---
title: "Chapter 1. Kinds That Hold"
book: "Why Minds Break: An Evolutionary and Developmental Programme for Psychopathology"
author: Ian Pitchford
url: https://evidencepress.org/why-minds-break/chapter-1/
---

# Chapter 1. Kinds That Hold

Asks what a psychiatric diagnosis must achieve to support scientific inference, and proposes keeping clinical categories apart from an evidence-graded record of proposed explanations.

The chapter opens in April 2013, when Thomas Insel, then head of the US National Institute of Mental Health, faulted the forthcoming fifth edition of the Diagnostic and Statistical Manual for resting on agreed symptom descriptions, and pointed research towards the Research Domain Criteria instead. A joint statement with the American Psychiatric Association soon followed, and the chapter finds the two positions compatible. Its guiding distinction is between reliability, meaning that clinicians agree on a diagnosis, and validity, meaning that the diagnosis picks out what it claims to.

It then traces how agreement became the goal. After studies in which American and British psychiatrists labelled the same filmed patients differently, Robert Spitzer's team rebuilt the 1980 manual around checklists of criteria. Early field trials reported respectable agreement under favourable conditions, while trials for the fifth edition gave mixed results, with low figures for major depression and generalised anxiety, and later work showed how much such figures depend on method. Diagnosis is not arbitrary, though: a first diagnosis of schizophrenia usually persists. It follows the serotonin account of depression from guarded hypothesis to popular belief, describes Joanna Moncrieff's 2022 umbrella review and the unresolved dispute it provoked, and keeps that question apart from trial evidence that antidepressants, on average, outperform placebo.

To say what a sound category requires, the chapter draws on Richard Boyd's homeostatic property clusters and on the mechanistic version developed by Kenneth Kendler, Peter Zachar and Carl Craver. A category need not have an essence, but causal processes should hold its features together, and whether a given diagnosis meets that standard is a question for evidence. Survey, register and cohort data show diagnoses often accumulating in the same people, and genetic studies find shared influences across schizophrenia, bipolar disorder and depression. Explanations must account for this overlap, though the overlap does not choose among them. HiTOP, Denny Borsboom's network theory and the general factor of psychopathology are each judged by the task in hand, not ranked above diagnoses by default.

Following Ian Hacking, the chapter notes that people respond to being classified, which complicates testing any category. It weighs contested accounts of dissociative identity disorder, tic-like presentations reported during the pandemic and Nick Haslam's idea of concept creep. The 1973 removal of homosexuality from the manual shows that explaining a variation is a different matter from showing it to be a disorder.

The chapter closes by proposing two linked classifications: clinical descriptions that serve care, and a research registry recording proposed explanations alongside the evidence for each. A claim earns mechanistic status only when it states the process, level and conditions involved and has evidence that sets it apart from rival accounts.

## Questions the chapter asks

- If psychiatrists can agree on a diagnosis, does that show the diagnosis is real?
- Was the idea that depression is a chemical imbalance ever supported by the evidence?
- What should be done about people who meet the criteria for several disorders at once?

This is a summary of the chapter, written for this site; it is not an extract from the book. Coming soon from Amazon in Kindle, paperback and hardback editions.

*Why Minds Break* and these summaries are © 2026 Ian Pitchford, all rights reserved. The cover drawing is by Santiago Ramón y Cajal (1915), public domain.