This chapter seeks to examine and further clarify the relationship between two main conceptual approaches to causality in psychiatric research: counterfactual and mechanistic. The author suggests that psychiatry may pose some relatively unique challenges for causal inference not shared with other branches of medicine. At their core, counterfactual approaches ask “what if” questions while mechanistic approaches ask “how” questions. The chapter also seeks to evaluate the Russo-Williamson Thesis for psychiatric research, which argues that causal inference requires evidence of counterfactual and mechanistic causal effects, by examining three research papers that examine causal effects on psychiatric disorders. Two of these are from epidemiological samples and employ counterfactual methods, and one is from molecular genetics and molecular neuroscience and utilizes a mechanistic approach. Kendler argues that these two methods are complementary and often mutually reinforcing. In particular, the demonstration of counterfactual evidence of causation naturally raises the question of how such an effect occurs at a mechanistic level. However, the author suggest that the Russo-Williamson Thesis is too high a threshold for psychiatry and that, in at least some cases, high-quality counterfactual evidence can be actionable.

In his chapter, Kenneth Kendler succinctly explains some difficult-to-understand philosophical concepts and then shows how they can be used to illuminate psychiatric research and practice. The concepts he explains are those relating to causality, specifically the contrast between counterfactual and mechanistic approaches.

I have, for many decades, read widely and deeply within the rich corpus of descriptive psychopathology. It has enriched my appreciation of the variations of the syndromes in psychiatry that we diagnose, study, and attempt to treat. It has helped me develop interview assessments used in a wide variety of psychiatric genetic studies, from family to twin, high-density pedigree, and now Genome-Wide Association Study case–control studies. I have taught two generations of psychiatric residents on the importance of this tradition and our need to know our rich descriptive inheritance. My teachings have been met with substantial (but not universal) appreciation. I have served on various DSM committees from DSM-III-R onward and tried to use my psychopathological expertise to improve our nosology, in part by increasing the empirical rigor of the evaluation for change using the concept of validators.

The concept of ‘stereotypes’ refers to generalisations that are made about the behaviour adopted and/or the characteristics possessed by the members of a particular group. Involving presumptions about human actions and attributes, a stereotype provides ready-made narratives as to how and why some events unfold as they do. Thus, stereotypes, especially when they operate ‘undetected’, hamper an objective analysis of the factual situation. In the courtroom, they tend to have a polluting effect on the assessment of evidence, leading to relevant pieces of evidence being ignored, irrelevant circumstances being given weight, and higher standards of proof being imposed than would have been the case in their absence. This chapter focuses on the approach of the CEDAW Committee in examining the impact of gender stereotypes on the evaluation of evidence performed by domestic courts. It provides an in-depth analysis of the views adopted by the Committee in selected individual communications.

In their chapter, Josef Parnas and Maja Zandersen critically examine the etiological claim that schizophrenia is a brain disease. They are not arguing against a claim that it is a confirmed brain disease, but rather against some implications that follow from the inference that schizophrenia is a disease even though the biopathology is not yet mapped out. To focus the discussion, Parnas and Zandersen describe what they consider a folk psychological definition of disease in medicine. A disease is an unwanted affliction caused by a biopathological process. It has a characteristic pattern of onset, time course, and outcome – including the potential death of the patient. As unwanted afflictions, diseases are experienced as deserving of treatment.

I recall the enthusiasm with which magnetic resonance imaging was greeted by the field of psychiatry in the 1980s, the stunning quality of the images, almost magical. This technology fed into the long desire by our field to have an innovative technology that would, in one fell swoop, solve the main problems of our field, particularly etiology and diagnosis. I recall claims such has “Now we can do neuropathology on living people,” “Now we have a tool to solve the mind–body problem,” and “It is now just a matter of time until we find the causes of schizophrenia.” A decade or more earlier it had been monoamine neurotransmitters (serotonin, norepinephrine, and dopamine) as the revolutionary scientific breakthrough, and one or two decades later it would be molecular genetics. In each case, the “hype cycle” (Dedehayir & Steinert, 2016) produced a crest of zealous excitement (and dramatic overpromising) followed by a period of disillusionment and then some stabilization and emerging maturity. The field came to recognize that the application had merits but the problems – very hard ones – of psychiatric diagnosis and etiology were not going to yield easily. Of note, the field of psychiatric neuroimaging, although now 40-plus years old, underwent another recent round of substantial criticism about the use of underpowered sample sizes (Marek et al., 2022) and inadequate corrections for multiple testing.

The network approach to psychopathology has been introduced in psychometrics to offer an alternative to the dominant approach that sees mental disorders as caused by underlying mechanisms or pathogenic pathways. Instead, the network approach conceives of mental disorders as causally connected symptoms, locked in stable, dysfunctional states. The authors of this chapter aim to chart the explanatory potential of network models in psychopathology. They argue for the following (related) points. Network models (1) afford explaining the covariation between two or more symptoms as being part of a network and (2) should be conceived as heuristic devices for hypothesis testing in individual therapeutic settings. (3) This heuristic perspective allows for two types of inferences – from a generic model to specific individuals and from specific individuals to a generic model – making system boundaries of networks of causally related symptoms and of underlying mechanisms fluid, and (4) thus showing that a rigid understanding of constitutive relevance is not viable. The presented analysis is informed philosophically by two approaches: a mechanistic property cluster-understanding of mental disorders, and an understanding of mechanistic modeling as a toolbox for hypothesis generation. The authors illustrate their claims through an empirical study on panic disorder.

The development of new therapeutic methods for intervening in brain function has the potential to influence aspects of a person’s agency, including their autonomy and authenticity. To determine whether or not this is the case, we need an assay that can measure aspects of agency before and after neurointervention. This chapter introduces a framework for thinking about agency as a multidimensional construct, and describes a research project aimed at measuring agency. The authors develop an Agency Assessment Tool (AAT) using survey methods and present preliminary results from the project that shows that the AAT can characterize control and patient groups along several dimensions with a factor analysis; the results show some interesting patterns among putative conceptual dimensions with network analysis. Although the data do not go so far as to include interventions to help understand causal relations, the authors discuss the prospects for using data-driven tools to come up with ontologies of agency.

In a thoughtful essay that follows, Peter Zachar interweaves several philosophical issues that relate to classification in general and psychiatric classifications specifically. I think it will be most useful to interested readers if, in this introduction, I steer away from a largely conceptual/philosophical critique of Zachar’s essay and rather comment from the perspective of a working clinician and psychiatric nosologist who has, for better or worse, spend several decades working on various versions of the DSM – the accepted nosology for American psychiatry.

The Prologue gives examples of various kinds of hiddenness, focusing especially on the way the artist or the author is both indicated by and hidden behind a work.

Introduction. The Introduction sets out the aims of the book. No attempt has previously been made to investigate the detailed operations of any major figure as evidenced by his or her day-to-day practice. This is an effort to fill that gap by exploring the career of Edmund Curll, and to go beyond the mainly biographic treatment he has received so far. It describes how the succeeding pages seek to consider various aspects of Curll’s publications in the context of the London publishing industry as it then operated. An account is given of the content of various chapters, as this will be given in the abstracts below. Among topics briefly outlined are the range of Curll’s lists, his dealings with the law, his connections with contemporary authors (major and minor), with a review the most important scholarship on the condition of publishers in the eighteenth century.