Therapy and Its Effect

Does it help anyone? It is the question this archive runs toward, and it is younger than the field likes to admit. Psychoanalysis treated for half a century without counting its successes; its form of proof was the interpreted single case, and whoever asked for control groups was held to be naive. That treatment must let itself be measured like a drug is an achievement of the century’s second half — and three folders of this archive tell how it came about.

The first step was taken by humanistic psychology, and it was technical in nature: Carl Rogers had therapy sessions recorded and transcribed — the treatment became a dataset. His conditions — warmth, genuineness, close listening — proved necessary and not sufficient; and from the comparisons of procedures with one another comes the dodo verdict: properly conducted psychotherapies differ astonishingly little in their effect.

The second step Beck took with tools. A questionnaire that gave depression a number; a manual that made the treatment reproducible; and in 1977 the trial in which a talking treatment held up for the first time, randomised, against the standard drug. That the first study was too good and the large comparative trial came out soberer belongs to the pattern: the effect is there, it is meaningful, and it is smaller than the early figures promised. Beck’s school co-published the corrections itself — the rarest virtue in this collection.

Ellis was there earlier and left the testing to others; Frankl brought meaning into treatment and left the measuring of his school to successors; Seligman built a model of depression out of the animal laboratory, reversed it himself fifty years later — and founded with positive psychology a movement whose interventions lost considerable ground in the replication crisis. The span of this theme reaches from the manual to the bestseller, and the balance sheets differ accordingly.

What the large numbers show is sobering and consoling at once. Psychotherapy works — in hundreds of randomised trials, with medium effects, beyond the end of treatment. Why it works is open: behavioural activation alone achieves in several studies about as much as the whole cognitive package, and the lead of individual schools largely vanishes in clean comparisons. And under everyday conditions — the British routine-care programme has measured it at scale — the results stay below those of the studies from which they were derived.

For the archive the circle closes here: what Freud began in the consulting room is today counted, compared and corrected. The question “does it help?” has ceased to be a question of schools — it has become a question of measurement, and that is the progress this theme documents.