Reconsidering the Relationship Between Criminal Insanity and Delusions

Criminal Insanity and Delusions

The criminal insanity doctrine demarcates lack of criminal responsibility and inapplicability of punishment. It is commonly considered to identify those who cannot reasonably be blamed and should be excused for their otherwise criminal actions because, at the time of the act, they were not deemed capable of practical rationality.^13,–,16 More specifically, criminal insanity is commonly understood in terms of impairments in the capacities of recognizing and responding to reasons for actions provided by legal (and moral) norms.^14,17,18 This doctrine mirrors the criminal law’s premise that a responsible person is able to understand and respect the law and act accordingly.^19,20

As mentioned above, current insanity doctrines commonly require that the defendant is experiencing a serious mental disorder at the time of the act, but the criteria for the legal relevance of mental disorders differ between countries.¹ At the same time, it is commonly recognized that having a diagnosis, or a symptom, of a severe mental disorder is neither sufficient nor necessary for criminal insanity. What principally matters to criminal insanity is certain functional impairments that are considered to negate capacities needed for criminal accountability.¹⁹ Because psychiatric notions of delusions are legally related to such impairments, we need to address these notions before we proceed to the legal questions.

To describe the psychiatric notions of delusions is, however, not straightforward. Despite the descriptions of delusions given in diagnostic manuals of mental disorders, there is currently not a unified definition of this phenomenon within the psychiatric discourse.²¹ To some extent, different diagnostic manuals are also applying different criteria to account for delusional disorder.²² In addition, there are different approaches to the understanding of mental disorders, from biological to phenomenological perspectives, as well as to the classification of mental disorders.^23,24

Within clinical psychiatry, the writings of the psychiatrists Kraepelin and Bleuler from the late 19th and early 20th centuries have been of central importance to the classification of mental disorders, such as schizophrenia (a term coined by Bleuler, first described as dementia praecox by Kraepelin). They did not directly define delusions but described them in their classifications. For example, they described typical delusional symptoms in persons with schizophrenia as believing that they were spied upon, that the atmosphere was poisoned (Kraepelin), or that their brain was being sawn to pieces (Bleuler).²⁵ Later, the influential philosopher and psychiatrist Karl A. Jaspers proposed three criteria to define delusions: false (or impossible) ideas that are uncorrectable and held with extraordinary certainty.^26,27 Today, the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) defines delusions in a similar way as “fixed beliefs that are not amenable to change in light of conflicting evidence” (Ref. 10, p 87), a definition that has generated broad discussions and been criticized for being too narrow.²⁸ Still, it emphasizes an important aspect for our further discussion; that delusions involve, at least to some degree, an inability to form alternative beliefs. This inability, in turn, may constitute the link to the legal notion of criminal insanity in so far as it may hamper the ability to recognize and respond to legal norms.

It is important to recognize the distinction between the content and form of the delusions.²⁵ The delusional content can be of persecutory, referential (e.g., one is not controlling one’s own actions), somatic, religious, or grandiose character. The content of the delusions is understood to be dependent on the experiences and context of the person, and delusions may therefore sometimes be difficult to distinguish from strongly held ideas.^29,30 Delusions can also be bizarre, incomprehensible, or nonsensical, i.e., the content is clearly implausible.^27,31

The form of a delusion, on the other hand, denotes its structural components. In this context, the origin of the delusion has been considered of great importance. There are different accounts on this matter.²⁸ According to early writings by Schneider and later Jasper, delusions may arise from a real perception followed by a delusional interpretation of that perception, from a delusional interpretation of a real memory, or from the patient’s mood (Wahnstimmung), or they may be autochthonous (i.e., they arise spontaneously).^32,33 Other perspectives held today are that delusions arise from a reasoning bias, i.e., the person bases decisions on insufficient information,³⁴ or by a reduced ability to understand other people’s mental states and intentions.³⁵ From a mechanistic point of view, delusions have been suggested to be caused by errors in the interplay between prior beliefs and “correction” signals.³⁶ According to this view, the brain generates internal models of the world that are continuously being updated based on sensory information that confirms or contradicts them. Disrupted ability to adjust these predictions to the incoming sensory information may lead to the formation of delusional beliefs.^37,38 Taken together, these different views suggest that the formation and persistence of delusions may be correlated with functional impairments in different cognitive domains independent of the specific content of the delusion. Moreover, disregarding the distinction between content and form, delusions may be related to the presence of other psychopathologies, such as hallucinations, self-disturbances, anxiety, depression, or mania.³⁹ The profile of concurrent symptoms may also affect the form, content, and severity of the delusions.

Finally, it is important to note that delusions can vary in intensity over time.⁴⁰ That is, they may be held with stronger conviction and affect behavior in some periods but be less pronounced and debilitating in other periods. Thus, the functional impairments associated with delusions and their effect upon persons show great heterogeneity, not only between different individuals but also over time for each individual experiencing delusions.

The Anglo-American Approach

Within the Anglo-American paradigm, rules and doctrines of criminal insanity are built on a mixed model that attributes a weak relevance to mental disorder as an excusing condition.² This means that, in addition to the presence of a specified condition of mental disorder, it is required that this condition resulted in certain functional impairments that influenced the commission of the crime. In current criminal insanity doctrines, the latter typically concerns the lack of ability to understand and, sometimes, to control one’s actions.¹ Although there are different legal accounts of the relevance of delusions in different jurisdictions, they all entail the requirement that the delusion produced, or exemplified, functional impairments relevant to the commission of the crime. Within Anglo-American law, delusions are thus considered as proof of a link between the defendant’s disorder and the committed crime(s). It is not sufficient to have delusions; the defendant’s delusional state must reach a certain level of severity according to legal standards for criminal insanity. Still, this construction with a causality requirement has led to a focus on how the defendant’s delusional beliefs formed motives for actions, i.e., the content of the delusions comes to the forefront.^7,11,41,42

When and why someone who is acting from delusional motives should be understood as criminally insane is, however, debated. The controversies regarding the specific relevance of delusions have revolved around legal criteria regarding the defendant’s understanding of the nature and quality of the act and of the defendant’s understanding of the wrongfulness of the act. These criteria are found in many jurisdictions, such as in England and Wales, Scotland, South Australia, New Zealand, many states in the United States, and Canada, although with somewhat different formulations and interpretations with implications for the legal relevance of delusions.¹

According to Scottish law, to give an example, delusions may be relevant to the defendant’s understanding of the act. What matters is the defendant’s understanding of the quality of the act, which is understood to include situations where the defendant’s delusional beliefs were providing reasons for committing the crime. Scottish law thus recognizes that delusional states often do not negate defendants’ intent to commit wrongful acts but provide reasons to form this intent.⁴³ In English law, as a contrast, the criterion of understanding of the action has been interpreted in a more restrictive sense, referring only to the physical character of the act and not to its moral or legal quality.⁴⁴ As a result, delusions are basically not relevant to the defendant’s understanding of the act, not even when the delusions may negate mens rea.

The most complicated, and hence debated, factor regarding the relevance of delusions concerns how delusions affect defendants’ understanding of the wrongfulness of their actions. In so far as defendants’ delusional motives render them incapable of understanding that their actions were wrong, they may in many jurisdictions be exempted from punishment. The discussion about the proper understanding of “wrong” in relation to criminal insanity has, in this regard, generated a vast amount of literature.^12,45,–,51 It has been suggested that wrong should denote either the legal or the moral wrongfulness of the act or even both. According to the M’Naughten rules as interpreted in English law, wrong denotes unlawful. Only in so far as defendants did not understand that their actions were against the law can they be acquitted by reason of criminal insanity.⁵² This means that a defendant can be in a psychotic and delusional state of mind that influenced the act but still not qualify for being criminally insane.^53,–,55 This narrow interpretation of wrong was established in England through the Windle case in 1952⁵⁶ and is also explicit in the model penal code, where it is stated that a person who “as a result of mental disease or defect…lacks substantial capacity either to appreciate the criminality [wrongfulness] of his conduct or to conform his conduct to the requirements of law.”⁵⁷ At the same time, the reasoning in the Windle case and the restrictive interpretation of wrong have been criticized.⁵⁸ For example, in 1975, the U.K. Butler Committee observed that it is a very narrow ground of exemption “since even persons who are grossly disturbed generally know that murder and arson are crimes” (Ref. 59, p 218). Moreover, many other countries that base their insanity defense on the M’Naughten rules have chosen a standard that denotes moral wrongfulness. The High Court of Australia in Stapleton v. The Queen⁶⁰ already in 1952 expressly declined to follow Windle and instead concluded that the decisive question was whether defendants believed their act to be right according to the ordinary standard of reasonable people.⁶⁰ In Canada, the Supreme Court initially followed Windle, but in 1990, in R v. Chaulk,⁶¹ the standard was reversed to that of wrong according to the ordinary moral standards of reasonable members of society. New Zealand has explicitly implemented such an interpretation of wrong in its Crimes Act.⁶²

As the emphasis has thus turned to morally wrong in these jurisdictions, recent developments in English law have seemingly led to an even more restrictive standard for criminal insanity than before. According to the Keal case⁶³ from 2022, wrong has been construed to mean that defendants must prove that they did not know their act was legally and morally wrong to succeed with the insanity defense. This means that defendants who delusionally believed that their acts were in accordance with the law but knew that they were morally wrong cannot be acquitted by reason of insanity.⁴²

To give a feasible account of what it would mean not to know that the act is morally wrong, several scholars have modeled their accounts on what typically can constitute a justification for doing something that is under normal circumstances considered wrong. One reason to believe that hurting another human being is not wrong is that it is somehow justified. This line of reasoning is, for instance, expressed by Stephen J. Morse who explains:A person who delusionally believed that she was about to be killed by another person, and kills the other in the mistaken belief that she must do so to save her own life, does not rationally understand what she is doing. She of course knows that she is killing a human being and does so intentionally, but the rule against unjustifiable homicide will be ineffective because she delusionally believes that her action is justifiable. (Ref. 64, p 152)

Morse is referring to a legally valid justification, but if we are to interpret wrong as morally wrong, we must look for further moral justifications for committing an otherwise wrongful act, which is more challenging. To take the U.S. case of Andrea Yates⁶⁵ as an example, the defense expert opined that Ms. Yates delusionally believed that, if she did not kill her children, they would suffer a far worse destiny. Her understanding of the murder was hence altruistic.⁵⁰ This suggestion was controversial for several reasons, including that whether an altruistic motive morally justifies murder is dependent on what normative theory one endorses.

In sum, there are many controversies within the Anglo-American paradigm about how and why delusions are relevant to criminal insanity. They all have in common an emphasis on how the content of the defendants’ delusional beliefs relate to the specific crime. In our view, these controversies reveal profound challenges in specifying and applying legal standards for criminal insanity in a way that considers the complex manifestation of delusions. Delusions may be associated with functional impairments in different cognitive domains, and the extent to which a person’s general functional ability is affected by the delusions varies. Furthermore, because delusions also have a form, and because their content is influenced by experience, personality traits, and context, we cannot easily determine from their content how incapacitated a person experiencing delusions is to assess the veracity of the belief. In such a perspective, an emphasis on the content of the delusions may even result in unequal treatment before the law of equally mentally disordered and functionally impaired defendants.

To this point, the Norwegian approach brings different perspectives. Within Norwegian law, all that matters for insanity is if the defendant was impaired enough while committing the crime. This may, as we next discuss, provide an alternative way of linking criminal insanity to delusions.

The Alternative Approach of Norwegian Law

The Norwegian insanity doctrine differs from the insanity doctrines of most countries in so far as it does not operate with any requirement of a causal link between mental disorder and crime. The decisive question in the court’s evaluation is how mentally disordered the defendant was at the time of the act, disregarding if and how the defendant’s condition influenced the commission of the crime.⁶⁶ In this way, insanity is associated with a sufficiently severe state of mental disorder and, as in other countries, criminal insanity is here especially associated with psychosis.² This medical model approach, which has a long tradition in Norwegian law, became known to the wider international audience through the case concerning Anders Behring Breivik, who killed 77 people in Oslo and at Utøya on July 22, 2011.⁶⁷ At that time, the Norwegian Penal Code Section 20 letter b defined criminal insanity in terms of being psychotic as a matter of having significantly impaired reality understanding (“The offender is not criminally responsible if, at the time of committing the act, he or she is…psychotic” (Ref. 68, Section 20 letter b, in the wording before January 10, 2020)). In the aftermath of the Breivik case, however, a law reform process was initiated, and a new rule in the Penal Code section came into force October 1, 2020.^68,69 The psychosis criterion has now been replaced by criteria requiring that the defendant was “unaccountable due to” a “severely deviant state of mind” at the time of the act (Ref. 68, Section 20, second paragraph letter a). Now, the Norwegian rule requires a two-step evaluation.

The first step is to assess whether the defendant’s mental disorder at the time of the act qualifies as “a severely deviant state of mind” (Ref. 68, Section 20, second paragraph letter a) rather than referring to being “psychotic” as in the prior version. This criterion, in the same way as the previous one, primarily targets psychotic states of impaired reality understanding and requires obvious symptoms, such as delusions, at the time of the act. Other psychotic pathologies, such as hallucinations and disorganized thought process, are also relevant. In addition, the category includes other conditions equally serious as a psychotic state, measured in relation to how they affect the functional abilities and reality understanding of the defendant. In the bill from the Ministry of Justice, severe autism disorders and dementia conditions are mentioned as possible candidates (Ref. 70, p 76, 229) whereas personality disorders are excluded (Ref. 70, p 13, 67, 76). Except for personality disorders, a severely deviant state of mind can be the result of many different psychiatric and somatic disorders, such as brain tumors or infectious diseases, and the etiology of the condition is irrelevant. The focus on failures in functional abilities and reality understanding also allows for a severely deviant state of mind to be the result not only of one but of several coexisting disorders or dysfunctions.

If the first criterion is met, the second evaluation is whether the defendant was unaccountable because of this condition. In this evaluation, emphasis is given to the “degree of the person’s failure in appreciation of reality and functional capacity” (Ref. 68, Section 20, third paragraph). Still, there should be no evaluation of how the defendant’s disorder influenced the crime. The required evaluation of the defendant’s deficient appreciation of reality and the defendant’s functional capacity only concern the severity of the disorder at the time of the act considering the symptom load (Ref. 70, p 69). To illustrate, if the original M’Naughten case were adjudicated under current Norwegian law, the assessment would focus on whether he had a mental disorder at the time of the act that was sufficiently severe considering the impairments in cognitive functional abilities and reality understanding. In this assessment, his delusional beliefs about the criminal act would have been relevant only as proof to this matter in an overall assessment of his symptom load. Conversely, if the Breivik case were judged in England, the discussion of his alleged delusions would not only concern the presence and severity of his state of mental disorder at the time of the act but also whether his possibly delusional mental state influenced his understanding of his actions.

The Norwegian legislature has, however, provided the courts with a significant scope for discretion to ensure a reasonable and fair demarcation of criminal incapacity (Ref. 70, p 13, 69). Except for the almost unnecessary clarification from the Supreme Court that it is cognitive mental impairments that matter to criminal insanity,⁷¹ there is, to date, little legal clarification about the relevant impairments in reality understanding and functional capacity.⁶⁵ The Ministry’s bill explains that a significant failure in one of these capacities can be sufficient (Ref. 70, p 230). At the same time, it is noted that “to have a delusion may not be sufficient” if the defendant’s general functioning is relatively good (Ref. 70, p 69). Interestingly, it also seems like the character of the crime may affect the assessment of the defendant’s accountability. With the Breivik case as the background, the Ministry states in its bill that “it can be hard for the public to accept that a relatively well-functioning person with the ability to plan and execute complicated crimes is not punished” (Ref. 70, p 69). The Supreme Court has followed this route and has stated that, for psychotic but cognitively well-functioning defendants with delusions, the question of whether they have planned their actions or not can be decisive for whether they should be acquitted or punished.^71,72 In our view, this is a problematic approach that may lead to unequal treatment before the law, bearing in mind that severely psychotic persons may retain the ability to plan and commit seemingly complicated acts.⁷³ Similar to the content focus in Anglo-American law, the result may be that equally impaired defendants are treated differently depending on the manifestation of their symptoms.

In this context, it should also be noted that the threshold for being acquitted by reason of criminal insanity is high also in Norway, resulting in sentencing many offenders with severe mental disorders and disabilities. Many of those that are acquitted by reason of insanity are also sentenced with special criminal sanctions of compulsory treatment to prevent them from committing new crimes, like dispositional measures that are used in Anglo-American systems.⁶⁶ Here, an intriguing question is whether the different approaches to criminal insanity result in different forensic patient populations and thus different approaches to treatment and rehabilitation.

Still, as a regulative model for approaching the association between criminal insanity and delusions, the Norwegian approach carries some interesting perspectives. According to its rationale, it allows for a seriously disordered defendant to be acquitted, even without evidence that the disorder influenced the crime. Thus, a defendant can be acquitted even when the content of the delusions did not morally or legally justify the criminal act. They may also be punished when the delusions seemingly provided such a justification, whereas the defendant overall displayed good cognitive capacities at the time of the act. This approach evades focus on the link between disorder and crime and the content of delusions and suggests that the only consideration is whether a relevant and sufficiently severe disorder was present. Hence, the Norwegian model solves the causality problem by assuming that serious enough and legally relevant disorders, when present, generally impair cognitive capacities relevant to responsible actions. This seems reasonable, as there will always be epistemic insecurity in determining whether and how a mental disorder influences a crime.¹⁹

To this point, the Norwegian approach may have some advantages over the Anglo-American approach. The Norwegian approach will not differentiate between two defendants with similar degree of functional impairments who happen to have different manifestations of the content of their delusions and therefore seemingly different motivations for their crimes. It could be argued that equal treatment before the law is better ensured by an approach focusing on the severity of the functional impairment itself, at least if criminal insanity is to be associated to psychiatric notions of mental disorder. As explained, this is currently the case also in Anglo-American systems, because of the influence of forensic psychiatry in legal practice.^2,3 Considering how challenging this association has become, the Norwegian approach asks us at least to reconsider the legal relevance of delusions. This, in turn, however, requires further work in clarifying the link between the manifestation of delusion and a sufficiently severe and legally relevant state of mental disorder.

Reconsidering the Relevance of Delusions

With the Norwegian approach as a point of entrance, the key question for the relevance of delusions is not how they influenced the commission of a specific crime but rather how delusions indicate a state of mental disorder that, when present at the time of the act, entails criminal insanity. To answer this question, we must concretize the relevant state of mental disorder as well as what we mean by delusions and, on this basis, explain their practical association, i.e., how delusions can be proof of legally relevant mental states. This is an ambitious task, and we will thus only sketch some possible ways forward for further discussion.

We propose, as a start, to investigate the possibility of utilizing psychiatric constructs as proxies for legally relevant functional impairments. In Norwegian law, insanity-relevant impairments have, as explained, become impairments in reality understanding or cognitive functional capacity. Thus, the question of the relevance of delusions becomes a question about how delusions relate to such impairments in a cognitive domain associated with the state of being psychotic. To concretize the relevance of delusions then requires us to explore what we know about how delusions may indicate or correlate with cognitive impairments, especially in terms of reality understanding.

This is not a straightforward exercise, not least because it asks for investigations in mental health research. The challenge is that impaired reality understanding and cognitive impairments are subject to different definitions and interpretations. Both cognitive impairment and impaired reality testing are broad psychological terms that are also core features of schizophrenia.⁷⁴ It is thus to a large degree unclear both how relevant functional impairments should be defined and how they relate to psychiatric symptoms, such as delusions.⁷⁵ Moreover, also with given definitions of the cognitive impairments in question, it is challenging to disentangle if and how delusions affect or are a sign of these impairments.

The question becomes even more demanding as the law must decide if the defendant is disordered enough. In other words, given that delusions are relevant per se, we need to understand them according to a severity spectrum of mental disorder. The challenge is then how we should define and assess this spectrum. One possible approach is here provided by the continuum model of psychosis, where some scholars argue that delusions exist along a continuum from normal experiences and thought processes to severe psychopathology.^76,77 Following this line of thought, one may argue that the functional impairments delusions indicate could be assessed as more severe the more the content deviates from normal beliefs. The most severe forms of delusions may indicate not only an incapacity to form justifiable beliefs but also a severely impaired ability to perceive reality accurately. Focus is again on the content of delusions, but not with respect to how it is related to the crime (as beliefs, motivation, and possible justifications) but to how extraordinary the beliefs are in relation to what is deemed as normal in a clinical sense.

The continuum model of psychosis has, however, been criticized for being unscientific and posing challenges to the need for setting thresholds for more or less severe psychosis symptoms, such as delusions.^78,79 An alternative approach to assess the severity of a state of mental disorder is provided through what are called Schneiderian first-rank symptoms (e.g., delusions of thought insertion or of being controlled or the original concept of delusional perception). Originally thought to be pathognomonic for schizophrenia, these symptoms presuppose a radical qualitative change of consciousness that fundamentally involves a disorder of the self (“Ichstörungen” or self-disorders)⁸⁰ that is understood as distinctly different from normal beliefs in both form and content.⁸¹ This approach provides a more global perspective on certain types of delusions as indicators of mental disorder that highlight the fundamental breakdown of the delusional person’s capacity to perceive reality accurately, which may affect the person’s entire cognitive functioning. Intuitively, this seems to provide a possible link between delusions and a sufficiently severe and legally relevant state of mental disorder.

A practical challenge is that the first-rank symptom concepts are not included in the DSM-5 and minimized in the International Classification of Diseases (ICD)-11, which challenges translation to diagnostic categories that are the common basis for forensic experts in criminal proceedings. At the same time, this challenge should not be exaggerated as the severity dimension of delusions is not necessarily linked to the presence of a specific diagnosis and diagnoses are not, as such, legally relevant. More challenging is the fact that the Schneiderian first-rank symptoms exclude certain types of delusions. For example, first-rank symptoms do not include the delusions of persecution, reference, or grandiosity (in their pure form). The omission of these forms of delusion could, to some extent, be seen as a return to a content focus, but again, not in relation to the crime. It is the type of content (persecutory) rather than the specific content (X is spying on me and trying to kill me) that is of relevance. Moreover, the fact that the Schneiderian first-rank symptom has a specificity for schizophrenia introduces a diagnostic categorical bias where, in practice, delusions as part of other disorders (i.e., affective or other psychotic disorders) would be excluded. This does not align with our understanding of the legal relevance of mental disorders, which is based on the extent of functional impairments rather than on specific diagnoses.

The ultimate question of what kind of functional impairment is relevant and sufficient is inevitably normative. After all, we cannot define the threshold of insanity as if it is related to natural distinctions between different mental categories. From psychiatry’s perspective, symptoms and functional impairments occur on a continuum, whereas the threshold between the responsible and nonresponsible must be defined by the law. Furthermore, one should be cautious to attach decisive weight to delusions as symptoms of mental disorder. At least, we must keep in mind that deviations from reasonable beliefs are commonplace in the nonclinical population. It is also well known how difficult it can be to distinguish psychotic delusions from extreme ideas^82,83 and that defendants with extremist views can experience a mental disorder. These observations hold regardless of specific construction of criminal insanity.

A tentative conclusion is thus that we need to clarify the legal relevance of delusions within a legal and normative domain and thus to look beyond delusions as broad psychiatric constructs. Instead of using psychiatric constructs as proxies for criminal insanity, we need to seek clear normative definitions of the relevant functional impairments. In other words, we must return to the law’s concept of criminal insanity and operationalize and possibly revisit the relevance of delusions through this lens. One key to such operationalization is, in our view, empirical study of legal reasoning in concrete cases. In addition, philosophy, as well as mental health research, can help to illuminate law’s normative and empirical assumptions and their consistency with the knowledge perspectives to which the law relates.¹⁹ It may be that, in the end, this will lead us to move beyond focusing on the content of delusions and even beyond the paradigmatic legal view of the relevance of delusions and mental disorders as such.

Future Directions for Law and Research

We started by outlining the problem of linking criminal insanity to the psychiatric construct of delusions. We have argued that the paradigmatic focus on the content of delusions raises several problems and suggested that the Norwegian approach provides an alternative point of entrance to the legal relevance of delusions as it does not involve any causality assessment and therefore evades the focus on delusional content. We have further shown that this approach reveals deeper problems requiring clarification. In fact, there seems to be a more profound problem in law’s association between criminal insanity and mental disorders, which consists in the unclarity in this very association. This challenge appears to be present across different legal systems with significantly different insanity doctrines, and it largely seems to be related to how contemporary criminal insanity doctrines and related legal practices relate to psychiatry.⁸⁴

It is fundamental for the criminal insanity doctrine to identify defendants who should not be blamed and punished. From a constitutional point of view, the most important consideration may even not be precisely how we define the threshold for criminal insanity but rather that this threshold is defined clearly enough to secure equal treatment before the law. Also in such a perspective, it is imperative that we challenge the current paradigmatic emphasis on the content of delusions as evidence of criminal insanity. We hope, therefore, that our paper can stimulate research engagement and discussion about the relevance of delusions for criminal insanity and more generally about why and how mental disorders are and should be relevant to criminal law.