DSM-5: The Diagnostic and Statistical Manual of Mental Disorders

The DSM-5 is the primary reference used by psychologists, psychiatrists, and mental health professionals worldwide. Published by the American Psychiatric Association (APA), it is the fifth edition of a project that began in 1952. Its full title is the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, and if you work anywhere near clinical psychology or psychiatry, you will encounter it constantly.

But the DSM-5 is not simply a list of mental disorders. Behind each entry sits decades of research, scientific debate, and hard-won shifts in how we understand the human mind. Whether you are a general reader trying to make sense of a diagnosis you have received, a psychology student working through your first clinical placement, or a practising clinician who uses it daily, the DSM-5 has something different to offer each of you. This article tries to serve all three.

A brief history: where it came from

Before the DSM existed, there was no shared language between mental health professionals. A psychiatrist in New York and one in Edinburgh could discuss the same patient and mean two entirely different things. Diagnoses were subjective, inconsistent, and largely useless for research purposes.

The first edition, DSM-I, appeared in 1952. It ran to 130 pages and listed 106 disorders. By later standards, it leaned heavily on psychoanalytic theory rather than empirical observation. DSM-II followed in 1968, and it was during this period that homosexuality remained classified as a mental disorder. Its removal in 1973 became one of the most significant moments in the manual’s history, demonstrating that the DSM could and should change in response to scientific evidence rather than social prejudice.

DSM-III, published in 1980, was a genuine revolution. The descriptive, criteria-based approach replaced theoretical interpretation. For the first time, diagnoses rested on observable, measurable signs rather than on an analyst’s reading of a patient’s psyche. This shift standardised diagnosis in a way that made meaningful research possible. DSM-III-R made minor revisions in 1987, and DSM-IV arrived in 1994, remaining the standard for over two decades.

The DSM-5 was published in 2013 after fourteen years of work, hundreds of specialist committees, thousands of clinical trials, and scientific debates that occasionally made mainstream news. In 2022, DSM-5-TR (the text revision) appeared, updating descriptive text, epidemiological figures, and making targeted corrections rather than overhauling the diagnostic criteria themselves. One new diagnosis was added: Prolonged Grief Disorder.

The official DSM-5 website remains active and provides supporting resources for clinicians.

How the DSM-5 works

The core logic of DSM-5 rests on diagnostic criteria. For each disorder, a set of symptoms is defined, with a minimum number required over a specified period before a diagnosis can be made. These criteria come from clinical research, and they exist so that two different clinicians in two different countries can reach the same conclusion about the same patient.

Take major depressive disorder as an example. At least five of nine specified symptoms must be present during the same two-week period, and one of those five must be either depressed mood or loss of interest or pleasure. Without that kind of precision, two psychiatrists can examine the same person and prescribe entirely different treatments.

One thing worth being clear about: DSM-5 is a diagnostic tool, not an explanation of cause. A diagnosis of ADHD tells you what pattern of symptoms is present. It does not tell you why they are there, where they came from, or what the best treatment is. The manual is a map that tells you where you are. The route from there is worked out separately.

The fundamental criterion: distress and functional impairment

Every diagnosis in the DSM-5 shares a foundational requirement: the condition must cause clinically significant distress or impairment in social, occupational, or other important areas of functioning. Someone with unusual traits or behaviours who functions well and experiences no distress does not meet diagnostic criteria. This matters because it draws a line between human variation and clinical disorder, and that line is not always obvious.

The shift towards spectra

DSM-5 moved several conditions away from rigid categories and towards a spectrum model. Autism is no longer divided into separate subtypes (classic autism, Asperger syndrome, PDD-NOS) but sits on a single continuum from mild to severe. Substance use disorders dropped the DSM-IV split between “abuse” and “dependence” in favour of a single diagnosis graded by severity. These changes reflect a more honest picture of how these conditions actually present in clinical practice.

Mental disorders in DSM-5: the full picture

The manual contains around 157 main diagnoses across roughly 600 diagnostic codes, organised into 20 chapters. What follows covers each category with the level of detail that is actually useful, not just a list of names.

Neurodevelopmental disorders

This chapter covers conditions that emerge during development and affect how the brain grows and functions. It includes autism spectrum disorder (ASD), attention-deficit/hyperactivity disorder (ADHD), intellectual disability, specific learning disorders (including dyslexia and dyscalculia), communication disorders, and motor disorders such as tic disorders and Tourette’s syndrome.

One of the most debated changes in DSM-5 was the folding of Asperger syndrome into the autism spectrum. Many people who had built an identity around that diagnosis felt it was being taken from them. Clinically, the change was justified by evidence that no reliable boundary exists between Asperger’s and autism at the higher-functioning end of the spectrum. The shift was not about invalidating experiences but about diagnostic honesty.

For ADHD, DSM-5 raised the age of symptom onset from seven to twelve, and made it meaningfully easier to diagnose in adults. This was not ADHD becoming fashionable. It was the field acknowledging that large numbers of adults had been missed throughout childhood and were suffering the consequences.

Schizophrenia spectrum and other psychotic disorders

Schizophrenia, schizoaffective disorder, schizophreniform disorder, brief psychotic disorder, delusional disorder, and substance-induced psychotic disorder all sit here. The core features are delusions (fixed false beliefs), hallucinations (perception without external stimulus), disorganised thinking, disorganised behaviour, and negative symptoms such as reduced emotional expression and motivation.

DSM-5 removed the schizophrenia subtypes that appeared in DSM-IV: paranoid, catatonic, disorganised, residual, and undifferentiated. Research showed these subtypes were unstable over time and did not predict treatment outcome. In their place, DSM-5 uses symptom dimensions (psychosis, negative symptoms, disorganisation) that provide a more accurate and clinically useful description of what each patient is actually experiencing.

For general readers: schizophrenia is not what most media portrayals suggest. It is not dissociative identity disorder. The vast majority of people with schizophrenia pose no threat to others. With appropriate treatment, a meaningful and fulfilling life is achievable.

Bipolar and related disorders

Bipolar I disorder, bipolar II disorder, cyclothymic disorder, and bipolar disorders induced by substances or medical conditions make up this chapter. DSM-5 positions bipolar conditions as a conceptual bridge between the psychotic spectrum and the depressive spectrum, which reflects the biological and genetic evidence rather than any arbitrary editorial decision.

The distinction between bipolar I and bipolar II matters considerably in clinical practice. Bipolar I requires at least one full manic episode, lasting a minimum of seven days or requiring hospitalisation. Bipolar II involves hypomanic episodes (lasting at least four days, less severe than full mania) but no full mania. This distinction directly affects medication choices, and it is commonly missed: many patients with bipolar II present during depressive episodes and leave with only a depression diagnosis.

DSM-5 added a clarifying requirement: for hypomania and mania, the change in behaviour must be observable to others, not just a self-reported mood shift. This reduced the number of false positives.

Depressive disorders

Major depressive disorder, persistent depressive disorder (formerly dysthymia), disruptive mood dysregulation disorder (DMDD), premenstrual dysphoric disorder (PMDD), and substance or medically induced depressive disorders belong here.

DMDD is a DSM-5 addition created for children under eighteen who show chronic severe irritability and frequent temper outbursts. Its introduction was driven by concern that bipolar disorder was being over-diagnosed in children, particularly in the United States.

The removal of the bereavement exclusion was among the most debated changes in DSM-5. DSM-IV protected against a depression diagnosis in the two months following bereavement. DSM-5 removed this exception. The argument was that grief can trigger genuine clinical depression regardless of cause, and that withholding treatment because the precipitant was a death made no clinical sense. Critics argued the change would medicalise normal human grief. The debate has not entirely settled.

Anxiety disorders

This chapter includes separation anxiety disorder, selective mutism, specific phobia, social anxiety disorder, panic disorder, agoraphobia, generalised anxiety disorder (GAD), and substance or medically induced anxiety disorders.

Two notable changes from DSM-IV: obsessive-compulsive disorder and PTSD were moved to their own chapters. This reflected neurobiological and genetic research suggesting they have a fundamentally different character from anxiety. They share some surface features with anxiety but operate through different mechanisms.

The requirement that patients recognise their fear as unreasonable was also removed. Many people with genuine phobias believe their fear is entirely proportionate. Insisting on insight as a diagnostic criterion was excluding patients who clearly met every other standard.

Generalised anxiety disorder is one of the most misunderstood conditions in public conversation. Everyday worry is not GAD. GAD involves persistent, excessive, uncontrollable worry about multiple areas of life, lasting at least six months and accompanied by physical symptoms including muscle tension, fatigue, and disrupted sleep. The key word is uncontrollable: patients with GAD typically know their worry is disproportionate but cannot stop it.

Obsessive-compulsive and related disorders

OCD, body dysmorphic disorder (BDD), hoarding disorder, trichotillomania (hair-pulling), excoriation disorder (skin-picking), and substance or medically induced variants are gathered here.

The creation of this chapter as a standalone category was one of DSM-5’s more intellectually interesting moves. These conditions look quite different on the surface but share a common neural circuit involving the orbitofrontal cortex and striatum. That shared architecture helps explain why they tend to respond to similar treatments.

Hoarding disorder appeared as an independent diagnosis in DSM-5 for the first time, rather than as a symptom of OCD. Research had established it has distinct neural pathways and a different cognitive profile. Treating hoarding as OCD-adjacent had not served patients well.

A word on OCD for general readers: the phrase “I’m so OCD” used to describe a preference for tidiness is not just inaccurate but unhelpful. Clinical OCD involves intrusive, ego-dystonic thoughts (obsessions) that cause genuine distress, and compulsions that provide only temporary relief before the cycle restarts. Many people with OCD experience intrusive thoughts about harming others that are completely at odds with their own values and cause them profound shame. The public image of OCD as neat shelves and symmetrical objects misses most of what the condition actually involves.

Trauma and stressor-related disorders

PTSD, acute stress disorder, reactive attachment disorder, disinhibited social engagement disorder, adjustment disorders, and unspecified trauma or stressor-related disorders sit in this chapter.

DSM-5 restructured PTSD from three symptom clusters to four: intrusion, avoidance, negative alterations in cognitions and mood, and alterations in arousal and reactivity. The addition of the fourth cluster, covering persistent negative beliefs, distorted blame, emotional numbing, and feelings of alienation, acknowledged experiences that many PTSD sufferers reported but that fell outside the old criteria.

DSM-5 also added a separate subtype for children under six, with modified criteria reflecting the fact that young children express trauma differently. This matters because earlier criteria were making childhood PTSD systematically harder to diagnose.

Dissociative disorders

Dissociative amnesia, dissociative identity disorder (DID), and depersonalisation/derealisation disorder are the three diagnoses in this chapter. These conditions most commonly arise following severe trauma, particularly repeated trauma during childhood.

DID is probably the most misrepresented diagnosis in popular culture. What is usually depicted on screen is several generations removed from the clinical reality. DID involves a system of distinct identity states that alternately take control of behaviour and awareness, with significant amnesia between states. It is a dissociative response to overwhelming trauma during early development, a survival mechanism rather than a curiosity. The treatment is long-term, relationship-based, and the outcomes with good therapy are considerably better than public perception suggests.

Depersonalisation/derealisation disorder received standalone status in DSM-5. Many people experience brief episodes of feeling detached from themselves or from the world around them, particularly under stress or with poor sleep. In this disorder, those experiences are chronic, persistent, and severely distressing.

Somatic symptom and related disorders

Somatic symptom disorder, illness anxiety disorder (which replaced hypochondriasis), functional neurological symptom disorder (formerly conversion disorder), psychological factors affecting other medical conditions, and factitious disorder belong here.

The renaming of hypochondriasis to illness anxiety disorder was not cosmetic. “Hypochondria” carried connotations of delusion or exaggeration that made patients feel dismissed. “Illness anxiety disorder” places the emphasis on the genuine distress the person experiences, regardless of whether a physical cause is found.

The renaming of conversion disorder to “functional neurological symptom disorder” reflects the current understanding that the problem lies in neurological functioning rather than structure. Symptoms such as non-epileptic seizures, functional paralysis, and psychogenic blindness are real. They are not being manufactured or performed. The mechanism simply does not involve structural damage of the kind that appears on a scan.

Feeding and eating disorders

Anorexia nervosa, bulimia nervosa, binge-eating disorder, pica, rumination disorder, and avoidant/restrictive food intake disorder (ARFID) appear in this chapter.

Binge-eating disorder gained independent diagnostic status in DSM-5, having previously been listed only in an appendix. It is actually the most common eating disorder in the general population, considerably more prevalent than anorexia. Its clinical importance had been under-recognised for a long time.

ARFID was another DSM-5 addition. It covers restricted eating driven by sensory sensitivity, fear of choking or vomiting, or a traumatic eating experience, not by concerns about weight or body image. It affects children more commonly but adults are not immune.

For anorexia, DSM-5 dropped the requirement for amenorrhoea (loss of menstrual periods) as a diagnostic criterion, which had been excluding men and prepubertal girls from diagnosis. It also specified severity levels based on BMI: mild (17.0 to 18.5), moderate (16.0 to 17.0), severe (15.0 to 16.0), and extreme (below 15.0).

Elimination disorders

Encopresis (faecal incontinence) and enuresis (urinary incontinence) are both diagnosed here, with enuresis not typically considered clinically significant until after age five. Both are more common in childhood and frequently co-occur with other conditions, particularly ADHD and anxiety.

Sleep-wake disorders

Insomnia disorder, hypersomnolence disorder, narcolepsy, breathing-related sleep disorders (including obstructive sleep apnoea), circadian rhythm sleep-wake disorders, parasomnias (including sleepwalking, sleep terrors, nightmare disorder, and REM sleep behaviour disorder), and restless legs syndrome are covered in this chapter.

REM sleep behaviour disorder deserves specific attention. It involves people physically acting out their dreams during REM sleep, sometimes injuring themselves or a partner. Clinically, it has a strong association with Parkinson’s disease, Lewy body dementia, and multiple system atrophy. In many patients it precedes neurological diagnosis by years, sometimes by decades. Some specialists consider it a prodromal marker rather than a separate condition.

Sexual dysfunctions

Delayed ejaculation, erectile disorder, male hypoactive sexual desire disorder, female orgasmic disorder, female sexual interest/arousal disorder, genito-pelvic pain/penetration disorder, and substance or medication-induced sexual dysfunction are classified here.

DSM-5 merged several DSM-IV diagnoses for women. Hypoactive sexual desire disorder and female sexual arousal disorder became a single diagnosis (female sexual interest/arousal disorder) because research showed these two dimensions are typically intertwined rather than separable. This was an acknowledgement of clinical reality rather than a conceptual compromise.

Gender dysphoria

This chapter replaced “gender identity disorder” from DSM-IV. The change in terminology was deliberate. DSM-5 does not classify being transgender as a disorder. What falls within this chapter is the distress that can arise from the mismatch between an individual’s gender identity and their sex assigned at birth. Not every transgender person experiences this distress, and those who do not fall outside the diagnostic criteria. The distinction matters both clinically and in terms of how patients are treated within healthcare systems.

Disruptive, impulse-control, and conduct disorders

Oppositional defiant disorder, intermittent explosive disorder, conduct disorder, antisocial personality disorder, pyromania, and kleptomania are gathered here. What unites them is difficulty regulating behaviour and emotions in ways that harm the individual or others.

Intermittent explosive disorder involves recurrent, impulsive aggressive outbursts disproportionate to the situation. It typically begins in late childhood or adolescence and shows high rates of co-occurrence with anxiety disorders and depression, which are often missed because the anger presentation is more obvious.

Substance-related and addictive disorders

This is one of the largest chapters in the manual. It covers alcohol, tobacco, caffeine, cannabis, hallucinogens, inhalants, opioids, sedatives/hypnotics/anxiolytics, stimulants (cocaine, amphetamines), phencyclidine, and other substances. Gambling disorder appeared in this chapter for the first time in DSM-5, the only behavioural addiction included. Internet gaming disorder is listed in the appendix as a condition requiring further study.

The collapse of “abuse” and “dependence” into a single substance use disorder was a significant structural change. DSM-5 replaced two separate diagnoses with one, graded by severity: mild (two to three criteria met), moderate (four to five), and severe (six or more of eleven criteria). The eleven criteria cover four domains: impaired control, social impairment, risky use, and pharmacological criteria (tolerance and withdrawal).

The removal of “legal problems” as a criterion was worth noting. Whose behaviour results in legal consequences is as much a function of social circumstances and policing as it is of the disorder itself.

Neurocognitive disorders

Delirium, mild neurocognitive disorder (MCI), and major neurocognitive disorder (which replaced “dementia”) appear here. Subtypes are specified by aetiology: Alzheimer’s disease, vascular disease, Parkinson’s disease, traumatic brain injury, HIV, prion disease, Lewy body disease, frontotemporal lobar degeneration, and others.

Replacing “dementia” with “major neurocognitive disorder” was partly about clinical accuracy and partly about reducing stigma. The word dementia carried connotations that went beyond its clinical meaning in many languages, often implying madness or loss of personhood rather than a specific set of cognitive deficits.

The introduction of mild neurocognitive disorder acknowledged the clinically important territory between normal ageing and dementia. Many people with MCI never progress to major neurocognitive disorder. Some do. Identifying the condition early opens the possibility of intervention and informed planning.

Personality disorders

DSM-5 retains ten personality disorders divided into three clusters. Cluster A (odd/eccentric): paranoid, schizoid, schizotypal. Cluster B (dramatic/emotional/erratic): antisocial, borderline, narcissistic, histrionic. Cluster C (anxious/fearful): avoidant, dependent, obsessive-compulsive.

These diagnoses remain categorical in the main body of the manual, but DSM-5 introduced an alternative model in Section III (the appendix) called the Alternative Model for Personality Disorders (AMPD). This dimensional model assesses two broad domains: impairments in self and interpersonal functioning, and pathological personality traits. It is better supported by research and is gradually gaining clinical traction, particularly in specialist settings.

The comorbidity challenge with personality disorders is substantial. Borderline personality disorder co-occurs with depression, PTSD, and substance use disorders at very high rates. Narcissistic personality disorder often presents alongside depression, particularly when the person’s sense of status or superiority is threatened. Treating the presenting disorder without addressing the underlying personality structure tends to produce limited results.

Paraphilic disorders

Voyeuristic disorder, exhibitionistic disorder, frotteuristic disorder, sexual masochism disorder, sexual sadism disorder, paedophilic disorder, fetishistic disorder, and transvestic disorder are classified here. DSM-5 draws a careful distinction between a paraphilia (an atypical sexual interest, which is not in itself a disorder) and a paraphilic disorder (which causes personal distress or involves harm to others). This distinction has both clinical and legal implications.

DSM-5 and ICD-11: what is the difference?

Two major classification systems exist. The DSM is published by the American Psychiatric Association. The International Classification of Diseases (ICD), currently in its eleventh edition, is published by the World Health Organisation and covers all diseases and health conditions, not just mental disorders. In the UK and most countries outside North America, ICD is the system used for official purposes and insurance coding.

DSM-5 provides more detailed and specific diagnostic criteria, which makes it better suited to clinical research and training. ICD is broader in scope and more widely used for population health statistics. The two systems have been brought into closer alignment over successive editions, but meaningful differences remain. ICD-11, for instance, includes Complex PTSD as a distinct diagnosis. DSM-5 does not. Whether Complex PTSD deserves its own category or falls within existing diagnoses is an active area of debate.

Serious criticisms of DSM-5

Any serious engagement with this manual requires engaging with the criticisms levelled at it.

Thomas Insel, during his tenure as director of the National Institute of Mental Health, wrote in 2013 that DSM lacks validity because its diagnoses are based on symptom clusters rather than biology. His analogy was apt: grouping together a cough, a sore throat, and a fever and calling it a syndrome tells you nothing about whether the cause is bacterial, viral, or allergic. Symptom-based taxonomy may improve reliability (different clinicians agreeing on a label) without improving validity (the label actually corresponding to a coherent biological entity).

Allen Frances, who chaired the DSM-IV task force, wrote at length about his concerns that DSM-5 had lowered diagnostic thresholds in ways that pathologise normal human experience. His worry was most pointed around ADHD, autism, and bipolar disorder in children. When clinical criteria become too broad, genuine need gets diluted, resources are spread more thinly, and people who do not need treatment receive it anyway.

The relationship between the DSM process and the pharmaceutical industry has been a persistent source of concern. Research examining the financial conflicts of interest among DSM-5 committee members found that a significant proportion had ties to drug manufacturers. This does not amount to corruption, but it creates conditions where those ties can influence, consciously or not, which diagnoses are created and how broadly they are defined.

Cultural bias is another real limitation. DSM-5 rests primarily on research conducted in Western, and particularly North American, populations. Mental distress is expressed differently across cultures. DSM-5 includes a section on cultural concepts of distress, which is more than its predecessors offered, but critics argue the integration of cultural considerations into the main diagnostic criteria remains insufficient.

None of this means DSM-5 should be abandoned. There is no better alternative. Its limitations are documented, debated, and actively worked on. That transparency is itself a scientific virtue.

How DSM-5 is used in clinical practice

A common misconception is that DSM-5 tells clinicians how to treat. It does not. It identifies what pattern of symptoms is present. Treatment selection requires additional resources, including empirically supported treatment protocols, clinical training, and professional judgement.

The diagnostic process using DSM-5 involves several steps: a structured or semi-structured clinical interview to gather comprehensive information; comparison of reported symptoms against DSM-5 criteria; ruling out alternative explanations (medical conditions, substance effects, other disorders); assessing comorbidity, since most patients meet criteria for more than one disorder; and applying clinical judgement that goes beyond ticking boxes.

Comorbidity: the rule, not the exception

In real clinical settings, comorbidity is the norm. Somewhere between 50 and 60 per cent of people with major depressive disorder also have an anxiety disorder. PTSD rarely presents alone. Personality disorders frequently co-occur with what DSM-IV used to call Axis I conditions. DSM-5 allows and encourages multiple simultaneous diagnoses, which better reflects what clinicians actually encounter.

The person behind the criteria

Every clinical training programme worth its name emphasises this point: a DSM diagnosis is a starting point, not a destination. The criteria tell you something about pattern. They tell you nothing about the person’s history, their strengths, the context in which the disorder developed, or what they most need from treatment. Reducing someone to their diagnostic label is not good clinical practice. It is not what DSM-5 asks clinicians to do.

Why DSM-5 remains the standard

Despite everything, DSM-5 is the gold standard for psychiatric diagnosis in international research, and that matters because science requires a common language. A trial in Tokyo studying major depressive disorder needs to be using the same definition as one in Amsterdam, otherwise the results cannot be compared and the evidence base cannot build.

Every medication currently in use, every psychotherapy with an evidence base, every advance in understanding the neuroscience of mental illness, has been built on research that used shared diagnostic criteria. The manual made that research possible.

DSM-5 does not claim to be final. Its own preface acknowledges that the classifications are provisional and will change as science advances. That scientific honesty, the willingness to sit with imperfect categories while working towards better ones, is not a weakness. It is what good science looks like.

For anyone working in mental health, or living with a mental health condition, or trying to understand someone who is, DSM-5 is an essential tool. Worth knowing well. Worth knowing critically. And worth treating as what it is: a sophisticated but imperfect map of territory that remains only partly understood.

Frequently asked questions about DSM-5

Is DSM-5 a treatment guide? No. DSM-5 contains diagnostic criteria, not treatment recommendations. For evidence-based treatment protocols, separate clinical guidelines are required. DSM-5 tells you what you are dealing with. How to address it is worked out elsewhere.

Can someone use DSM-5 to diagnose themselves? The diagnostic criteria are publicly available, but self-diagnosis without clinical training is unreliable. Many disorders share overlapping symptoms, and determining whether impairment meets clinical significance requires experience and context that a checklist cannot replicate. Self-assessment can be a useful starting point for seeking help, but it should not substitute for professional evaluation.

How many disorders does DSM-5 contain? Around 157 main diagnoses and approximately 600 diagnostic codes. A single disorder can generate multiple codes depending on severity specifiers, subtypes, and whether the condition is primary or substance-induced.

How does DSM-5-TR differ from DSM-5? DSM-5-TR, published in 2022, updated descriptive text, revised epidemiological data, and made corrections to several criteria sets. It did not restructure the diagnostic system. One new diagnosis, Prolonged Grief Disorder, was added. It is an update rather than a new edition.

When will DSM-6 be published? The American Psychiatric Association has not announced a timeline for DSM-6. Preparatory work for new editions typically begins a decade or more before publication. DSM-5-TR may function as a bridge while that longer process develops.

Are mental disorders real or socially constructed? Both elements are present. The suffering involved in mental disorders is real. The impairment is real. The neurobiological correlates being studied are real. The diagnostic categories used to organise those experiences, however, are human constructs that have changed over time and will continue to change. The DSM is not discovering natural kinds in the way a biologist discovers a new species. It is creating clinically useful categories from complex, continuous human experience. That is not a reason to distrust it. It is a reason to understand what kind of tool it is.