poor linkage can permit more victims and leave a clearer historical series. Good linkage may stop the path earlier and leave a harder counterfactual.

The number on the page is therefore a residue of offending, access, interruption, recognition and time. It measures crime imperfectly and the machinery used to discover crime at the same time.

How It Actually Works

The first death

A body is found, or a person disappears, and no official file yet contains a serial-murder case. There is one event and several possible explanations. The death may be recognised as homicide. It may resemble overdose, accident, suicide, illness or exposure. There may be no body, no prompt report and no agreement about where the missing person was last seen. This is the ordinary logic of a first case, not a reconstruction of one particular death.

The first decisions are classificatory. Is the person missing by choice or at risk? Is the death suspicious? Which police force owns the case? What evidence deserves preservation? Whose account is treated as reliable? Each decision can be reasonable in isolation and still delay the recognition of repetition. Adults may leave voluntarily. Older patients often die naturally. People using drugs can overdose. Bodies found far from home may remain unidentified. Rare predation hides among common explanations because the common explanations usually are correct.

If homicide is recognised, investigation normally begins locally. Detectives work from relationships, witnesses, movements, communications and physical evidence. That is rational because much homicide arises from known relationships, conflict and circumstances unlike the stranger-predator story. A base-rate problem appears at once. Starting every murder inquiry by hunting a serial killer would consume resources and misdirect attention. Treating every case as familiar can miss the rare exception. Good investigation must respect the common case without turning it into a prison.

The offender's role is not passive. After the first killing, that person may choose never to repeat, may want to repeat but lose access, may be interrupted for another reason or may deliberately seek another victim. News coverage can reveal whether a body was found and how police classified the death. Silence can supply information too. If no appeal appears and no questions reach the offender's circle, the event may seem to have vanished. The lesson need not be brilliance. It may be that this route appeared to work.

A survivor can close the pathway before the dead reveal a pattern. A non-fatal assault may supply a description, vehicle, contact method, location or sequence that later makes another case legible. Yet attempts are often recorded under different offence categories and may sit in systems separated from homicide files. Definitions that count only completed murders are necessary for clean statistics and poor guides to active investigation. Assaults, abductions and failed attacks may expose a continuing offender before a second death creates the formal category.

Whitechapel sells the pattern

In 1888 a cluster of murders in London's Whitechapel district gave the modern city an early serial-killer spectacle. The women were poor, several sold sex, and the killings arrived in a district already crowded with anxieties about migration, poverty, policing and urban disorder. Newspapers competed for readers while a reorganised police system tried to make sense of crimes that appeared connected. Letters claiming to come from the murderer supplied a name that proved more durable than any identified offender.

The case did not invent repeated murder, and it did not establish who committed the canonical killings, how many belonged together or whether the famous letters came from the killer. Those uncertainties helped rather than hindered the mythology. An unknown offender can absorb every fear because no disappointing person arrives to limit him. The blank space filled with medical men, foreigners, aristocrats, butchers, conspiracies and later fictional geniuses.

Whitechapel established a cultural bargain. The victims became a set, while the possible killer became an identity. Repetition invited the public to search for a pattern across bodies, wounds, dates and places. The press turned the search into serial entertainment, releasing each development into a market hungry for the next instalment. The crimes remain historically important, but their detailed suspects, letters and police failures belong to Jack the Ripper in a Hurry. Here they mark the moment when an unsolved series became a reusable public form.

The form rewards uncertainty. Each disputed victim extends an argument about the killer's beginning and end. Each suspect supplies a new biography. Each alleged letter gives the absent offender a voice. The case can therefore generate fresh certainty without new evidence. That pattern recurs throughout serial-killer culture: ambiguity is marketed as depth, while the people whose deaths created the case remain fixed at the edge of the frame.

For much of the next century, repeated killers did not sit inside one settled category. Police and psychiatrists used terms such as multiple murderer, lust murderer and mass murderer. Cases were known, but their boundaries varied. A poisoner inside a family, a doctor killing patients and a stranger attacking women in public could be filed under different moral and professional problems. The public image narrowed around the most dramatic version long before the data justified it.

The category is assembled

The FBI did not invent the language from nothing. Berlin police official Ernst Gennat used Serienmörder in professional writing in 1930, and James Brophy used serial murder in English in 1966. Other writers distinguished repeated killings from mass murder before the FBI programme took shape. What changed from the late 1970s through the 1980s was not the first appearance of two words. It was the assembly of a category with institutional force and a mass audience.

The United States supplied the setting. Several heavily publicised offenders were active, recently captured or still unresolved. Ted Bundy, the Green River case and BTK entered public memory while interstate mobility and fragmented jurisdiction made the problem feel national. Television carried local crimes across the country. In the 1990s, The Silence of the Lambs fixed the profiler-versus-killer duel as a dominant fictional form. The cultural image and the investigative speciality grew together.

Within the FBI, the Behavioural Science Unit interviewed imprisoned offenders and tried to convert case experience into usable investigative knowledge. The work encouraged investigators to compare crimes separated by place and time, examine victim selection, preserve behavioural detail and ask whether one offender might explain several cases. Motives were grouped. Crime scenes were described as organised or disorganised. The profiler emerged as an expert who seemed able to move backwards from behaviour to the person who produced it.

The achievement was real. Repeated murder became a problem for structured comparison rather than folklore alone. The limitations were present from the start. Early typologies drew heavily on small, selected groups of captured male sexual murderers who agreed to interviews. Their accounts were retrospective and open to lying, minimisation, boasting and memory change. The cases were already solved, which made coherence easier to find. Categories built from that material were then applied to unknown offenders far beyond the original sample.

The interviews also entered a feedback loop. Investigators asked questions shaped by existing theories. Offenders learned which answers made them appear calculating, damaged or unique. Books, films and later documentaries circulated the same language. A statement about fantasy or ritual could be sincere, strategic and culturally borrowed at once. The more familiar the category became, the less safe it was to treat an offender's self-description as untouched evidence from inside the mind.

The organised-disorganised distinction carried the promise in a memorable form. An organised scene was associated with planning, control and social competence; a disorganised one with impulsivity, chaos and poor functioning. Many crimes contain both kinds of feature. Some apparent organisation is required by the offence. Some disorder is produced by victim resistance or interruption. In 2004 David Canter and colleagues analysed thirty-nine features across one hundred United States serial murders. They found common organised features but no distinct pair of organised and disorganised types.

That finding did not make behaviour irrelevant. It reduced the claim. Crime-scene information can generate questions, suggest links and direct attention to probabilities. It cannot reliably reconstruct a complete stranger's biography. The category became useful when it made separated evidence comparable. It became myth when comparison was sold as mind-reading.

Making institutions remember together

A series crossing jurisdictions exploits an administrative fact: police forces remember locally. Two murders can be similar in the world and unrelated in the records because different forms were used, one force did not submit the case or nobody had reason to search the other jurisdiction.

The FBI established the Violent Criminal Apprehension Program, ViCAP, in 1985 to collect and compare information on serious violent crimes. The ambition was to give separate investigators a shared memory. Details about victims, circumstances, methods, locations and evidence could be searched beyond one officer's experience.

The work is harder than matching unusual wounds on a screen. Common behaviours can create false similarities. Rare details may be recorded differently. An offender can change method, and two offenders can imitate the same method because the environment makes it practical. Analysts have to compare both resemblance and difference, then ask whether the pattern is more informative than the base rate. A cluster is a lead, not a conclusion.

Analysts need rival explanations for any apparent cluster. The same pattern may reflect one offender, a shared victim setting, widely available means, copied publicity or chance. Apparent differences do not rule out a link if access or method changed. Confidence should depend on independent convergence and a model that anticipates anomalies, not one that explains them away after the fact. When every contradiction is recast as evidence of cunning, the hypothesis has become impossible to test.

Physical evidence added another form of memory. DNA profiling could connect scenes even when no suspect was known, while national databases could compare profiles across cases. Later, investigative genetic genealogy reopened some dormant series by linking crime-scene DNA to relatives in consumer or public genealogical data, subject to law and policy that vary by jurisdiction. Phones, payment records, cameras and vehicle data can also supply movement, contact and transaction records that earlier investigations lacked.

These tools do not abolish the human choices before them. A sample must have been collected and preserved. A case must have been entered. An unidentified body must be examined and compared. A missing-person report must survive assumptions about voluntary absence. The best database cannot retrieve a life that the system never represented.

The 2014 FBI study Serial Murder: Pathways for Investigations shows both the advance and the limit. It coded 480 cases attributed to 92 offenders and examined how victim encounter, killing location, body disposal and discovery related. The structured data offered investigators testable patterns. Yet the cases were a selected United States sample of single male offenders from 1960 to 2006 whose files reached the FBI's specialist centre. It was research on a defined set, not a portrait of every serial murderer.

When a long series becomes visible

Long series are often discovered through an accumulation rather than one masterstroke. A survivor describes an offender. A detective reopens an old file. A family keeps pressing. An arrest for another crime produces DNA. A suspect confesses and supplies details only the offender should know. Analysts compare a cluster. Remains receive a name. Each route changes what the series is thought to contain.

Samuel Little's confessions show the process in reverse. He was already imprisoned when investigators began connecting his accounts to unresolved deaths across the United States. He described victims and locations, sometimes drawing faces from memory. In 2019 the FBI reported ninety-three confessions, judged them credible and said fifty had then been verified. Verification demanded more than belief in a prolific offender. Investigators had to locate a matching death or disappearance, reconcile dates and places and find corroborating detail. Some cases could not be closed because records were thin, bodies absent or jurisdictions unable to establish enough.

Confession creates its own risk. Some offenders exaggerate, adopt unsolved crimes or enjoy controlling investigators. Others minimise. A credible account still needs testing. “Confirmed victim” can mean conviction, strong police attribution, corroborated confession or another standard depending on the database. One offender can therefore have several totals: convicted, admitted, linked, suspected and claimed. Repeating the largest number without its label converts uncertainty into publicity.

Harold Shipman's series emerged through a different route. Concerns about death patterns and a forged will led to investigation and conviction for fifteen murders. The later public inquiry examined hundreds of patient deaths and reached conclusions far beyond the trial. The first report attributed at least 215 deaths to him and identified more that raised serious suspicion. The widening total was not a discovery that his mind had changed. It was a change in the evidence examined and the standard used.

Both cases demonstrate why a serial killer's victim count is not printed at the moment of arrest. It is reconstructed, often for years, from records created under different assumptions. The number may rise, narrow or remain a range. Certainty attaches to victims one by one.

Identification comes before attribution. Unnamed remains cannot be placed into the person's movements, relationships and last known contacts. A missing person without usable records cannot be compared confidently with a body. Advances in dental comparison, fingerprints, DNA and genealogy can restore a name, but the administrative work is as important as the laboratory result: locating relatives, correcting dates, merging duplicate files and sharing information across agencies. A series can become visible because one victim becomes a person in the system again.

The decline and the afterlife

A 2019 peer-reviewed study of assembled United States cases found a marked fall in known serial-homicide series from the 1980s peak into the 2010s. The study was not a national census and the newest period had less time to mature, but the assembled record still supports a substantial directional decline. It sits awkwardly beside a culture in which serial killers seem permanently abundant.

Several changes could have reduced repetition. DNA and linked databases made disconnected scenes easier to compare. Cameras, mobile phones and digital transactions reduced some forms of anonymity. Hitchhiking declined. Street-based sex work and other victim environments changed. Long prison sentences removed violent offenders. Police became more alert to cross-jurisdictional series, while homicide itself changed. These are plausible contributors rather than measured shares. The trend may differ across offender types, countries and victim groups.

The data contain a built-in delay. Cases from the 1970s have had decades to be linked. Recent cases have not. Cold-case work can add an offender or victim to an old decade today. The 2019 study also examined claims that hidden serial killers explain large proportions of unresolved homicide and missing people, but its mixed methods and incomplete narrative records limited exact extrapolation. The defensible conclusion is directional: known United States series are far less common than at the late twentieth-century peak, while the exact curve and causal balance remain unsettled.

Culture moved the other way. Films, novels, podcasts, documentaries and online communities keep the peak decades in permanent circulation. Old offenders produce new content. The archive is replenished without new crimes, so public exposure can rise while identified incidence falls. The category has become a genre with conventions stronger than its evidence: the brilliant loner, the ritual, the profiler and the final duel.

A decline creates a counterfactual problem as well. A person arrested after one killing may have intended another, or may never have killed again. No ethical experiment can leave the offender free to find out. Prevention cannot be measured by counting the long series that failed to occur. This uncertainty limits causal claims without weakening the practical case for faster linkage, independent scrutiny and attention to threatened groups.

The cultural afterlife affects investigation. Public tips may help, but rumours can attach unrelated disappearances to a supposed series. Families may feel compelled to make a case media-friendly before it receives attention. Offenders can seek notoriety or borrow scripts already waiting for them. Serial murder is both a rare criminal sequence and a cultural object capable of shaping what police, offenders and audiences expect to see.

The United States cannot stand for the world. Countries differ in homicide recording, police capacity, mobility, missing-person systems, media freedom and the likelihood that old cases are digitised. A low recorded count may mean low incidence, weak linkage or both. International comparison is useful only when definitions and institutions remain visible. Otherwise a league table ranks data systems as much as offenders.

How we know

The strongest evidence concerns named, detected and well-documented cases, which means it is selected. Official reports describe investigative practice and known offenders, not the unseen population. Prison interviews exclude those never caught and rely on people who may lie, boast, minimise or reinterpret their past. Psychological studies are often small or limited to sexual homicide, convicted men or files sent to specialist units. Their findings cannot silently become rates for serial murder worldwide.

Victim totals vary with evidential thresholds. Demographic tables inherit the biases of solved cases. Recent trend data are incomplete because links emerge years later. Cross-country comparison is weakened by different definitions, police systems, homicide records and publication practices. Historical cases add retrospective labels and lost evidence.

This book therefore uses exact figures only where the definition, population, period and source are visible. It treats FBI reports as bounded practitioner evidence, peer-reviewed studies as tests of specified samples and specialist compilations as research tools rather than censuses. The most secure findings are structural: serial murder consists of separate events, arises through varied pathways and is measured through detection. Available estimates place it as a small share of homicide, but not as a globally comparable rate. The least secure claims are one universal mind, one hidden total and one settled cause of the decline.

What People Get Wrong

“A serial killer has killed at least three people with a cooling-off period”

The rule still sounds scientific because it has a number and a psychological interval. It is a historical convention. Different agencies and researchers have used two, three or more victims, while current United States federal law uses three or more killings in a provision governing federal investigation. The threshold does not create a new kind of mind.

Cooling-off is no firmer. Time between offences may reflect planning, loss of access, imprisonment, illness, work, family life or chance. Researchers cannot infer emotional cooling from a calendar alone. The FBI's 2005 symposium adopted two or more unlawful killings by the same offender or offenders in separate events and omitted a required cooling-off period.

The correction matters because rigid thresholds can hide prevention. A person arrested after two separate murders may present the same investigative problem as someone who reached three. It also prevents timing from being mistaken for motive. Separate events define the structure. The inner state remains a question requiring direct case evidence. A definition is useful when it makes cases comparable; it becomes misleading when the convenience of counting is mistaken for a boundary in nature.

“They are all white men”

The image comes from a selected canon of United States sexual stranger murder: the cases most heavily covered, studied and dramatised during the late twentieth century. Many influential research samples also consisted of captured male offenders, sometimes by design. The sample then became a portrait of the population.

Documented serial murderers include women and people from many racial groups. Women have appeared in caregiving, family, financial and partnership contexts as well as stranger cases. Men remain the majority in most large datasets, but the proportions vary with country, period, definition and detection.

The correction is not a replacement stereotype. One dataset's racial percentages do not describe humanity, and an all-male study cannot establish that women are absent. Demographics help test specific claims only when the denominator and selection process are visible. Treating the familiar white male face as a diagnostic clue risks missing offenders who do not match the cultural casting. It can also distort victim review, since investigators may overlook repeated deaths in domestic, medical or financial settings that do not resemble the stranger-hunter script.

“Psychopath means serial killer”

Psychopathy describes a pattern of traits involving callousness, manipulation, shallow emotion and antisocial behaviour. It does not name a crime. Most people with elevated psychopathic traits are not serial murderers, and not every serial murderer meets a high psychopathy threshold. The same is true of antisocial personality disorder, which overlaps with psychopathy but is a different clinical construct.

The myth persists because psychopathy seems to explain the emotional impossibility of repeated killing. In some cases it helps explain deception, lack of remorse and instrumental cruelty. It does not explain victim access, motive, timing or why an offender remained free.

Using the word as a moral superlative produces false expertise. A cold act is not a diagnosis, and a diagnosis is not a forecast. The useful question is which traits affected which decisions in this case. Anything broader turns a probabilistic construct into a monster label. It also reinforces stigma against people with mental-health diagnoses, most of whom are not violent and whose conditions tell us little about this rare offence.

“The Macdonald triad reveals them in childhood”

The triad joins animal cruelty, fire-setting and persistent bed-wetting into a supposed childhood warning pattern. It survives because three memorable signs make later violence feel retrospectively legible. Biographies can nearly always locate disturbing behaviour once the adult outcome is known.

A critical review found that the three behaviours rarely co-occur and that the evidence does not substantiate the package as a predictor of later violence. Individual behaviours may signal distress, maltreatment, conduct problems or a need for safeguarding. They do not identify a future serial murderer.

The base rate destroys the fantasy of screening. Serial murder is exceptionally rare, while childhood adversity and behavioural problems are far more common. Even a moderately associated sign would generate an overwhelming number of false predictions. Children need help because of what is happening to them or what they are doing now, not because adults imagine a murderer hidden decades ahead. Animal cruelty and deliberate fire-setting can demand urgent intervention, but the intervention should address harm, safety and underlying distress rather than attach an adult criminal destiny.

“The organised-disorganised profile identifies the offender”

The distinction offers a clean reversal from scene to person. Planning, control and concealment imply an organised, socially competent offender; chaos and impulsivity imply a disorganised, impaired one. It became persuasive through FBI authority, memorable cases and stories in which a profiler's inference lands exactly.

The empirical structure is weak. David Canter and colleagues examined thirty-nine features in one hundred United States serial-murder cases and found no distinct organised and disorganised types. Many organised features were common across cases, while mixed scenes were normal. Victim resistance, interruption and circumstance can produce disorder without revealing a stable personality.

Behavioural analysis can still generate leads, compare cases and test geographical or practical hypotheses. The correction is about confidence. A profile should be a provisional set of probabilities, not a suspect description treated as evidence. The moment contradictions become proof that the offender is unusually clever, the method has stopped being testable. Its performance should be judged against alternatives and misses, not by collecting the memorable details that happened to fit after arrest.

“They always escalate and leave a signature”

Narratives prefer a rising curve: fantasy intensifies, violence worsens, intervals shorten and a personal ritual becomes clearer. Some offenders do change in those directions. Others vary, pause, reduce risk, switch victims, alter methods or stop for reasons investigators never learn. Arrest, age, health, relationships and access can interrupt a pattern without satisfying a psychological arc.

Modus operandi is practical and can change through learning. Signature is an interpretive term for behaviour thought to express needs beyond the minimum required. It is not a fixed barcode. The same offender can omit an act, and different offenders can produce similar behaviour. Evidence may also be lost or altered by the scene.

The myth matters because it turns inconsistency into exclusion. Investigators may separate linked crimes because one detail changed, or join unrelated crimes because one dramatic feature matches. Patterns are strongest when several independent facts converge and the proposed link explains both similarities and differences. Courts and media can overread the language too, converting an examiner's cautious behavioural comparison into the claim that one unique personality must have produced both scenes.

“There are dozens of active serial killers near you”

Claims about an exact hidden population circulate because they combine fear with apparent insider knowledge. They are rarely traceable to a census, because no census exists. Estimates may extrapolate from unsolved murders, missing people, unidentified remains or old clearance rates, then assume that a chosen fraction belongs to serial offenders.

The same 2008 FBI estimate put the share below one per cent of annual murders. A 2019 peer-reviewed study also found a marked decline in known United States series from their 1980s peak. Neither statement proves that every series is detected or that present risk is zero. Together they show why a precise claim about how many hidden offenders are operating now carries more certainty than the records allow.

Active is itself ambiguous. It can mean a known offender whose case remains open, an unidentified offender linked to recent deaths, a person believed to be between offences or a statistical estimate of offenders nobody has recognised. Those are different populations. Adding them creates a dramatic total without one evidential standard. Excluding dormant, geographically scattered or misclassified cases can miss real harm.

Unsolved homicide and missing-person cases deserve attention without being recruited into a thriller. Hidden victims exist, and sensational totals can still be wrong. Uncertainty is a reason to improve reporting, identification and linkage, not a licence to invent the number. Fear grows from the numerator alone. Judgement begins by asking what was counted, over which period, against which denominator and through which system of detection.

Use It

Start with the denominator

A number about serial killers should trigger two questions before an emotion: counted out of what, and counted by whom?

“Fifty victims” measures scale inside one case. It does not tell you the share of homicide, the risk to the public or whether the total combines convictions with suspicions. “One-third had this trait” means little without the sample: convicted men, sexual murderers, files sent to one specialist unit, offenders who agreed to interviews, or every documented case in a database.

Apply the same discipline to warning signs. A behaviour may be common in a selected offender group and useless for prediction if it is common outside that group and the outcome is rare. Ask for the base rate and false positives. The stranger the event, the more seductive the numerator and the more work the denominator must do.

This does not minimise victims. It stops their deaths being converted into inflated public fear or false expertise.

Map the sequence

When separate harms recur, do not begin by searching for one damaged personality. Reconstruct the path from one event to the next.

What renewed choice or intention was present? Who had repeated access? Which role supplied trust, privacy or mobility? What interrupted the path, and what failed to? Which explanation ended scrutiny after the first event? Where did responsibility change hands? Who was exposed but weakly represented in records? What signal would have appeared if the events had been compared?

These questions move prevention from prophecy to design. Independent death review can reduce one doctor's control over care and certification without predicting temperament. Comparing assault reports, missing-person files and unidentified remains can reveal a pattern without a flawless psychological portrait. A willing offender may remain, but one or more gates to repetition can close.

Use the sequence carefully. Opportunity is not blame, and institutions do not author the crime. Describing how somebody became reachable should increase scrutiny of the offender and the conditions around the offence. The decision to kill remains the offender's.

Separate trait, diagnosis, motive and act

Four claims are often collapsed into one. A person can possess a trait, meet criteria for a diagnosis, experience a motive and commit an act. None proves the next.

Callousness is a trait. Antisocial personality disorder is a diagnosis. Sexual control or financial gain may be a motive. Murder is an act with legal and moral meaning. Psychopathy is a structured construct that overlaps some categories without fusing them. Severe mental illness can shape perception in particular cases without becoming a synonym for violence.

When somebody explains a crime by naming a label, ask what work the label performs. Does it identify a tested mechanism, or restate horror in clinical language? Does the evidence come from assessment, observed behaviour or a commentator's impression? Would the label have predicted anything before the crime?

The distinction protects accuracy and people who share a diagnosis or trait but not the act. It also forces explanation back towards decisions, circumstances and evidence.

Treat profiles as hypotheses

A useful profile should narrow inquiry while remaining capable of being wrong. Write its claims as probabilities and attach a test.

Instead of declaring that an offender is socially isolated, ask which scene features make isolation more likely than the alternatives and what would weaken that inference. Instead of announcing a shared signature, identify the independent similarities, their rarity and the meaningful differences. Compare the proposed link with the chance that common methods recur in the same victim environment.

Keep the profile away from the evidential throne. Behavioural advice can guide searches, interviews and data review. It does not identify a person, establish guilt or excuse weak physical and testimonial evidence. Record misses as well as hits. A prediction visible only after arrest is interpretation, not a demonstrated forecast.

Intuition can generate questions. Procedure must restrain answers. Mystery fiction reverses the order because certainty is more entertaining.

Read the record through selection

Every serial-murder dataset has an entrance policy, written or hidden. Cases enter because somebody recognised a homicide, identified a victim, linked events, named an offender and preserved enough information to code. Each gate selects.

Before accepting a demographic portrait or trend, reconstruct those gates. Are unsolved cases included? Are women, accomplices, medical killings and profit cases sought as actively as male sexual stranger murder? How are historical cases handled? Which countries publish usable records? How long has the latest decade had to mature?

Then find the missing comparison group. Interviews describe captured offenders who spoke, not those who escaped detection or refused. A study of body disposal may guide cases resembling its sample without describing offenders who left no recoverable body. A large database can reduce random noise while retaining systematic blind spots.

Add one phrase to every result: among the cases that became visible in this way. That phrase often changes the conclusion.

Refuse the celebrity bargain

The standard bargain offers understanding in exchange for attention to the offender. Learn the nickname, body count, rituals, childhood and quotations, while the victims appear as proof that the story matters.

Refuse it. Names may be necessary for history, but notoriety should not determine explanatory weight. Ask whether a detail clarifies mechanism, evidence or consequence. A lurid detail that does none of those jobs is taking space from the conditions that allowed repetition and the lives reduced to a count.

Notice the cover image, episode title and whose voice receives dramatic reconstruction. Check whether an uncertain victim total is presented beside the offender's largest claim. Prefer work that distinguishes confirmed from suspected cases, explains source limits and gives victims more than a category.

Attention can help identify the dead and solve cases. It can also reward performance, spread myths and turn harm into a franchise. Judge it by what it enables.

The limits

A one-hour account cannot explain why any named person killed. That requires case records, clinical evidence, history and uncertainty specific to the individual. General mechanisms identify conditions and recurring patterns, not private motives by remote diagnosis.

The data remain disproportionately American, male, solved and modern. Sexual stranger murder receives more analysis than medical, domestic, financial and team offending. Definitions differ. Victim totals remain disputed. Recent decades are incomplete. Unknown cases cannot show how unlike known cases they are.

Opportunity and institutional failure must not swallow agency. Systems can make repetition easier or harder, but the decision to kill belongs to the offender. Better linkage cannot remove every series. Some crimes leave little evidence, some victims remain unidentified and some institutions improve unevenly.

Explanation can also become fascination wearing a lab coat. A precise account of fantasy or victim choice may teach less than it appears while giving the offender another stage. The subject deserves analysis because prevention and justice matter, not because cruelty becomes meaningful when classified.

The one thing to keep

Keep the sequence.

The label serial killer makes a person look like a completed type. The sequence reveals events that might have ended earlier. Someone killed, chose or prepared to kill again, reached another person, passed through failed interruption and remained unconnected to the previous event long enough to repeat. Change any part and the later category changes with it.

That does not make the offender a passenger in a system. Violence begins with agency. Minds, motives, fantasies and traits matter because they affect decisions. They do not place the next victim within reach, preserve the offender's freedom or guarantee that separate evidence remains separate. A person can be willing and unable. Another can possess access and never abuse it. Explanation requires both halves without allowing either to erase responsibility.

The sequence changes how famous cases look. A long victim count stops being a medal for criminal intelligence and becomes a record of people reached, signals missed, evidence delayed, luck accumulated and choices repeated. The total can widen years after arrest when a confession is corroborated, remains are identified or an old record is compared. The number is an investigative conclusion with a date, not a quantity engraved into the offender.

It also changes prevention. Success may be a survivor believed, a suspicious death reviewed, a DNA profile linked, a missing-person report treated as urgent, an unrelated arrest or a family member who keeps asking. The intervention may prevent the third killing and the future label. The counterfactual victim never enters a database, which makes success hard to count. It remains the result that matters.

Popular culture asks what kind of monster could do this. Keep that question, but do not let it close the inquiry. Ask what decision came next, who was reachable, what might have interrupted the route and why the events remained separate. The mind explains part of the first gate. The myths hide the rest. The numbers are what the sequence leaves behind.

Terms

Serial murder. The practical FBI definition begins at two victims killed unlawfully by the same offender or group in distinct events. Research definitions still vary substantially.

Serial killer. A person attributed a qualifying series of murders. The label describes repeated acts under a chosen definition; it is not a psychiatric diagnosis or stable personality type.

Mass murder. Multiple killings within one event or closely connected incident. The structure differs from serial murder because the deaths present themselves as one collective scene from the beginning.

Spree murder. An older category for killings at two or more locations with little meaningful interval. Its reliance on inferred cooling-off creates overlap, so many modern definitions avoid it.

Separate event. A killing episode distinguishable in time or circumstance from another. Separation creates the investigative problem of recognising that apparently completed crimes share an offender.

Cooling-off period. A supposed emotional return towards normality between murders. Elapsed time cannot establish an offender's internal state, so the concept is no longer mandatory in the main FBI working definition.

Victimology. Study of a victim's life, routines, relationships and circumstances for investigative understanding. It should explain contact and exposure without turning circumstance into blame or worth.

Victim selection. The process by which an offender encounters, chooses or accepts a target. Selection can reflect motive, access, opportunity, prejudice, convenience and assumptions about institutional response.

Case linkage. Assessment of whether separate offences were committed by the same person or group. It uses converging behavioural, geographical, forensic and contextual evidence while retaining alternatives.

Linkage blindness. Failure to recognise related crimes because information is fragmented, cases cross boundaries, descriptions differ or investigators assume local explanations. It is an institutional condition, not a diagnosis.

Modus operandi. The practical methods used to commit an offence and manage its risks. It is learned and adaptive, so it may change with experience, resistance, circumstance or available tools.

Signature. Behaviour interpreted as expressing psychological needs beyond what was required to commit the crime. It may assist linkage, but it is neither fixed nor unique enough to identify a person.

Confirmed victim. A victim attributed to an offender under a stated evidential standard. Conviction, corroborated confession and strong investigative linkage are not identical, so totals need labels.

Cold case. An unsolved investigation no longer receiving continuous active work, though definitions differ by agency. New evidence, identification or comparison can return it to active inquiry.

Investigative genetic genealogy. Use of genealogical searching and family-tree analysis to generate leads from crime-scene DNA. A genealogical lead requires lawful access, conventional confirmation and jurisdiction-specific safeguards.

Psychopathy. A research and forensic construct involving interpersonal, affective and behavioural traits such as manipulation, callousness and persistent antisocial conduct. It neither predicts nor defines serial murder.

Antisocial personality disorder. A clinical diagnosis involving a pervasive pattern of disregarding and violating others' rights. It overlaps with psychopathy but places different weight on behaviour and history.

Sexual sadism disorder. A diagnosis involving recurrent sexual arousal from another person's suffering plus specified distress, impairment or action with a non-consenting person. It applies to some cases, not the category.

Paraphilia. An atypical sexual interest. It is not automatically a disorder, crime or danger; significance depends on distress, impairment, consent, conduct and the specific interest.

Fantasy. An imagined scene, role or sequence carrying emotional or sexual reward. Fantasy may organise some offending, but most fantasies never become acts and cannot establish intent alone.

Instrumental violence. Violence used as a means towards another goal, such as money, concealment or control. The term describes function in an event, not a complete personality or motive.

Expressive violence. Violence through which emotion, grievance or identity is discharged or communicated. Real offences can combine expressive and instrumental elements, so the terms are analytical tools rather than boxes.

Predatory violence. Planned or opportunistic aggression in which an offender seeks, approaches or waits for a target. It contrasts with immediate reactive conflict but can still contain emotion and improvisation.

Organised-disorganised typology. A classification linking orderly or chaotic scene features to offender characteristics. Its clean two-type structure has not survived strong empirical testing, though individual behaviours may still inform inquiry.

Offender profiling. Inference about an unknown offender from crime and victim information. It can organise hypotheses and priorities, but it does not identify a suspect or replace corroborating evidence.

Behavioural investigative advice. Structured specialist input on behaviour, linkage, interviewing, risk or priorities. The term avoids implying that every useful product is a personality portrait.

Geographic profiling. Analysis of offence locations to estimate areas associated with an offender's activity or anchor points. It ranks search space probabilistically; it does not reveal an address from a map.

ViCAP. The FBI's Violent Criminal Apprehension Program, established in 1985 to collect and compare serious violent-crime information across jurisdictions and support recognition of linked cases.

Base rate. The underlying frequency of an event or trait in the relevant population. Rare outcomes make plausible warning signs produce many false alarms unless specificity is exceptionally high.

Right-censoring. Incomplete observation of recent cases because their final outcome or full series is not yet known. It makes the newest period look more settled than it is beside mature historical data.

Go Deeper

James Alan Fox, Jack Levin and Emma E. Fridel, Extreme Killing: Understanding Serial and Mass Murder, fifth edition (2023). Start here for a broad, readable scholarly overview. It distinguishes serial from mass murder, covers motives, partnerships, victim choice, fantasy, female offenders and prevention, and keeps the subject connected to wider homicide rather than sealing it inside folklore. The fifth edition is current enough to reflect recent debates, though its centre of gravity is the United States and it covers mass killing as well. Read it for range, then carry this book's warning about selected cases into every profile and prevalence claim.

Robert J. Morton and Mark A. Hilts, editors, Serial Murder: Multi-Disciplinary Perspectives for Investigators (FBI, 2008). This is the clearest primary institutional source for the modern working definition, common myths, investigative problems and the FBI symposium's rejection of one generic offender template. It is free, concise and written for practitioners rather than tourists. Its authority should be used precisely: it records expert consensus and investigative experience, not a global census or experimental proof. Read it to see which popular claims professionals themselves discarded, and to understand why separate events and case linkage matter more than the old cooling-off formula.

Philip Jenkins, Using Murder: The Social Construction of Serial Homicide (1994). Jenkins asks how a real form of violence became a late twentieth-century public crisis, professional speciality and cultural commodity. The phrase social construction does not mean the murders were invented. It means categories, estimates, experts, media incentives and political claims shaped what counted and how large the threat appeared. The book is older and focused on the United States, but that distance is useful because it captures the category while it was hardening. Read it beside the FBI report to watch institutional knowledge and public mythology develop together.

Kenna Quinet, “The Missing Missing: Toward a Quantification of Serial Murder Victimization in the United States” (2007). Read this article for the hardest numerical problem in the field: victims who never enter the count because they are not reported missing, remain unidentified, are misclassified or are never linked. Quinet's attempt to estimate the hidden total relies on assumptions that should be handled cautiously, and this book does not adopt its largest extrapolations. Its lasting contribution is methodological and moral. A victim database begins after several institutions have already decided who is visible. Read it to understand why undercount can be real without making every dramatic estimate sound.

Notes and Sources

Definition, terminology and scale

The working definition follows Serial Murder: Multi-Disciplinary Perspectives for Investigators, the FBI monograph based on a symposium held from 29 August to 2 September 2005: the unlawful killing of two or more victims by the same offender or offenders in separate events. Current 28 U.S.C. § 540B separately defines serial killings as three or more killings for a provision authorising federal investigation at the request of state or local authorities. That statutory threshold serves jurisdiction and should not be mistaken for the only behavioural definition. Specialists continue to debate motive exclusions, reliable linkage, accomplices and the boundary around rapid sequences.

Wayne Petherick, Shuktika Bose, Amber McKinley and Candice Skrapec's 2021 review documents the unstable history of both the term and the definition. It identifies Ernst Gennat's use of Serienmörder in professional literature in 1930 and James Brophy's English use of serial murder in 1966. The book therefore credits the FBI with institutionalising and popularising a modern category, not coining language that had never existed.

Jeffery Osborne and C. Gabrielle Salfati's 2015 study, together with Petherick and colleagues' review, supports replacing claims about emotional cooling with observable intervals. Time between offences can reflect access, interruption, incarceration, planning, illness or circumstance. It cannot establish a psychological reset by itself.

The FBI's statement that serial murder comprises less than one per cent of murders appears in the 2008 monograph. It is retained as a dated order-of-magnitude estimate, not a current annual statistic, global rate or offender headcount.

Samuel Little and deaths outside the expected pattern

The FBI page dated 6 October 2019 reported that Samuel Little had confessed to ninety-three murders, that analysts considered the confessions credible and that fifty had then been verified. These figures describe the verification status on that date. They are not presented as a permanent final count.

The same source states that many victims' deaths had been ruled overdoses or attributed to accidental or undetermined causes, and that some bodies were never found. Its account supports the mechanism of misclassification, missing records and cross-jurisdictional review. The body text paraphrases rather than quotes Little's claim that he believed nobody was accounting for his victims.

Minds, motives and mental state

The FBI practitioner literature and Fox, Levin and Fridel's synthesis support substantial heterogeneity in motive, victim relationship, setting, partnership and offender presentation. The manuscript does not present any motive typology as exclusive or exhaustive. Named cases illustrate different pathways rather than represent a population.

Austin Lam, Stephanie Penney and Alexander Simpson reviewed eight studies of psychosis in sexual homicide. Estimates ranged from zero to 27 per cent across diverse samples, with five studies reporting five per cent or less. Sexual homicide is narrower than serial murder, and the review did not establish the causal or motivational role of psychosis. The manuscript therefore says only that psychosis appears in some individual cases, is uncommon in most samples of that narrower literature and cannot supply a serial-murder prevalence rate.

Eric Elbogen and Sally Johnson's population research is used only for the broader warning that diagnosis alone is a poor explanation of violence and that substance use, life circumstances and other factors matter. It does not provide a rate for serial murder.

Psychopathy, sadism and prediction

Jennifer Skeem, Devon Polaschek, Christopher Patrick and Scott Lilienfeld's 2011 review supports the treatment of psychopathy as a heterogeneous research and forensic construct rather than a synonym for violence or a simple diagnosis. Structured measures require trained assessment and collateral evidence. The manuscript does not diagnose named offenders from public material and does not claim that psychopathy has no association with violence.

Definitions of antisocial personality disorder, paraphilia and sexual sadism disorder follow the Diagnostic and Statistical Manual of Mental Disorders, fifth edition, text revision. The visible definitions are deliberately compressed. Psychopathy is not listed under that name as a DSM-5-TR diagnosis.

Charlotte Hannah Parfitt and Emma Alleyne's critical review of the Macdonald triad found that animal cruelty, fire-setting and persistent enuresis rarely co-occur as a package and that the literature does not substantiate the triad's predictive premise. The book preserves the separate safeguarding significance of harmful or distressed childhood behaviour while rejecting a serial-murder screen.

The base-rate argument is mathematical. When an outcome is exceptionally rare, a marker with imperfect sensitivity and specificity will identify far more people who never produce the outcome than people who do.

Opportunity, access and healthcare settings

The Shipman Inquiry's first report, Death Disguised, published on 19 July 2002, concluded that Harold Shipman had killed at least 215 patients and identified additional deaths that raised serious suspicion. The manuscript uses the secure lower finding from that report rather than larger totals assembled under mixed standards.

Department of Health and Social Care guidance, updated 14 August 2024, states that from 9 September 2024 every death in England and Wales requires independent review by a medical examiner or investigation by a coroner. The guidance cites the Shipman Inquiry's third report among several reports and inquiries that underlined reform. Publication date, implementation date and geographical scope are therefore separated in the narrative.

Elizabeth Wettlaufer pleaded guilty in Ontario in 2017 to eight murders, four attempted murders and two aggravated assaults involving people receiving long-term or home care. Justice Eileen Gillese's 2019 public inquiry concluded that the offences would not have been discovered without Wettlaufer's confession and that systemic vulnerabilities permitted them. Chris Frank's 2020 review places the case within the rarer and poorly measured problem of healthcare serial murder. The manuscript uses the inquiry for the case facts and systemic finding; it does not infer an incidence rate or a female offender type from one case.

Opportunity is used in the criminological sense of access, routine, guardianship and institutional conditions. It does not reduce responsibility or imply that people exposed to risk caused their victimisation.

Behaviour, profiling and linkage

David Canter, Laurence Alison, Emily Alison and Natalia Wentink tested the organised-disorganised distinction using thirty-nine features from one hundred United States serial murders. They found no discrete pair of types; organised features were common and disorganised features did not form a separate cluster. This finding attacks the dichotomy, not every form of behavioural analysis.

Brent Snook and colleagues' 2008 review supports the warning that belief in offender profiling has outrun its theoretical and empirical support, assisted by anecdotes, expert labels, repetition and selective attention to hits. The manuscript preserves a narrower role for behavioural and geographical analysis as a source of ranked, testable investigative hypotheses.

Modus operandi and signature follow practitioner usage but are treated as inferences with limits. Practical methods can change through experience and circumstance. Behaviour thought to express psychological needs can be absent, imitated, interrupted or misread. No scene feature is treated as a unique personal barcode.

The FBI's Violent Criminal Apprehension Program began at the FBI Academy in July 1985 to support the collection and comparison of serious violent-crime information across jurisdictions. It is used as an example of institutional memory, not as a claim of complete national coverage.

Zena Rossouw and Ted Palys's 2025 qualitative study drew on interviews with six experienced practitioners from Canada and the United States who had worked across more than forty serial-homicide investigations. Establishing a defensible connection between cases emerged as the primary hurdle, alongside missing persons, changing methods, limited resources and complex scenes. The sample is small, selected and centred on North American experience. It supports the practical account of linkage difficulty, not a frequency estimate or universal sequence.

Ellen Greytak, CeCe Moore and Steven Armentrout's 2019 review supports the high-level description of investigative genetic genealogy as a lead-generating method. The manuscript does not treat a genealogical lead as identification by itself. Legal access, database rules, conventional confirmation and privacy safeguards vary by jurisdiction.

The Pathways sample

Robert J. Morton, Jennifer M. Tillman and Stephanie J. Gaines's Serial Murder: Pathways for Investigations used 480 cases attributed to 92 offenders. The cases occurred in the United States from 1960 to 2006 and were submitted to, or solicited by, the FBI's National Center for the Analysis of Violent Crime. Files with missing or minimal information were excluded. The report states that the resulting sample was not representative of the United States and limits inclusion to single offenders acting alone; the analysed set was male.

The manuscript gives the population, observation period, route into the sample and exclusions wherever the result carries weight. Its patterns can support questions inside comparable cases. They cannot establish universal percentages for all offenders or victims.

Victims, missing people and unequal visibility

Kenna Quinet's 2007 article examines possible omissions from serial-murder counts through people never reported missing, unidentified remains, deaths assigned another cause and cases never linked. The methodological correction is retained. Its numerical extrapolations depend on assumptions about hidden populations and are not repeated as established totals.

Categories such as sex worker, runaway, patient or person using drugs matter because they can alter exposure, reporting, identification and institutional urgency. They do not reduce culpability for the crime or the value of the victim. The FBI monograph itself describes why series involving transient or weakly protected victims across multiple jurisdictions can be harder to identify.

Philip Jenkins's account of claims-making around serial homicide and L. Perry Curtis's study of the Whitechapel murders in the London press support the discussion of media selection. Publicity can aid identification and investigation while allocating fame, simplifying victims and stabilising myths.

Whitechapel, landmark cases and the modern category

The Whitechapel discussion is intentionally bounded. The canonical group of 1888 murders, authorship of the letters and the offender's identity remain disputed. Curtis supports the account of competitive press, urban anxiety and a durable killer identity. The book does not call Whitechapel the first serial murder and leaves suspects, letters and detailed police chronology to Jack the Ripper in a Hurry.

The FBI's 2008 monograph names the Green River case, Ted Bundy and BTK as cases associated with renewed public interest in the 1970s and 1980s, and The Silence of the Lambs as part of the 1990s cultural expansion. They appear here as landmarks in category formation, not as a complete history or evidence that those offenders were typical.

Jenkins supports the claim that serial homicide is both a real pattern of violence and a modern category shaped by experts, institutions, media and political incentives. Social construction here describes how boundaries, prevalence claims and public imagery formed. It does not deny the murders.

Counts, trend and data vintage

Enzo Yaksic and colleagues' peer-reviewed article was published in 2019 and compared assembled United States case records across decades, reviewed a defined sample of unresolved homicides and interviewed experts. It found a marked decline in known series from the 1980s peak into the 2010s. The article's observation periods, selected data and missing narrative detail prevent a national-census interpretation or a precise decomposition of causes.

The book does not use the unverified June 2026 Radford/FGCU report inherited from Stage 11. No current active-offender total is retained. The newest source publication date is therefore not mistaken for a newer or more complete observation window.

Right-censoring remains material. Older cases have had more time to be solved, confessed, linked or reclassified, and later cold-case work can add victims to earlier periods. The newest decade is an unfinished record, not a directly comparable end point.

International comparison is weaker. Definitions, homicide registration, missing-person systems, police capacity, digitisation and publication differ. No reliable global census or universally comparable current count of active serial killers was found as of 4 September 2026.

Evidence limits

The field is selected towards detected offenders, solved cases, identified victims, modern records, sexual stranger murder and United States material. Prison interviews add detail but cannot represent offenders who were never caught and may reflect strategic self-presentation. Historical cases are reconstructed through records made for other purposes.

The book therefore distinguishes established structure from contested prevalence. Separate events, heterogeneous pathways, the limits of psychopathy as a synonym, the failure of the Macdonald triad as a predictive package, the weak organised-disorganised dichotomy and detection-dependent measurement are supported strongly enough for direct prose. Exact hidden totals, one universal psychology and a ranked causal explanation for the decline are not.

Bibliography

Official records and institutional sources

American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 5th ed., text rev. Washington, DC: American Psychiatric Association Publishing, 2022.

Department of Health and Social Care. An Overview of the Death Certification Reforms. Updated 14 August 2024. Current guidance checked 4 September 2026.

Federal Bureau of Investigation. “Samuel Little: Confessions of a Killer.” 6 October 2019.

Federal Bureau of Investigation. “Violent Criminal Apprehension Program.” Programme history and case-linkage material, checked 4 September 2026.

Morton, Robert J., and Mark A. Hilts, eds. Serial Murder: Multi-Disciplinary Perspectives for Investigators. Quantico, VA: Federal Bureau of Investigation, 2008.

Morton, Robert J., Jennifer M. Tillman and Stephanie J. Gaines. Serial Murder: Pathways for Investigations. Quantico, VA: Federal Bureau of Investigation, 2014.

The Shipman Inquiry. First Report: Death Disguised. 19 July 2002.

United States Code. Title 28, § 540B, “Investigation of Serial Killings.” Current preliminary edition checked 4 September 2026.

Gillese, Eileen E. Report of the Public Inquiry into the Safety and Security of Residents in the Long-Term Care Homes System. 4 vols. Toronto: Queen's Printer for Ontario, 2019.

Modern scholarship and research

Canter, David V., Laurence J. Alison, Emily Alison and Natalia Wentink. “The Organized/Disorganized Typology of Serial Murder: Myth or Model?” Psychology, Public Policy, and Law 10, no. 3 (2004): 293-320. DOI: 10.1037/1076-8971.10.3.293.

Curtis, L. Perry. Jack the Ripper and the London Press. New Haven: Yale University Press, 2001.

Elbogen, Eric B., and Sally C. Johnson. “The Intricate Link Between Violence and Mental Disorder: Results from the National Epidemiologic Survey on Alcohol and Related Conditions.” Archives of General Psychiatry 66, no. 2 (2009): 152-161. DOI: 10.1001/archgenpsychiatry.2008.537.

Fox, James Alan, Jack Levin and Emma E. Fridel. Extreme Killing: Understanding Serial and Mass Murder. 5th ed. Thousand Oaks, CA: SAGE, 2023.

Frank, Chris. “Health Care Serial Murder: What Can We Learn from the Wettlaufer Story?” Canadian Family Physician 66, no. 10 (2020): 719-722.

Greytak, Ellen M., CeCe Moore and Steven L. Armentrout. “Genetic Genealogy for Cold Case and Active Investigations.” Forensic Science International 299 (2019): 103-113. DOI: 10.1016/j.forsciint.2019.03.039.

Jenkins, Philip. Using Murder: The Social Construction of Serial Homicide. New York: Aldine de Gruyter, 1994.

Lam, Austin A., Stephanie R. Penney and Alexander I. F. Simpson. “A Review of the Prevalence of Psychosis in Sexual Homicide.” Journal of the American Academy of Psychiatry and the Law 48, no. 1 (2020): 56-64. DOI: 10.29158/JAAPL.003887-20.

Osborne, Jeffery R., and C. Gabrielle Salfati. “Re-Conceptualizing ‘Cooling-Off Periods’ in Serial Homicide.” Homicide Studies 19, no. 2 (2015): 188-205. DOI: 10.1177/1088767914526716.

Parfitt, Charlotte Hannah, and Emma Alleyne. “Not the Sum of Its Parts: A Critical Review of the MacDonald Triad.” Trauma, Violence, & Abuse 21, no. 2 (2020): 300-310. DOI: 10.1177/1524838018764164.

Petherick, Wayne, Shuktika Bose, Amber McKinley and Candice Skrapec. “A Rose by Any Other Name: Problems in Defining and Conceptualising Serial Murder with a New Proposed Definition.” Journal of Mass Violence Research 1, no. 1 (2021): 4-24. DOI: 10.53076/JMVR85123.

Quinet, Kenna. “The Missing Missing: Toward a Quantification of Serial Murder Victimization in the United States.” Homicide Studies 11, no. 4 (2007): 319-339. DOI: 10.1177/1088767907307467.

Rossouw, Zena, and Ted Palys. “‘Baggage in the Business’: The Investigative Challenges of Serial Homicide.” The Police Journal 98, no. 1 (2025): 249-268. DOI: 10.1177/0032258X241269482.

Skeem, Jennifer L., Devon L. L. Polaschek, Christopher J. Patrick and Scott O. Lilienfeld. “Psychopathic Personality: Bridging the Gap Between Scientific Evidence and Public Policy.” Psychological Science in the Public Interest 12, no. 3 (2011): 95-162. DOI: 10.1177/1529100611426706.

Snook, Brent, Richard M. Cullen, Craig Bennell, Paul J. Taylor and Paul Gendreau. “The Criminal Profiling Illusion: What's Behind the Smoke and Mirrors?” Criminal Justice and Behavior 35, no. 10 (2008): 1257-1276. DOI: 10.1177/0093854808321528.

Yaksic, Enzo, Clare Allely, Raneesha De Silva, Melissa Smith-Inglis, Daniel Konikoff, Kori Ryan, Dan Gordon, Egor Denisov and David A. Keatley. “Detecting a Decline in Serial Homicide: Have We Banished the Devil from the Details?” Cogent Social Sciences 5, no. 1 (2019): 1678450. DOI: 10.1080/23311886.2019.1678450.