KRAEPELIN'S SYSTEM OF mental diseases has contributed substantially to the foundation of modern psychiatric diagnosis in DSM-IV and ICD-10. A return to Kraepelin's ideas, so-called ‘neo-Kraepelinianism’, has inspired a long list of publications and new research approaches towards Kraepelin's teachings, methodology, understanding of science, nosology, worldview and towards the figure of Kraepelin himself. As in the past, this new research deals with the implications of Kraepelin's nosology for modern psychiatric classification and diagnosis.
This review's enquiry arose against this background of the correlation between modern psychiatry and Kraepelin's work: how did Kraepelin's contemporaries and successors react to his nosology, bearing in mind that Kraepelin's nosology was a major target of criticism at the time? Our goal was to investigate the extent to which the critical figures of Kraepelin's era concur with today's evaluation of Kraepelin's scientific results.
Thus there are two sides to the question of how experts in psychiatry – particularly German experts – reacted to the appearance of Kraepelin's teachings:
We examined individual phases of varying reception of Kraepelin's work by differentiating between criticism voiced, on the one hand, by authors of his own era, and on the other hand, by contributors following his death in 1926 up to 1960.
Between 1900 and 1960, three periods of differing perception emerged:
- 1In the sixth edition of Kraepelin's Textbook of Psychiatry, published in 1899, his nosological concept is presented in its entirety. The first criticism of the work appeared in 1902. The period until Kraepelin's death in 1926 was a phase of harsh criticism of his nosology. This period includes the First World War, after which criticism was less prevalent.
- 2The second period of the reception of Kraepelin's teachings lasted from 1926 to 1928. It is characterized by a return to Kraepelin and includes, together with the ‘rehabilitation’ of Kraepelin and his work, only criticism of individual items of his classification system.
- 3The third period, from 1929 to 1960, includes Kraepelin's centenary in 1956. During this period, Kraepelin's authority was broadly accepted; criticism was sparse and confined to individual points. Most reactions to Kraepelin's teachings from about 1920 onward were positive, and Kraepelin was held in high professional esteem.
This review takes into account rare historical literature in medical and general libraries and archives in Munich, Berlin and Paris to elucidate early reactions to Kraepelin's nosology. Most of the literature is not available in web-based libraries, and some of the books and papers in the historical archives are unique. Articles and books were searched with the keywords ‘Kraepelin’, ‘dichotomy’, and ‘manic-depressive disorder’. All references and cross-links were also searched. Arguments pro and contra Kraepelin from different countries and eras were compared and are commented on. Overall, early critics of Kraepelin's work argued against the whole concept of his nosology, comparing it to classical psychiatric schools of the 19th century. Debates were carried out with many single arguments based on differing historical and regional psychiatric concepts, whereas contemporary authors argue against the dichotomous classification without offering alternative options of psychiatric classification. Despite all the discontent with Kraepelin's dichotomy, psychiatric classification is still based on his nosology.
Reception of Kraepelin's ideas during three periods
The first period, 1900–1926, was initially characterized by massive criticism of Kraepelin's dichotomy, which he had clearly formulated in the sixth edition of his textbook in 1899. The main target of criticism was the exclusion of melancholia from the group of the manic-depressed insane, and the unmethodical survey of the group of manic-depressed conditions. Various opinions questioned whether simple mania and depression, in view of the composition of the manic-depressive group, still have their justification as separate forms. Fewer opinions were expressed on Dementia praecox than on manic-depressive conditions, some critical, some disapproving, but generally covering different issues. There was harsh criticism regarding morphology and disease categories as well as mild suggestions for improvement. A positive general assessment of Kraepelin's work was completely lacking in the first years, as the academics were too caught up in discussing the validity of Kraepelin's theses. Although some psychiatrists had already recommended Kraepelin's work prior to 1914 and Soutzo, a psychiatrist in Bucharest, called Kraepelin's work the ‘quintessence’ of modern psychiatry,1 it was only in 1926 that Gaupp first made an attempt to integrate Kraepelin's work into an historical context at a respectful distance from the aforementioned points of conflict.2
The most frequent criticism was aimed at the group of manic-depressive disorders, and especially at the exclusion of melancholia from the group of manic-depressive illness. The first to address this topic was Thalbitzer, in the original Danish version of his doctoral thesis of 1902 and in the years 1905 and 1908 (German translation of his thesis).3–5 For Thalbitzer the isolation of circular depression from melancholia as being a psychosis of later years (involution melancholia) was unnatural and arbitrary. In 1908 Kraepelin abolished this delineation between age-dependent depression and his group of affective disorders. Thalbitzer deemed Kraepelin's two conditions of the manic-depressive group, mania with motoric excitation and depression with motoric arrest, as being too simplistic. Infinite combinations of the symptoms could be possible.5 He also stated that the differential diagnosis of melancholia as a true mood psychosis and ‘depressed insanity’ as a depression caused by delusions, had to be determined. Other authors also expressed the opinion that the transition between manic-depressive psychosis and melancholic psychosis was too smooth to divide,6 and that Dementia praecox also showed manic-depressive symptoms.7,8 A debate on this topic in the Berlin Society for Psychiatry and Nervous Diseases between Abraham, Dreyfus, Urstein and Forster in 1909 is referred to in the journal Neurologisches Centralblatt.8 Abraham was of the opinion that the mixed states had to be separated from the manic-depressive group. Dreyfus, former disciple of Kraepelin, replied that Kraepelin's categorization allowed many, ‘originally incomprehensible clinical pictures’ to be rendered understandable and precise prognoses to be made.
The exclusion of melancholia from the group of manic-depressive conditions caused difficulties in the classification of juvenile depression. Fauser, medical officer in Stuttgart, stated that the prognosis of a favorable depression in youth cannot always be grouped in the class of manic-depressive psychoses, Dementia praecox, hysteria or ‘degenerative insanity’, but belongs in the melancholia group.9 By contrast, Hübner, senior house officer at the Bonn University Hospital, claimed that juvenile depression should not be included in the class of involution melancholia.6
Overall, in the view of German critics, the group of manic-depressive conditions was too large and too complex,10 and was enlarged at the cost of Dementia praecox and contained invalid aspects.8
In France, there was an intensive debate against the background of the traditional concept of monomania and melancholia. But Kraepelin's dichotomy quickly had some supporters because of its simplicity and clarity compared to the fragmented traditional French psychiatric system.
Régis, in a discussion with colleagues, denied the given exclusiveness of manic-depressive psychosis, and stated that mania and depression should not appear as separate conditions but with their accompanying symptoms in the mixed states (‘état mixte’), composed of mania and depression.7 In his view, only 26% of all basic forms were recurrent, while Kraepelin assumed that nearly all were recurrent. For this reason, mania and depression had their justification as separate elements of the manic-depressive psychosis and, as opposed to Kraepelin's theory, could exist in simple forms (‘état simple’). Régis rejected Kraepelin's proposal that mania and depression are the same conditions, having the same ‘origin’ and ‘pathomechanism’. In his view, both belonged to the same family of conditions, but they were differentiable and even opposable. In this discussion, Deny contradicted Régis by stating that simple forms of mania and depression hardly exist. They occur as relapses of both forms in manic-depressive mixed states, and for this reason Kraepelin's expression of the manic-depressive disorder was justified.7
In comparison to the German discussion, which focused on the embedding of different formerly independent diseases in the manic-depressive group, the French reaction focused on the recurrent character of the manic-depressive disorder.
Opinions on Kraepelin's Dementia praecox also differed, but were much less prevalent than the opinions on manic-depressive disorder and tended to be more favorable. Bumke, Kraepelin's disciple and his successor at the Munich University Psychiatric Hospital in 1923, believed that Dementia praecox was too often falsely diagnosed in youth,10 while Walker, director of the Psychiatric Hospital Waldau near Berne, Switzerland, welcomed the fact that, due to Kraepelin's categorization, many cases that were earlier classified as mania, melancholia or circular psychoses were now attributed to Dementia praecox, because the melancholic or manic mood only represents a momentary situation, but the basic character shows ‘decay’ and ‘dulling’.11 Bleuler, Kraepelin's greatest opponent, at the time professor in Zurich and eponymist of schizophrenia with his book ‘Dementia praecox or group of Schizophrenias’, declared that the Dementia praecox group was a ‘true definition of illness’ with homogeneousness and clear delimitation.12
A few colleagues were already arguing about Kraepelin's morphology and his work in general. Hoche, professor of psychiatry in Freiburg, Germany, could not see any advantages of Kraepelin's teaching. With cynical and polemic commentaries he rejected Kraepelin's morphology, claiming it was diagnostically and prognostically unsatisfactory, with ‘numerous’ transitional forms and indistinct borders, and raised the question of entities.13,14 The search for pure types in psychiatry was invalid and would make the prognosis in numerous single situations more difficult, e.g. in diagnostic borderline conditions.15 Homburger summarized that Kraepelin's system lacked the causal connection of commencement, development and outcome of an illness.16 Jaspers challenged the basis of Kraepelin's system in his famous book ‘Allgemeine Psychopathologie’ (‘General Psychopathology’).17 Jaspers' anastomosis of psychiatry and philosophy still provides the direction for psychopathological differential diagnosis today and reveals criticism of Kraepelin's rigid dichotomy. He wondered whether Kraepelin had considered all aspects of the illnesses and the transitions between them in determining his categories:
The diagnosis of a general condition can only be made if one knows of a delimiting illness to be diagnosed. But one cannot find a sharply delimiting illness from the complete picture; only types which in single situations show ‘flowing borders’.
Jaspers saw a ‘remaining core of truth’ and the ‘intensification of diagnostic efforts’, but Kraepelin's illness entities were ‘no attainable assignment’ and just a point of orientation for future research. Jasper's fundamental criticism of Kraepelin's rigid system was unique in the German pre-war era and anticipated – more than Hoche's multidimensional approach – the current discussion on transnosological approaches.
The first period of Kraepelin's reception was interrupted by the First World War 1914–1918. During the war there was little debate on Kraepelin's teaching, especially not outside Germany. This fact was mentioned by Ion and Beer in the epilogue of their work ‘The British reaction to Dementia praecox 1893–1913’, because at that time psychiatrists were more occupied with treating the war traumatized than with engaging in fundamental debates.18 Ion and Beer estimate the year 1913 for the general circulation of Kraepelin's Dementia praecox concept in Great Britain.
The tone of the discussions of Kraepelin's work changed after 1918. It almost seems as if the harsh discussions on Kraepelin in the pre-war era had, at least in Germany, ‘dried out’ during the war. Hoff writes about these observations in his article ‘Psychiatrische Diagnostik: Emil Kraepelin und die ICD-10’ (‘Psychiatric diagnosis: Emil Kraepelin and ICD-10’):19
After the First World War and in the 1920's the originally significant ‘gravitational energy’ (to use a physical picture) of Kraepelin's teaching subdued. It could function less and less as a programmatic brace, which although being criticized in part, was basically accepted.
After the First World War until Kraepelin's death in 1926, Kretschmer and Meyer discussed general morphology and claimed that there was no necessity for an illness uniformity but rather for an ‘illness diversity’, because group isolation prevented the view to the ‘interaction of free emotional individual energies’.20 The monographic concept was useless and the dichotomy too rigid, not doing justice to the clinical course of illness, and therefore a more dynamic model was necessary.21 Kehrer, Head of the Psychiatric University Hospital in Münster, Germany, explained that the search for ‘Krankheitsformen’ (illness forms) was falsely based on one-sided criteria and that the results were not objectively evaluated.22 Only Gaupp, Kraepelin's disciple in Heidelberg and Munich and later professor in Tübingen, favored Kraepelin's concept, pointing out that, despite the deficiencies of the system, it had brought about progress and the opening of new research possibilities.2
The French reaction during this period was sparse and only concerned Dementia praecox. Halberstadt condemned the excessive enlargement of the Dementia praecox group at the cost of other groups,23 while Claude, psychiatrist at Sainte-Anne Hospital in Paris, saw an indistinct differentiation between paranoia and Dementia praecox.24
Again, as in the pre-war era, the German reaction from 1918 to 1926 covers the question about the legitimacy of the unidimensional nosological concept, whereas the French reaction refers to details of the grouping. Gaupp's commentary from 1926, still before Kraepelin's 70th birthday and his death, already points out the historic impact of his work.
In the second period 1926–1928, both commemorative publications for Kraepelin's 70th birthday appeared as well as the necrology, which was published soon after his unexpected death. Most of these contributions contain recognition of Kraepelin's work, of him personally and of his research methods.
The borders of manic-depressive disorder and Dementia praecox were again subject to discussion. Bumke, who in 1909 had warned of the blind acceptance of Kraepelin's theses, now stated that only Kraepelin's system of pre-defined illnesses allowed a distinction between the overlapping of manic-depressive and schizophrenic disorder.25 By contrast, Halberstadt found that pre-senility and the degenerative forms were wrongly classified in the manic-depressive group.26 There would be the necessity of providing sub-groups in the single groups. Meyer, a student of Kraepelin, later director of the New York State Hospital and professor at Johns Hopkins University, described a series of realistic differences.27 He was surprised at the pooling of mania and melancholia (without involutional melancholia) in the group of periodic, circular afflictions without deterioration and which generally only occur once in a lifetime. Claude considered the group of Dementia praecox to be too imprecise, especially regarding paranoia.28
Fundamental criticism concerned Kraepelin's nosological system. Henneberg questioned the binding of diagnosis and prognosis.29 The borders of the groups were too widely set and too vague. Meyer mentioned the lacking incorporation of the constitutional background into Kraepelin's research methods. Birnbaum, professor at the Charité University in Berlin and, following his emigration, lecturer at the New School for Social Research in New York, compared the popular schools of thought at the beginning of the 20th century and concluded that Kraepelin's work, although needing an enlargement of multi-dimensional criteria, is valid in its entirety as a diagnostic basis.30
Overall, the points of view in the second period after Kraepelin's death did not go beyond the well-known arguments of the pre-war era with regard to delineation of disease entities or dimensional approaches, and were even attenuated. Authors' descriptions seem to be affected with a certain resignation regarding the impact of Kraepelin's system.
Kraepelin's 100th birthday falls within the third period, 1929–1960. While the publications by Gruhle and Mayer-Gross, appearing 3 years after Kraepelin's death, highlighted the imperfections in Kraepelin's work and methods,31,32 publications written on the occasion of his centenary in 1956 were generally favorable and pointed out the historical perspective. Gruhle, a disciple of Kraepelin and later professor at Bonn University, and Mayer-Gross, professor in Heidelberg and, following his emigration, fellow in the Bethlem Royal Hospital in London and the Birmingham Medical School, criticized the lack of psychology in Kraepelin's research. Apart from this, Kraepelin was too strongly oriented towards clinical details without integrating these into a theory.32 De Boor, professor at the University of Cologne, explained that Kraepelin's endogenic group was only ‘a basic convention’.33 Each era of research should try to determine single forms. In the anniversary year of 1956, articles were published by Gruhle, Schneider and Kahn, as well as two articles by Wyrsch. Although Gruhle approved Kraepelin's schizophrenia, he made some minor comments on alcoholic hallucinosis and on the difference between paranoia and paranoid schizophrenia.34 However, Gruhle relativized these ‘special cases’ and welcomed that,
aside from all the small temporal inadequate single views this generous categorization of Kraepelin remains.
Schneider, author of the benchmark book ‘Klinische Psychopathologie’ (‘Clinical Psychopathology’) and director of the former Kaiser-Wilhelm-Institute for Psychiatric Research in Munich, which was founded by Kraepelin and financially supported by James Loeb, claimed that Kraepelin's work still remained valid, although it had lost much of its significance:35
The era of Kraepelin is not yet over. The posts set by him, stand. If these posts waver, it is not because they are brittle, but because they are elastic.
The only clearly formulated point of criticism concerned the unsuitable term ‘insanity’ in the description of psychiatric illnesses, since this term had a spate of ‘negative valuations’ associated with it and did not justify the ‘peculiarities of the schizophrenic or peculiar/characteristic personalities.’ Apart from this point, Schneider considered Kraepelin's work as the basis of modern diagnostics. Kahn, Kraepelin's disciple and later professor at the Baylor College of Medicine in Houston, Texas, praised Kraepelin's devotion to clinical research and called him ‘the master builder of psychiatry’.36 According to Wyrsch, director of the Berne University Hospital, the delineation of the schizophrenic and manic-depressed groups ‘stands somehow crooked and disassociated in the system’.37 All other groups – organic psychosis, arrested development, poisonings, cerebral lesions and psychogenic illnesses – are subject to the measurable realms of biology and psychometry, but not the manic-depressive group and the endogenic imbecility (‘endogene Verblödung’), inserted as a subgroup of Dementia praecox in the 8th edition of the ‘Lehrbuch’. Nevertheless, Wyrsch confidently emphasized the persistence of Kraepelin's work,
It still remains standing today without any dangerous fissures and seams and Kraepelin could end the matter in 1918 in his short history of psychiatry.
In his second article Wyrsch described Kraepelin's work as ‘historical’.38
Braceland, director of the Institute of Living in Hartford, expressed that Kraepelin's teaching was the basis of modern diagnostics.39 The influence of research based on Kraepelin's teachings had extended to the areas of institutional psychiatry, statistics, anatomy and genetics.
Kahn again emphasized that there had been no major changes regarding the question of endogenic psychoses since Kraepelin, and that Kraepelin's research results had been confirmed.40 The publications appearing on and following Kraepelin's centenary represent Kraepelin's teaching as generally being the fundament of modern psychiatry. One exception was Conrad, director of the Psychiatric University Hospital in Göttingen, Germany, who demanded a basic revision of the psychiatric classification.41 He especially attacked Kraepelin's dichotomy because of the sharp distinction between manic-depressive disorder and schizophrenia, which in clinical observation is not valid.
In conclusion, in the late third period of Kraepelin's reception, his work was set in an historical perspective without any hard criticism. Although most authors describe some inconsistencies in the group delineation, mostly known from the pre-war era of the first period, the harsh criticism of Kraepelin's scientific methods, his dimensional approach and his inflexible dichotomy – the main target in the second period – was mentioned less often. On the contrary: authors seemed to be content to use a diagnostic system that offered easy handling.
Modern research on Kraepelin's ideas, beginning in the late 1980s, focused on the absence of dependable statistics and his subjective hypotheses. Sheperd called this a ‘categorical mistake’.42,43 Due to these insufficiencies, various counter-models appeared soon after Kraepelin's nosology became known. Sheperd mentioned those of Wernicke,44 Hoche,15 Bumke,45 Jaspers,17 and Meyer.21,46 However, these authors took a different approach than did Sheperd, in that some presented their own psychopathological models (Wernicke, Jaspers) or they represented opposing views on single points of Kraepelin's dichotomy (Bumke, Meyer, and to some extent Hoche). In the early years after the appearance of Kraepelin's ‘Lehrbuch’, criticism was seldom based on the fundamental questions of method, statistical objectivity or scientific assumptions but on the discussion of Kraepelin's categories and summaries of diagnoses as well as the comparison of study data. Almost no one asked whether Kraepelin's measures were even measurable, or whether his clinical examples were based on objective testing criteria, or whether the analysis of results occurred without prejudice.
While a heated discussion on the contents and effects of Kraepelin's dichotomy raged after 1900, incorporating most of the contributions up to the First World War, a discussion of the methodological basis of Kraepelin's nosology was practically non-existent. One exception was Homburger who claimed early on that Kraepelin's teaching could only measure ‘mental based on the mental’.18 After the First World War this situation changed, and the ‘gravitational energy’ of Kraepelin's teaching deteriorated.19
The discussion of Kraepelin's scientific understanding gained importance after 1920, certainly due to the emerging psychopathological teachings by Jaspers, Kretschmer and Freud, while the criticism of Kraepelin's dichotomy decreased. Meyer rejected the rigid, biased categorization of the groups of illnesses.21 These were useless for diagnosis and prognosis and should be replaced by ‘dynamic formulations’. Gaupp described Kraepelin's research methods as being ‘half consciously and half unconsciously biased and incomplete’, being ‘too concerned with stocktaking’.2 Gaupp demanded the inclusion of constitutive biology in research, to do justice to atypical descriptions. Gruhle complained about the lack of psychological research in Kraepelin's psychiatry and called his working methods naïve and unconcerned with scientific thinking.31 Mayer-Gross blamed Kraepelin's anatomical viewpoint and his inability to see a multidimensional psychiatry.32
Recent discussion has focused on the limitations of dichotomy and the conceptual difficulties for the trans-syndromal approaches in the future classifications of ICD-11 and DSM-V. The deconstruction of the Kraepelinian categorical classification with two main groups of disorders towards a syndromal descriptive classification is based on results of recent research, especially genetic research, which showed similar results in formerly different disorders and led to transnosological thinking. Craddock and Owen suggest a new classification that takes into consideration the biological roots of schizophrenia and bipolar disorder.47 In their view, it is difficult to integrate the psychopathological varieties of schizoaffective disorders into a dichotomous classification, due to a genetic overlap between schizophrenia and bipolar disorder. Möller contradicts this opinion by pointing out that this genetic overlap will continue in syndromes and symptoms that are also based on common genetics.48 By deconstructing the dichotomous classification, genetic overlaps on a syndrome or symptom level would rather blur than clarify psychiatric diagnosis. Jablensky has already described how Kraepelin revised his system because of the comments of his critics, thus producing a model that also satisfies the modern demands of neuropathology and genetics.49 Kraepelin himself was aware of the bias of disease entities and the difficulties of classifying borderline cases, but in his opinion a mere syndromatology – as suggested by Hoche – seemed too simple for establishing complex psychopathological coherences.
In conclusion, today's questions on Kraepelin's system are similar to those in the years 1900–1920 regarding the exclusiveness of the dichotomous classification between manic-depressive disorder and schizophrenia. However, Kraepelin's contemporaries found counter-arguments in their own clinical data, whereas today's debate is based on research in neuroimaging and genetics. Interestingly, in the last two decades, several researchers have shown a resignation, similar to the authors of the obituaries, e.g. Berrios and Hauser.50 They complain of the inability to escape the ‘blinding embrace’ of Kraepelin's system: the categorical structure still influences diagnosis and prognosis in a rigid manner, but there seems to be no alternative classification system that could replace the simple, and for that reason embracing, Kraepelinian system.
Finally, all authors describe difficulties in grouping-in overlapping syndromes, but some of the formerly unclear diagnoses seem to have less importance today, e.g. involution melancholia, juvenile depression and endogenic imbecility. This might be due to the obsolete notion of some special forms, but could also be considered a marginal question of low importance, compared to the problem of a large group of disorders standing between the two Kraepelinian columns. In the last years of his life, Kraepelin himself was aware of this problem on which psychiatrists are debating again today, since the reappraisal of dichotomy in DSM-III in 1980. Also in DSM-IV, the Kraepelinian influence continued and comments cautiously addressed the question of reliability and validity of a dichotomous diagnostic system. For future classifications, modern research takes into account overlapping genetic linkage regions for schizophrenia and affective disorders and provides arguments for a syndromal or continuum approach. Integrating these unclear cases and overlapping syndromes, e.g. schizoaffective disorders, into a dichotomous or trans-syndromal approach still is and will be the main focus of nosological research. On the other hand, the lack of psychology and multidimensional approaches – the focus of criticism in the 1920s by Kretschmer and Meyer – is not the key problem of modern nosology, because they are implemented by psychological and biopsychosocial findings since the 1950s and the psychodynamic movement in the USA in the 1960s and 1970s and are represented in the DSM-IV axes. This led to a more holistic understanding of patients and their disorder, contrasting to Kraepelin's static formulation of ‘mental illness’. From today's perspective, Hoche's and Jaspers' discussion of disease entities versus syndromatology seems to be the remaining question on Kraepelin's nosology. Kraepelin established a system with a variety of flaws, but it was a simple and clinically applicable system that allowed distinct diagnoses instead of only syndromal description. For this reason it is still in use and one might claim, not for its superiority but for its simplicity.