Psychiatry, Insane Asylums, Mental Illness, ECT, Lobotomies, Freud & Jung | Lex Fridman Podcast #502

The evidence supports only partial symptomatic control of mental illness, not cure, and its diagnoses are symptom-based rather than pathological [04:10]. Historical abuses and modern trial data show efficacy claims often outrun evidence [50:51].

• DSM-III's 1980 symptom-counting system improved reliability, but reliability did not prove that labels like schizophrenia or major depression map to discrete diseases [06:41]. NIMH directors pushed genetics and neuroscience research, but after spending more than $20 billion, Thomas Insel said the lot of mentally ill patients had not improved [13:18]. DSM-5 therefore kept the same symptom-based approach instead of replacing it with pathological criteria [20:52].

• Asylum cure claims of 60-80% overstate outcomes: annual discharges were 35-40%, leaving chronic patients and creating a “degeneration” narrative used to justify confinement [38:43]. U.S. eugenic sterilizations exceeded 60,000 by the 1960s, with California disproportionately involved [41:32]. American eugenic thinking influenced Nazi T4, which killed up to 250,000 mentally ill people [43:29]. Today, public policy abandonment means people with serious mental illness die 15-25 years earlier, and the largest U.S. centers for psychiatric care are jails [28:12].

• Malaria therapy for general paralysis of the insane won the 1937 Nobel Prize for causing fevers, but it died after penicillin became available, not after controlled efficacy showed it worked [1:03:43]. Insulin coma therapy, used for schizophrenia from 1933, involved up to 60 comas, carried a 1-5% mortality rate, caused brain damage and obesity, and only stopped after 1950s controlled trials failed it [1:19:26]. Lobotomy extended from 1930s to 1970s, with Walter Freeman performing 20-30 operations in an afternoon, and became discredited partly through visible “zombie” failures and antipsychotics [54:20].

• Clinicians cannot predict who will respond to antipsychotics or SSRIs [05:51]. CATIE found newer antipsychotics not better than older drugs, and 67-82% of patients stopped because of lack of effect or side effects [3:09:52]. Antipsychotics mainly reduce positive symptoms, not negative symptoms, and tardive dyskinesia was a major first-generation harm [3:04:12]. SSRIs usually beat placebo statistically but often by only 1-2 points on 60-point scales, while 40%+ of depressed patients do not respond [3:25:23].

• Cochrane reviews rate CBT low-to-medium confidence [2:47:56]. NICE nevertheless recommends CBT as first-line for mild depression, where it can outperform drugs without some side effects [2:48:23]. ECT remains contentious: modern trials support benefit in severe/treatment-resistant depression, but it is not a cure, often requires maintenance, can impair memory, and is legally constrained for involuntary patients [1:30:06]. Psychoanalysis became a niche product accessible mainly to the wealthy [3:28:14].