From Melancholia to Depression: Disordered Mood in Nineteenth-Century Psychiatry

My book From Melancholia to Depression: Disordered Mood in Nineteenth-Century Psychiatry charts how melancholia was reconceptualized in the nineteenth century as a modern mood disorder and a precursor to clinical depression. The book shows how this occurred chiefly in two ways. First, the idea of disordered mood as a medical concept was created through the appropriation of language from experimental physiology into the realm of psychopathology. Second, the interplay of statistical and diagnostic practices formed the basis for modern psychiatric classification and facilitated the standardization of melancholia as a psychiatric diagnosis. These developments were key to the reconceptualization of melancholia and the subsequent emergence of clinical depression, and were foundational to modern psychiatric theory and practice.


Introduction
In the twenty-first century, clinical depression has become ubiquitous. The World Health Organization (WHO) considers it to be a leading cause of disability worldwide, and global consumption of antidepressant medication has risen steadily since such drugs were first introduced on the market (Organisation for Economic Co-operation and Development [OECD], 2015: 184;WHO, 2018;Winerman, 2017; see also Healy, 1997). Some have argued that this apparent epidemic of pathological low mood can be attributed to the ills of modern society, while others have suggested that it is the result of a medicalization of 'normal' sadness (Hidaka, 2012: 211;Horwitz and Wakefield, 2007). Another perspective purports that two types of depression exist. One is a milder, anxiety-based form, which has been referred to by different names over the centuries, and increasingly corresponds to depression as commonly understood today. The other is an endogenous, more severe depression, which medical literature has described with notable consistency since antiquity; prior to the twentieth century, this was known as melancholia, but it has become marginalized in diagnostic literature. This view has gained growing traction in recent years, with an international group of clinicians and scholars calling for melancholia to be reinstated in the major psychiatric manuals as a diagnosis distinct from Major Depressive Disorder (Parker et al., 2010). Proponents of the 'two depressions' doctrine point to historical evidence to support the argument that a clearly definable melancholic depression has always been part of the human condition, suggesting that it has been consistently described in medical literature throughout history (e.g. Carroll, 2012;Fink and Taylor, 2007, 2008a, 2008bShorter, 2007).
While melancholia has existed as a concept (or concepts) since antiquity, scholars disagree over whether it should be understood as a timeless, universal condition or whether the medical, cultural and linguistic frameworks of different time periods are too diverse for any meaningful equivalence to be made between, for instance, melancholia in the Middle Ages and the condition with the same name described in the nineteenth century. Stanley Jackson's Melancholia and Depression: From Hippocratic Times to Modern Times (1986) remains the most impressive chronology of melancholia to date. The book offers a continuity narrative, arguing that depressive illness, traditionally known as melancholia, has shown 'both a remarkable consistency and a remarkable coherence in the basic cluster of symptoms' across two millennia (p. ix). In a similar vein, Clark Lawlor (2012) has mapped melancholia's historical trajectory from ancient Greece to the 'paradigm-changing arrival' of the third edition of the American Psychiatric Association's Diagnostic and Statistical Manual of Mental Disorders (DSM-III) published in 1980. Lawlor equally argues for significant continuity across time, and laments the radical shift introduced with the new category Major Depressive Disorder, based solely on descriptive psychopathology and which, he argues, expanded into the realm of normal sadness. Edward Shorter is similarly critical of the DSM approach to mood disorders. A proponent of the two depressions narrative referred to above, he argues for the existence of two distinct forms of depression, one that is endogenous (melancholia), and a socially and culturally produced category, which is today the main form of depressive illness diagnosed in primary care (Shorter, 2013; see also Borch-Jacobsen, 2009).
In contrast to these narratives, a number of writers have highlighted historical change (e.g. Bell, 2014;Radden, 2003;Rousseau, 2000). Misbach and Stam (2006) have mapped a conceptual shift in the nineteenth century, whereby melancholia was 'gradually reconceptualized as depression' through a process of 'medicalization'. Their discussion follows on from Berrios (1988), who emphasized the role of French alienist Esquirol's idiosyncratic term lymemanie ('sadness mania') in the reconfiguration from intellectual to emotional insanity, arguing that the term helped bring about a change in meaning of melancholia, before the latter became gradually replaced by depression. As these histories suggest, there was no straightforward transition from melancholia to depression -the former was not simply replaced by the latter. Existing histories of melancholia in the nineteenth century have, however, almost completely overlooked two events that were fundamental to the reconceptualization of the diagnosis: the appropriation of language and concepts from experimental physiology to talk about emotion as a physiological phenomenon, and the role of asylum statistics in the development of diagnostic categories and criteria.
My book From Melancholia to Depression: Disordered Mood in Nineteenth-Century Psychiatry  seeks to redress these existing gaps in the history of melancholia and depression. In doing so, it also challenges the continuity narratives referred to above. The book maps a crucial chapter in the history of psychiatry: How was melancholia transformed in the nineteenth century from traditional melancholy madness into a modern biomedical mood disorder, paving the way for the emergence of clinical depression as a psychiatric illness in the twentieth century? From Melancholia to Depression charts two key events that were central to the development of psychiatric classification and which form a crucial part of the history of depression. First, it demonstrates how the modern medical concept of mood disorder was created through the merging of physiology and psychology into what scientists termed 'physiological psychology', showing how melancholia was reconfigured according to the new biomedical language that characterized this field of research. Second, it maps the origins of the relationship between statistical and diagnostic practices in psychiatry, which forms the basis for modern psychiatric classification, creating a framework that eventually culminated in the internationally recognized American Psychiatric Association's Diagnostic and Statistical Manual of Mental Disorders (DSM). In sum, these two developments were foundational to modern psychiatric theory and practice, and remain central to the discipline today.
The book draws on a range of sources, including asylum records, medical textbooks and articles, statistical reports, and directives from the English and Scottish Lunacy Commissions. The focus is primarily on the British context, with an emphasis on the uptake of physiology and German neuropsychiatry into British psychological medicine. At a time when modern science offered new and exciting tools for understanding the human body, language appropriated from experimental physiology allowed physicians to speak about the unseen internal operations of the mind using the analogy of sensory-motor activity. In this way, a new framework was created for explaining emotion as an automated, involuntary reaction that could be internally triggered. Following from this, 'disordered emotion' was constituted as a physiological process occurring when the brain was subjected to repeated 'irritation'. This was the medico-scientific foundation of the new medical concept affective insanity, the most common form of which was melancholia.
When it came to diagnosing asylum patients, however, internal biological explanations of disease were of little use. Instead, the focus was on externally observable symptoms of melancholia, chiefly depression, mental pain and suicidal tendencies. The late-nineteenth-century symptomatology of melancholia was in part constituted through statistical practices and was primarily descriptive. At the same time, however, the symptoms that emerged as defining criteria of melancholia were theorized within a biological explanatory framework. Diagnostic descriptions of melancholia travelled back and forth between the casebook and the textbook, producing a disease concept that on the surface displayed remarkable coherence yet simultaneously spoke volumes about the negotiations that take place when medicine seeks to label and classify the complexities of human life. In mapping these events, From Melancholia to Depression intervenes in current debates about the role of classification in psychiatry. It offers a critical and contextualized historical perspective on mood disorders, and contributes to emerging narratives in psychiatry, psychology and the medical humanities that argue for a more flexible and dynamic approach to the diagnosis and treatment of psychological distress.

Summary
From Melancholia to Depression maps the nineteenth-century reconceptualization of melancholia as a modern psychiatric disorder. In doing so, it also tells the story of how mood disorder became a possible and plausible medical concept. The book contributes to a rich scholarship on melancholia and depression in the history of psychiatry and psychology and the history of the emotions, as well as to current debates about the role of diagnosis in psychiatry. It does so through four key arguments: • • Experimental physiology and physiological psychology played a crucial role in the reconceptualization of melancholia and the creation of the modern category mood disorder, as ideas and concepts from these areas of research were appropriated by physicians to explain the aetiology of disordered emotion. • • The growth of asylum bureaucracy and the collection of asylum statistics helped to reshape the melancholia diagnosis in the nineteenth century. The relationship between statistical data and diagnostic categories has been central to psychiatry since the nineteenth century, and played a key role in the development of systems of classification of mental disease. • • The melancholia diagnosis was increasingly standardized toward the end of the nineteenth century, coalescing around four key symptoms: depression, suicidality, mental pain and religious delusions. This occurred in part as a result of a shift in recording practices, whereby narrative descriptions were gradually replaced by single keywords under which a vast range of acts and expressions were merged. • • The history of melancholia and depression highlights the problems attached to classification in psychiatry. The book places current debates about psychiatric diagnosis in the context of historical debates as well as the dominant socio-economic framework, arguing for a nuanced and flexible approach to the description, explanation and labelling of human emotionality.

Contents
Chapter

Chapter 2: The Scientific Foundation of Disordered Mood
This chapter maps how cutting-edge empirical research in experimental physiology carried out in the first decades of the nineteenth century provided a biological foundation for mental disorders in which no visible changes to brain tissue could be found. It builds on the work of Roger Smith (1971Smith ( , 1992, LS Jacyna (1982), Kurt Danziger (1982) and others by demonstrating how some of the core principles of physiological psychology were carried over into the realm of psychopathology. The chapter takes a close look at the idea of psychological reflex action, a key concept that facilitated a view of emotion as automated and involuntary, and thus prone to malfunction (for a history of the reflex concept in the early nineteenth century, see Leys, 1990). It traces the trajectory of psychological reflexion from internal scientific medicine to what became known as physiological psychology, where it provided mid-century British writers with the tools to create a biomedical framework for the phenomenon of disordered mood. The chapter begins by discussing the history of the reflex concept and its role in the development of a physiological model of emotion, highlighting the work of Archibald Billing in Britain and Johannes Müller in Germany. Billing discussed at length two key concepts which were taken up by the next generation of medical scientists to develop a biomedical model for disordered mood: 'morbid sensibility' and 'irritation'. He used the term morbid sensibility to describe 'that state of the nerves or central organs which renders them more susceptible to impressions than natural'. This inherent sensibility of nerves made them susceptible to 'irritation', understood as a 'diseased state' of the nervous system which would produce 'abnormal muscular contractions' (Billing, 1838: 108-9). The idea of irritation of the nerves causing an organ to become morbidly sensitive became central to biomedical explanations for disordered emotion put forward by later writers. Similar models were offered by other contemporaneous British writers who, like Billing, referred much of the current knowledge on the anatomy and physiology of the nerves to Bell and Magendie (e.g. Mayo, 1833, esp. 286, 350-8;Solly, 1836, esp. 150-8). Two recurrent themes in particular should be noted here: the idea that persistent irritation of the nerves caused them to become more sensitive and prone to morbidity, and a model of involuntary reflexive action that did not include the cerebral hemispheres in such activity. While Billing's framework was later taken up in, and adapted to, psychological medicine by the next generation of British physicians, in the early decades of British mental science the impact of work produced by continental writers such as Prussian physiologist Johannes Müller was at least as substantial, if not more so. Müller (1840Müller ( : 933, 1354 described how emotion could trigger involuntary bodily reactions such as 'crying, sighing, and sobbing'. Going further than Billing, he suggested that the brain could be excited by external stimuli. Müller's framework was built upon by Thomas Laycock and William Benjamin Carpenter in the 1840s. Laycock's first monograph, A Treatise on the Nervous Diseases of Women (1840), was based on his observations as a physician at the York County Hospital, and outlined a theory of unconscious mental reflexion. This was further developed in an article a few years later, 'On the reflex function of the brain' (Laycock, 1845). Laycock drew an analogy between motor nerves and those of the 'sensorium', suggesting that 'sensation [is] analogous to movements, abolition of consciousness to motor paralysis' (Laycock, 1840: 113-14). He used the example of hydrophobia (fear of water) to illustrate how involuntary emotional and motor reactions could be induced not just by external stimuli such as physical contact with, or sight of, water, but also 'by the mention of water', that is, by an idea alone. An involuntary emotional reaction -fear -would be triggered, and the patient 'immediately attempts to remove [the water]. This movement is strictly involuntary, and not the result of sensation'. (Laycock, 1845: 301-2). A similar model of reflexivity was offered by Laycock's former university colleague Carpenter (1842), who coined the term 'ideo-motor' activity to describe this process. However, while Laycock extended the reflex to include all mental functions, Carpenter maintained a clear division between the lower parts of the nervous system and a higher realm which was not subject to reflexive action. Tracing the work of Laycock and Carpenter into the realm of psychopathology, the chapter ends by looking at how physiological psychology was gradually taken up by mid-century asylum physicians writing on mental disease.

Chapter 3: The Classification of Melancholia in Mid-Nineteenth-Century British Medicine
At the time when Laycock first put forward the idea of psychological reflex action, the nosological status of melancholia was contested. At the turn of the nineteenth century, French alienist Phillippe Pinel (1806: xvi-xix, 242) had suggested that some forms of insanity manifested with only or primarily a disturbance of emotion, while the intellect was left partially or wholly intact. The argument that mental disease could affect the emotions without the presence of delusion soon became widely accepted, but at the same time new disease concepts proposed to classify such forms of insanity threatened to eclipse melancholia as a diagnostic category. J.E.D. Esquirol proposed the category 'monomania' to denote mental disease that manifested with only partial and limited delusion, with the subcategory 'sadness mania' (lypemanie) to replace melancholia (Esquirol, 1838(Esquirol, /1845. Bristol physician James Cowles Prichard trained under Esquirol in Paris and drew extensively on the latter's nosology in his Treatise on Insanity (Prichard, 1835), the most influential British textbook on mental disease published in the first half of the nineteenth century. Following Esquirol, Prichard classified melancholia as a form of partial insanity, a type of madness in which the intellect was only compromised in regard to one particular aspect or idea. This move was significant for the subsequent classification of melancholia in Britain. In the following decade, the Metropolitan Commissioners in Lunacy (a precursor to the nationwide Lunacy Commission set up in 1845) adopted a nosology in which melancholia was classified as a form of partial insanity alongside monomania and moral insanity (Metropolitan Commissioners in Lunacy, 1844). James Davey, who worked under John Conolly at Hanwell Asylum and subsequently ran a private facility near Bristol, merged this nosology with the new physiological psychology. He suggested that melancholia was best understood as a form of 'mania' in which 'the most prominent symptom is grief'. He also described melancholia as a type of moral insanity, which he provided with a biological explanation that drew on Billing's work. He noted that Billing had demonstrated that 'the consequence of the brain or spinal cord becoming in a state of morbid sensibility is, that their healthy actions are deranged', and concluded that '"mental excitement", such as anger, grief, fear, etc., which are analogous to the direct irritation of the brain or spinal cord by a depressed fracture or spicula of bone' would, if carried on over time, result in a 'state of morbid sensibility of the nervous centres' (Davey, 1855: 828-9, 831).
Other early-to mid-century writers also classified melancholia as a subcategory. For instance, John Conolly (1846: 1) referred to the condition as a 'variety of maniacal affection', whereas Joseph Williams (1852: 49-50) held melancholia to be a form of partial insanity alongside monomania and moral insanity, largely in accordance with the nosology adopted by the Lunacy Commission. This was also the approach taken by Henry Monro (1851: 12), who situated melancholia and disordered emotion firmly within a physiological framework. Echoing the work of German psychiatrist Wilhelm Griesinger (see below), Monro used the analogy of 'tone' to describe a healthy versus diseased mind, suggesting that in melancholia the 'tone' of the brain was slackened or lost. George Robinson offered a similar model, arguing that the brain could be excited into involuntary emotional reflexion by psychological factors (recall the 'mention' of water in Laycock's example of hydrophobia discussed above). External or internal stimuli would cause repeated 'irritation' of the brain, resulting in 'painful affections in parts far distant from the original seat of irritation'. Emotion would, in the same way as sensation, travel as an automated reaction and trigger 'an elaborate chain of thoughts and sentiments' (Robinson, 1859: 20-3, 43-5). Like Monro, Robinson followed the Lunacy Commissioners in classifying melancholia as a form of partial insanity together with monomania and moral insanity.
John Bucknill's and Daniel Tuke's widely read Manual of Psychological Medicine (1858) drew together knowledge and research from across Europe, but early editions gave only limited space to physiological explanations for mental disease, focusing instead primarily on a description of symptoms. The authors adhered to the prevailing view that melancholia could exhibit with or without delusion, and they largely followed the nosology favoured by the Lunacy Commission, with the exception that they perceived melancholia to be a form of monomania (pp. 157-9). In later editions, the authors gave greater emphasis to an internal, biomedical framework for mental disease and discussed melancholia as an independent disease entity.
This was the approach taken by John Millar in his Hints on Insanity (1861), which favoured a tripartite division of melancholia, mania and general paralysis. Millar's work illustrates a shift in the second half of the century toward classifying melancholia as an independent disease category, which received growing attention in medical literature. This was at least in part a result of rising concerns about suicidality and its association with melancholia. Millar made an observation that was relatively peripheral in medico-psychological literature in the first half of the century, but which later became widely accepted as a medical fact. 'Every case of melancholia', he suggested, 'should be looked upon as having a suicidal tendency' (pp. 22-3).
Chapter 3 ends with a review of existing literature on melancholia by William Sankey (1863), which indicated that, nosological disagreements notwithstanding, a coherent approach was beginning to emerge. In published material there was growing agreement on the validity of a biomedical approach to mental health and disease, and the view of emotion as cerebral and subject to disorder was becoming widely accepted. While disagreement on the nosological status of melancholia prevailed, descriptions of the disease were increasingly distinctive, precise and comprehensive.

Chapter 4: Melancholia and the New Biological Psychiatry
By the 1860s, the concept of disordered emotion as a physiological phenomenon was firmly established in British medicine. The physiological models of mental operations proposed by Laycock and others were foundational for this development. At the same time as Laycock put forward his theory of cerebral reflex action, German physician Wilhelm Griesinger (1843Griesinger ( /1968) had presented an almost identical model which he referred to as 'psychological reflex action' (psychische Reflexactionen). Unlike Carpenter and Laycock, Griesinger applied his theoretical framework for mental reflex action to melancholia and disordered mood. His textbook on mental pathology, first published in 1845 with a second, revised and expanded edition reaching a wide European audience in the 1860s, offered one of the century's most widely cited nosological descriptions of melancholia firmly anchored in a physiological model of emotion (Griesinger, 1861: 34, 213-16).
Griesinger's model for describing the internal aetiology of mental disease followed analogically the kind of physiological description offered by Billing and other contemporary European writers such as Müller. The 'irritation' of an organ leading it to become more sensitive or, in Billing's words, subject to 'morbid sensibility', served as a plausible aetiological explanation for affective insanity. Language used to describe observable disease in organic tissue was applied to make the unseen and unknown familiar and explicable. In this way, Griesinger was able to describe the internal events that produced a state of disordered emotionality, eventually leading to affective insanity such as melancholia. These ideas were further developed in his later work. Noting that mental images were accompanied by pain or pleasure, Griesinger suggested that a mind that had become disordered after prolonged irritation was in a continuous state of mental pain. This pain could be specific, relating to a particular idea, but more often it was vague and diffuse, affecting all 'emotion or the intellect as a whole'. 'Much like in a bodily state of general pain and discomfort, so in the mind a causeless feeling of trepidation, anxiety, etc., when long continued, will eventually develop painful ideas.' It was in such general mental pain and 'lowness of spirits' that melancholia consisted (Griesinger, 1861: 25-34, 213-29).
A few years later, Henry Maudsley's The Physiology and Pathology of Mind (1867) sought to merge physiology and psychopathology, drawing heavily on the framework developed by Laycock, Griesinger, Carpenter and others. Maudsley also endeavoured to settle the question of classification, presenting what he argued to be the most accurate nosology based on empirical judgement. A truly scientific system of classification, he argued, was one that divided mental disease into two broad categories: 'insanity without positive delusion and insanity with delusion, in other words, into affective insanity and ideational insanity', with further subdivisions of these (p. 2). In this way, Maudsley insisted on a marked division between mental disorders where partial or complete delusion was present, and those where only the emotions were affected. This move was particularly significant as it occurred decades before German psychiatrist Emil Kraepelin presented his famous affective/cognitive dichotomy with the diagnostic categories dementia praecox (later schizophrenia) and manic-depressive insanity. Maudsley's classification saw melancholia divided between the two umbrella categories, with simple melancholia in the former and delusional forms in the latter, but this did nothing to diminish its growing significance in psychopathological literature nor its status as an increasingly coherent disease category. On the contrary, the formal institution of non-delusional, affective insanity into diagnostic literature, and the considerable attention Maudsley awarded to the discussion of the symptomatology of both simple and delusional melancholia, was an important step towards the standardization of the diagnosis in the second half of the century.
As Sankey's (1863) literature review suggested, in the second half of the nineteenth century a coherent approach to the classification of melancholia was beginning to emerge across Europe. At the same time, important national and regional differences existed. For instance, in French literature the concept of circular insanity, or folie circulaire, gained greater currency than in Britain. In the 1850s two French alienists, Jules Baillarger and Jean-Pierre Falret, independently described a form of insanity in which patients oscillated between melancholia and mania. Falret suggested that it was not merely 'a variety, but a specific form of insanity' and gave it the name folie circularie. He viewed it as a more or less lifelong, chronic illness, but with milder symptoms than were often found in melancholia and mania proper (Falret, 1851;see also: Anon., 1854-55;Berrios, 1988: 301-2). While the work of both physicians was taken up by medical writers elsewhere, the concept of circular insanity only held minor presence beyond France. In Britain, a relationship between melancholia and mania was noted in both published literature and asylum case notes, but when the term circular insanity or folie circulaire was used, it was predominantly to talk of a sub-variety of melancholia or mania rather than a distinct disease category. George Savage (1884a: 123) echoed the views of many of his peers when he remarked that cases of genuine circular insanity were 'extremely rare' among English lunatics, suggesting that in most of the patients under his care who had exhibited such a circular symptom picture, the proper diagnosis was one of 'recurrent mania', sometimes ending in dementia.
If the French model of circular insanity failed to gain widespread theoretical popularity or practical use in Britain, German late-nineteenth-century conceptions of melancholia were more extensively appropriated. The second edition of Griesinger's textbook, published in 1861 and translated into English a few years later, was widely read across Europe and Britain. Its strong focus on a psycho-physiological model of disordered mood and on the aetiology and symptomatology of melancholia set the tone for the next generation of textbooks on psychopathology.
The Austrian neurologist and psychiatrist Richard von Krafft-Ebing made an important contribution to late-nineteenth-century diagnostic literature with his comprehensive general textbook on mental disease (Krafft-Ebing, 1879-1880); several editions were published, and it was also translated into English (Krafft-Ebing, 1903). The book incorporated significant sections from an earlier short monograph entitled Melancholia: A Clinical Study, which situated melancholia within a framework of physiological psychology (Krafft-Ebing, 1874). Krafft-Ebing argued that melancholics experienced a form of 'mental neuralgia' (psychische Neuralgie) analogous to, but functionally different from, physical neuralgia. This psychic pain defined and dominated the overall mental state of the melancholic, and arose from internal rather than external causes. Once the melancholic was trapped in a cycle of mental pain, every new impression generated more suffering, until the patient was 'unable to rejoice over anything, but . . . equally unable to experience sadness' (pp. 1-6). For many melancholics, the only escape from this intolerable pain appeared to be suicide. Here Krafft-Ebing expressed a view that was rapidly becoming dominant across Western Europe, that '[t]he majority of people who commit suicide are melancholics' (p. 65).
Krafft-Ebing's description of melancholia, which highlighted depressed mood, mental pain and suicidality, is illustrative of the increasingly coherent diagnosis that emerged in the last quarter of the century. While disagreement on the nosological status of melancholia prevailed, descriptions of the disease were increasingly distinctive, precise and comprehensive. By the turn of the twentieth century, the melancholia diagnosis was remarkably standardized for the time, coalescing around four key symptoms: depression, mental pain, suicidal tendencies and (in more severe cases) religious delusions.

Chapter 5: Statistics, Classification, and the Standardization of Melancholia
In 1883, Thomas Clouston suggested that 'a typical case of melancholia' would run 'a somewhat definite course, like a fever' and was characterized by 'mental pain, emotional depression, and sense of ill-being', often accompanied by a 'loss of self-control, or insane delusions, or uncontrollable impulses towards suicide', and 'with no proper capacity left to follow ordinary avocations, with some of the ordinary interests of life destroyed, and generally with marked bodily symptoms' (Clouston, 1883: 37). The frequent use of 'depression' to describe the general emotional state of the melancholic in the nineteenth century has led twentieth-and twenty-first-century scholars, both in the human and natural sciences, to attempt to equate melancholia with the condition commonly known today as clinical depression or Major Depressive Disorder. For nineteenth-century physicians, the term was applied in a way that suggested a much more literal meaning, that of something being 'pressed down'. In melancholia, the 'tone' of the mind was slackened and subdued. Doctors spoke of a 'depression of spirits', or an 'emotional' or 'mental' depression when describing melancholia, but the term was predominantly used as a way to describe the overall state of the melancholic mind, and as a way to contrast melancholia with mania -the opposite state of 'mental excitation'. Depression was not used in place of melancholia as a term denoting a specific disease, but rather it was a key symptom of the condition, alongside suicidality, mental pain and religious delusions. Clouston's definition of melancholia (see above) illustrates an emerging consensus on the defining features of the condition in the last quarter of the nineteenth century. While only a few decades earlier the status of melancholia as an independent and widely accepted disease category had been called into question, towards the end of the century physicians awarded significant attention to melancholia in diagnostic literature, where it was described in remarkably standardized terms. The creation of a coherent and precise diagnosis was the result of an interplay of different factors, including a near-universal adoption of a biomedical disease model, statistical analysis of symptoms, and a change in recording practices in asylums. Central to the latter was the relationship between asylum case notes and published literature (see below).
The statistical analysis of symptoms was of particular importance for the emergence of suicidality as a key symptom of melancholia. In the early decades of the century, suicidal tendencies were not highlighted as a prominent symptom of melancholia, and statistical analysis was primarily concerned with rates of completed suicides. The 1845 Lunacy Act (House of Commons, 1845) stipulated that every county in England and Wales must provide for the care of pauper lunatics, and created a national body, the Lunacy Commission, to oversee the management of county asylums. The Act also set out a range of new provisions for asylums, and standardized some existing ones, such as the documentation required for admission to an institution. This included a medical certificate of insanity (two for a private patient) stating the 'facts indicating insanity' as observed by a physician, and a reception order, usually completed by a magistrate, poor house official, or relative or friend of the lunatic. The latter asked for a yes or no answer as to whether the person to be admitted was 'epileptic, suicidal, or dangerous to others'. Upon admission, this information was transferred into the asylum casebook, where a diagnosis had to be recorded within a week. When asylum superintendents completed their annual reports to the Lunacy Commission, they collated statistical data pertaining to virtually every aspect of their institutions, including patient numbers and demographics, dietary regimens, financial records, and deaths, as well as diagnoses recorded. Physicians increasingly also included tables on the number of 'suicidal' patients under their care, a practice that was encouraged by the Lunacy Commission. When statistics on suicidality were merged with statistics on diagnoses, the data suggested that suicidal tendencies were more common in melancholia than in any other form of insanity.
This facilitated a growing focus on suicidality as a symptom of melancholia on asylum wards, as well as in published literature. Among British asylum physicians, consensus formed around the view that 'the majority of melancholics are suicidal' (Strahan, 1893: 104). The condition and the symptom became mutually constitutive, illustrated by Henry Maudsley's assertion that '[s]uicidal feelings and attempts are common in melancholia, so much so that one suspects their actual or possible existence even when they have not been openly manifested' (Maudsley, 1879: 384). As outlined in the next section, a wide range of different acts and expressions observed in melancholic patients were merged under the 'suicidal' label. This led some physicians to question the validity of asylum statistics for suicidality. For instance, George Savage, superintendent at Bethlem Hospital, remarked that while '20 to 30 per cent' of patients under his care were 'described as suicidal' on admissions documents and in case notes, only a fraction of these were, he argued, '"actively suicidal"' (Savage, 1884b: 17). Others expressed a similar view, and the usefulness of statistics also became subject to a heated debate at one of the meetings of the Medico-Psychological Association, the main professional body for British alienists. Inconsistent recording practices were highlighted as a key issue, and Henry Monro declared that 'of all the humbugs of the present day that of statistics is the greatest' (Royal Medico-Psychological Association, 1865: 415). Nevertheless, the correlation between melancholia and suicidality persisted, and was closely linked to another key symptom, mental pain.
The historical trajectory of mental pain as a symptom of melancholia is very different from that of suicidality. In the late nineteenth century, mental pain was constituted in biomedical terms, a 'psychalgia' analogous to physical pain. However, its relationship to melancholia and sadness is rooted in Christian doctrine, where it was closely linked to sin. At the turn of the eighteenth century, philosopher Mary Astell held that 'mental Pain is the same with Sin', in the sense that 'as a musical Instrument, if it were capable of Sense and Thought, would be uneasie and in pain when harsh discordant Notes are play'd upon it; so Man, when he runs counter to those Motions his Maker has assign'd him . . . must needs be in Pain and Misery' (Astell, 1705: 53). In a similar vein, clergyman Richard Fiddes (1720: 639) explained mental pain as 'that Anguish and Remorse of Mind, which Sinners so naturally feel, and all of them, more or less, when they call their own Ways to remembrance, and reflect upon their sins'.
This conception of mental pain is in stark contrast to that espoused by nineteenth-century physicians, who explained it in physiological terms. At the same time, early modern ideas about mental pain as closely tied up with sin were not simply erased with the advent of modern medicine. Patients continued to experience their pain and distress in spiritual terms, but physicians wedded to a psycho-physiological model of the mind and brain saw mental pain as functioning in the same manner as physical pain. In this way, it would arise when the brain was subjected to repeated irritation. At the same time, unlike physical pain and visible lesions, mental pain was an abstract concept, the perceived result of a physiological process that was theorized but which could not be observed. In this way, mental pain as understood in nineteenth-century medicine was both metaphorical and literal. As described above, concepts derived from empirical research were appropriated to explain unseen mental operations, which were nevertheless perceived to be real and tangible physiological events. The language of the physiology of mind, centred on words like irritation, tone and pain, was used to explain material cerebral processes as well as the abstract psychological manifestations of these; this allowed physicians to argue that the same relationship existed between psychological pain and involuntary action as between physical pain and automated muscular activity.
While physicians conceptualized mental pain as a physiological phenomenon, most patients continued to experience their pain in spiritual terms. Melancholic patients were often weighed down by overwhelming religious guilt, and frequently spoke of having committed 'the unpardonable sin', believing this to be the source of their suffering. For physicians, however, this was another symptom of illness, usually referred to as 'religious delusions'. While this symptom was generally not considered a defining feature of melancholia in the same way as depression, mental pain and suicidal tendencies, it was nevertheless awarded significant attention in diagnostic literature, and was frequently recorded in asylum case notes. Moreover, it allowed physicians to emphasize that only a trained professional could correctly identify the real cause of melancholic patients' suffering. Clouston (1883: 38) remarked that melancholics were usually able to give a comprehensive description of their symptoms, thus aiding the physician in reaching a diagnosis, but they were, he argued, almost always wrong about the cause of their depression due to 'their false, ungrounded beliefs'. Having 'analysed the "causes" assigned by melancholics' under his care over a number of years, Clouston concluded that they were 'wrong in ninety per cent of cases' (see also Maudsley, 1895: 360-1). For melancholic patients, the profound religious despair and fear of God's wrath they often expressed were very much real, but for physicians this was an example of patients assigning an incorrect cause to their suffering, the true source of which was an internal disorder understood in physiological terms.

Chapter 6: Diagnosing Melancholia in the Victorian Asylum
Towards the end of the century, a biomedical model of disordered emotion dominated in Britain and beyond. It underpinned the symptomatology of melancholia, and textbooks on psychopathology generally included detailed discussions on brain physiology. When it came to diagnosing patients, however, this model was of little use. While some physicians expressed the hope that biology would one day aid in the diagnosis of mental disease, they were nevertheless forced to rely on externally observable symptoms, in particular conversations with patients. Patient interviews were, Clouston argued, especially helpful in the case of melancholia as the 'patient is usually conscious that there is something wrong with him, which is not the case in most forms of insanity' (Clouston, 1883: 37-8). At the same time, the symptoms recorded by physicians, such as depression or suicidal tendencies, were not communicated in those words by patients. Rather, a significant amount of intellectual work was required to turn patient testimonies into precise and classifiable symptoms. As asylum recording practices were made increasingly efficient and standardized, events and experiences communicated by patients and their loved ones were translated into simple key terms on the pages of casebooks. In this way, the melancholia diagnosis was standardized and refined as it travelled back and forth between the casebook and the textbook.
In the late summer of 1874 Joseph Brown, assisting physician at the Royal Edinburgh Asylum at Morningside, came across a case of melancholia that he considered significant enough to warrant publication. His article on a 'Case of determined suicidal melancholia', which appeared in the Edinburgh Medical Journal later that year, described a patient whose actions offered a 'striking illustration of the great difficulty there exists in preventing a determined suicidal patient from accomplishing his object' (Brown, 1874: 402). The patient, Moses B., had been certified a lunatic and admitted to Morningside (as the asylum was known) because his father and brother felt that he could not 'be left alone' for the fear 'that he would seek to destroy himself'. According to the certifying physicians, Moses believed that his soul was lost, and he was reported to have taken 'a poisonous dose of Belladonna'. When examined on admission, his 'depression' was described as 'considerable', and his 'suicidal tendencies' were noted as consisting in 'taking belladonna, refusing food, &c'. The patient's recent mental symptoms were listed as 'delusions such as that his soul is lost, that he ought to die, and thinks he is committing great sins'. He was given the diagnosis melancholia, with a special reference made to his persistent suicidal tendencies (Royal Edinburgh Asylum, 1873-1894. It is not surprising that Brown decided to write up the case of Moses B. for publication; it served as an apt illustration of a typical case of melancholia, characterized by profound depression of mind, persistent suicidal tendencies, and delusions of religious guilt. According to Clouston, Morningside's superintendent, such cases were becoming increasingly common. Clouston had taken over the running of the asylum from his mentor, David Skae, the previous year, and immediately began instituting changes to recording practices that facilitated a standardization of diagnoses and symptoms. He gradually replaced Skae's (1863) aetiological system of classification, which had organized mental disorders according to their cause (e.g. 'mania of lactation', 'sun-stroke mania'), with the more commonly used descriptive nosology that included diagnoses such as melancholia and general paralysis. Clouston also introduced casebooks with pre-printed section headings, allowing for swifter and more uniform recording of symptoms. These changes facilitated a noticeable standardization of symptoms as lengthy narrative descriptions were increasingly replaced with brief key terms such as depression and suicidal tendencies (Jansson, 2021: ch. 6).
A similar picture emerged in other asylums across the country in the last quarter of the nineteenth century. A vast and highly varied landscape of human emotionality was increasingly flattened into singular key words and brief descriptions, which created the appearance of uniformity. This development was in part driven by a requirement to provide homogenous and comparable statistical data on patient populations and diagnostics, which created a need for standardized recording practices. In addition, these changes to how patient data were recorded were also facilitated by time constraints in a period when asylums and their populations were rapidly expanding across the country. They formed part of a broad shift in record-keeping within western medicine whereby longer, narrative descriptions of patients' physical and mental states gave way to briefer statements and descriptive keywords. John Harley Warner has aptly described this shift as a 'new epistemological and aesthetic sensibility, expressed as a narrative preference for what was universal and precise over what was individual and discursive' (Warner, 1999: 109). The implications for the melancholia diagnosis were that labels like depression and suicidality came to denote a wide range of acts and expressions, producing a diagnosis that was increasingly standardized on its journey between casebooks and published literature. Walker and O'Connell (2012) have noted the homogenizing function of depression as a single descriptive term for a range of emotional states associated with low mood, remarking that it 'not only replaced a wider vocabulary for a variety of experiences', but that it 'also flattened out the individuality' of those experiences. In the Morningside casebooks, the term depression was increasingly used without explanation. For instance, Dorothy D.'s 'depression' was observed in 'her conversation and the expression of her countenance', Jane Ann C.'s depression was 'exhibited in manner, appearance, and communication', Isabella Hutton's depression was 'marked, exhibited in appearance and conversation', and Catherine G.'s depression was simply 'present in her expression' (Royal Edinburgh Asylum, 1873-1894. The process of homogenizing the recording of symptoms of melancholia was also particularly evident in the use of the term 'suicidal'. A vast range of acts and expressions were merged under this label, as can be seen in the case of Moses B. referred to above (see also Jansson, 2013). He was deemed suicidal prior to admission after having consumed 'a poisonous dose of Belladonna', and according to Morningside's physicians continued to make attempts on his own life once inside the asylum. He was reported to have swallowed several stones, after which he complained of abdominal distress and declined his meals. This act of 'refusing his food' was seen as an attempt to commit suicide by starvation, which was followed by another suicide attempt consisting in taking 'an overdose of alcohol'. Around two weeks after this incident, from which he subsequently recovered, the patient 'snatched' from an attendant a bottle containing 'a solution of guttapercha in chloroform', which he proceeded to drink. According to Joseph Brown, the persistence with which Moses attempted to end his own life constituted a demonstrable example of the suicidal tendency in melancholics -so determined did the patient seem in his suicidal convictions that 'his one and only object in life is to destroy it' (Brown, 1874;Royal Edinburgh Asylum, 1873).
In this one case, then, a number of acts were recorded as evidence of suicidal intent, and other cases of suicidal melancholia widen the range further. Isabella M. was deemed suicidal because she would walk 'about the streets in a depressed state -found standing on the top balcony to jump over, &c.', and Alexander M. after he 'wanted to get a knife'. Jane C. was noted as having 'threatened' suicide after she 'wandered away to the Dean Bridges' but did not jump in the river 'as she says the water was not deep enough', and Alexander B. 'said that he was tired of life and wished to have done with it, but he has made no actual attempts'. In the case of Elizabeth R., the suicidal tendency appeared to be derived from reports that she 'wanders about at night' and 'refuses food'. Refusal of food was equally the reason given for the suspected presence of suicidal intent in Jane Ann C. and Isabella M., while in the case of Isabella H. no apparent reason was given beyond the 'usual' symptom of 'depression', deduced from the patient appearing 'dull and melancholy and despondent'. In a number of such cases where the 'suicidal' label came without explanation, such as those of James C. and James W., both admitted in 1888, few symptoms were given beyond the familiar 'depression', 'lowness of spirits ' and'despondency' (Royal Edinburgh Asylum, 1873-1894). In sum, the Morningside casebooks appeared to confirm the belief expressed by Maudsley (1879: 384) that '[s]uicidal feelings and attempts are common in melancholia, so much so that one suspects their actual or possible existence even when they have not been openly manifested'.
A similar picture emerges from county asylums in England at this time. Brookwood asylum in Surrey also introduced pre-printed headings in the 1870s, and many of its patients were labelled suicidal in the apparent absence of suicidal acts or openly manifested intent. In the case of John G., no specific reason was given for considering him to harbour suicidal tendencies beyond the patient stating that 'he feels very low sometimes but cannot give any reasons for being so'. In patients diagnosed with melancholia, any act or expression that could be construed as having self-injurious intent was likely to be labelled suicidal. Ann F. was labelled as suicidal because she 'seemed very much depressed and expressed an intense feeling of melancholy, as if she would do herself an injury'. Ann W.'s suicidality was reportedly manifested in 'persistent refusal of food', while Sarah Elizabeth L. was said to have 'attempted suicide by pressing her fingers slightly round her throat', which meant that she required 'constant watching' (Surrey County Asylum, 1868-1872). Brookwood's superintendent Thomas Brushfield (1880) expressed a similar scepticism regarding statistics on suicidal patients as that voiced by George Savage. Clouston (1883: 118) was equally wary of their reliability, noting that while the data suggested that melancholics were overwhelmingly suicidal, experience suggested that 'the actual risk of suicide being seriously attempted or accomplished is much less than those figures seem to show'. Nevertheless, the recording practices put in place by Clouston, Brushfield and others contributed to the widely accepted view of a close relationship between melancholia and suicidal tendencies, with consequences for the admission, diagnosis and care of melancholics in the late nineteenth century.
The developments outlined above have continued to shape medical perceptions of emotional distress into the twenty-first century. The emergence of suicidality as a category separate from suicide occurred within a wider context of psychiatric attention increasingly brought to bear upon a range of 'self-injurious' behaviours (see Chaney, 2017;Millard, 2015). In twenty-first-century psychiatry, 'non-suicidal self-injury' has become a firmly cemented diagnostic concept (see e.g. APA, 2013: 803-5); however, the parameters and definitions of what distinguishes suicidal from non-suicidal self-injury have shifted over time as these categories have been reconceptualized in the context of contemporaneous cultural tropes about group and individual behaviours. In nineteenth-century medicine, 'suicidal' was a multivalent and shifting concept, yet while physicians remarked on the ambiguity of the term and the statistical data it produced, they nevertheless continued to rely upon this category as a significant diagnostic criterion in the definition of melancholia.

Concluding remarks
The conclusion of the book briefly considers diagnostic shifts at the turn of the twentieth century, centred on Emil Kraepelin's (1899) separation of affective and cognitive insanity, and Adolf Meyer's proposal to replace the term melancholia with depression (see New York Neurological Society, 1905: 114). 1 It then goes on to consider the problem of diagnosis in psychiatry in the context of current debates around classification, as well as the relationship between psychiatric knowledge and wider socio-economic developments.
The story of how melancholia was reconceptualized as a modern biomedical mood disorder highlights two inherent and prevailing tensions within psychiatric diagnostics: on the one hand between biological disease models and descriptive nosologies, and on the other between precise medical categories and the infinite variations of human emotionality. While nineteenth-century doctors acknowledged the difficulty in attaching neat psychiatric labels to the diversity of acts and expressions encountered in patients, they nevertheless remained convinced of the necessity for psychiatric classification, no matter how flawed any such system was. In this way, they set the trend for psychiatric epistemology from then onwards.
Disagreement prevails within psychiatry over how to classify mental disorders, but the argument that classification is necessary remains pervasive. At the same time, the usefulness of psychiatric classification is increasingly being challenged, with some critics arguing that it does more harm than good (e.g. Kinderman, Read, Moncrieff and Bentall, 2013). Callard and Bracken have highlighted the dangers of 'diagnostic overshadowing' (physical symptoms being automatically attributed to existing psychiatric illness) as well as the negative consequences of long-term institutionalization that can result from some types of psychiatric diagnoses Callard, Bracken, Davies and Santorius (2013). They conclude that 'the [mental health] interventions that have arguably empowered people the most, such as innovative community services, have not been diagnosis specific'. It is also important to note the difficult relationship between psychiatric diagnosis and identity. Arguments for parity between physical and mental health often turn to biology for their defence, suggesting that these two areas of pathology should be valued equally because depression is no different from cancer or a broken leg. This argument is flawed and potentially harmful. Parity need not be rooted in sameness. Psychiatric disorders are evidently not like a broken limb; unlike the latter they concern the part of us that is at the core of our personhood -our minds. Regardless of whether the cause of a person's mental distress is primarily social or primarily biological, its manifestations tend to be predominantly (but not only) psychological, and consequently also relational and identity-based, in ways that a broken leg is not. If we frame mental disorders in strictly biological terms, we run the risk of imprinting human beings in all their complexity with simplistic, reductionist biological stamps, which can potentially have a self-perpetuating effect. 2 One the other side of the argument, however, there is also a danger in a wholesale rejection of existing diagnoses as not 'real' or 'true'. Depression currently exists in psychiatric literature as a mental illness; people are diagnosed with and treated for depression in large numbers every year, and millions of people across the world experience themselves as 'having depression' or 'being depressed'. These factors must all be seen as determinants of the existence of an illness, irrespective of how it entered medical knowledge. It is also undoubtedly the case that many people find it helpful to understand their suffering in medico-scientific terms, and that this can alleviate feelings of guilt, shame and personal responsibility. A more helpful and nuanced approach is one that accepts and validates both the concept and experience of depression (or melancholia) and other psychiatric diagnoses as legitimate medical conditions, while at the same time giving equal weight to other ways of understanding, experiencing and labelling psychological distress.

Declaration of conflicting interests
The author declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding
The author disclosed receipt of the following financial support for the research, authorship and/or publication of this article: The research for this book was supported by the Wellcome Trust, grant no: 092988/Z/10/Z.

Notes
1. For a discussion on the relationship between melancholia and depression in the twentieth and twentyfirst centuries, see Jansson, 2020. 2. Moreover, attempts to destigmatize mental illness by equating it with other medical conditions can lead people to feel 'less optimistic about their ability to get better' and increase public 'perceptions of dangerousness and unpredictability' of such disorders (Malla, Joober and Garcia, 2015: 147-50).