The third edition of Marijuana and Madness has over twice as many chapters as the first edition, which was published 20 years ago. This field of research is still very active and increasingly topical. Much has changed over the last two decades. Access to cannabis has been decriminalised in many jurisdictions. These changes have been made against the backdrop of a growing and convergent body of evidence about the association between cannabis use and adverse mental health outcomes. The new edition of this book maps the evidence in a comprehensive fashion.

Carl Bernstein, the Watergate journalist, coined the phrase ‘the best obtainable version of the truth’ to describe the responsibility of the media. Science also aims to provide the best obtainable evidence to support or reject hypotheses. Sometimes, the available evidence is scant. Sometimes, the body of evidence is inconsistent. However, sometimes, the data are convergent, which allows the research community to make decisions based on the balance of probability. Rarely, science can upgrade the quality of the evidence to beyond reasonable doubt.

In their chapter, Hall and Degenhardt make a compelling case for prudence with respect to public health recommendations. They argue the need to balance the potential benefits and harms from policy decisions about restricting access to cannabis. On balance, they state ‘[t]here is arguably an ethical imperative to inform young people of the probable mental health risks of cannabis use’ (p. 54). This evidence is also mapped out concisely by Power et al in Chapter 17.

The book provides compelling evidence that the potency of cannabis-related products has increased in many countries, and that synthetic forms of cannabis-related molecules pose a challenge with respect to potency and adverse events. These topics warrant ongoing scrutiny. Clues from genetics have also been informative in recent years – genetics variants associated with cannabis use/misuse are shared with the genetic correlates of psychosis and appear bidirectional. This pattern of finding is certainly thought-provoking. Cobert and Johnson (Chapter 21) remind the reader of an underappreciated feature of genome-wide association studies (GWAS). If an exposure (e.g. cannabis use) is a true risk factor for schizophrenia, then as the GWAS sample size for schizophrenia increases, variants associated with cannabis use will be blindly and automatically included in genome-wide significant hits associated with schizophrenia. Thus, GWAS studies can provide a Rosetta Stone to identify potentially modifiable risk factors (such as cannabis use), if we were only better at deciphering which variants impacted directly versus indirectly (i.e. via environmentally mediated exposures) on phenotypes such as schizophrenia.

There are excellent chapters that summarize the (often mixed) evidence from animal models, and post-mortem brain studies. The take-home message: the neurobiology of cannabis-related neuromodulation is complicated. However, progress has been made since the first edition. There are practical chapters about the management of people with cannabis use and misuse. There are also several new chapters that describe emerging topics related to cannabis (e.g. ‘medical use’, links with being a victim or perpetrator of violence, prenatal exposure and offspring outcomes, etc.).

D'Souza, Castle and Murray have edited an information-rich third edition of Marijuana and Madness (a very catchy title). They have outlined gaps in the evidence that need more research and summarised the ‘best obtainable version of the truth’.