Thomas A. Ban
Neuropsychopharmacology in Historical Perspective.

Education in the field in the post- psychopharmacology era. Te Wernicke - Kleit - Leonhard

Tradition with Special Reference to Mania, Melancholiaand Manic- Depressive Psychosis

Collated 7

Nassir Ghaemi’s comments


        Thomas Ban’s review of the Wernicke-Kleist-Leonhard (WKL) tradition brings us to central conflicts in the diagnostic assumptions of 19th and 20th century psychiatry in Germany and elsewhere. 

        We never think in a vacuum but are influenced unconsciously by our world around us.  In 2019, almost 40 years after DSM-III codified a version of Leonhard’s bipolar-unipolar distinction, many readers will have difficulty with any criticism of Leonhard. He is not a household name in the US, but his ideas are accepted unconsciously. He has become, like Freud in the past, a climate of opinion. 

        Contrary to popular belief, DSM-III was not neo-Kraepelinian. It was neo-Leonhardian. As Ban clearly shows, the bipolar-unipolar distinction is all about the WKL tradition and it completely contradicts Kraepelin’s basic view.  Contrary to what one often hears from contemporary Germanic psychiatrists, Kraepelin was not sympathetic at all to such radical rejections of his basic nosology, nor was it a later development. As noted, Kleist attacked Kraepelin’s scheme in 1911, and likely earlier. 

        The whole approach of Kraepelin and Wernicke are exact conceptual opposites. Wernicke was an expert in brain pathology; he came to clinical syndromes from the brain outwards (just as with the neo-Wernickian NIMH RDoC approach).  He had excellent and correct anatomic localizations of the brain; he then took what he knew around 1900 and generalized and speculated extensively from it so as to produce the nosology of his textbook. It was out-of-date when it was printed, much less a century later, as always is the case with such “neuromythology,” as Karl Jaspers called it. Such work is pseudoscience: it looks like science – with nice anatomic pictures and these days colorful brain MRI scans -  but it isn’t correct in the end.  There is an ancient basis for this way of thinking in medicine: it’s based in Galen’s approach, theorizing from biological concepts (four humors) to clinical constructs. That approach ruled medicine for two millennia and held it in place, backward, unscientific and damaging to sick patients (bleeding and purging were the consequence of the theory).

        Kraepelin took the opposite approach, from the clinical syndromes into the brain and body. He was honest enough to admit that we didn’t know the etiology in the brain in 1900, and still mainly don’t, and so he tried to describe clinical features as parsimoniously and empirically as possible, based on real-world objective features (especially course and heredity, as well as physical examination, not just symptoms), not biological speculation.  This attitude also is ancient in medicine, advocated by Hippocrates and his modern resuscitators (Sydenham, William Osler) who had to fight so hard to get doctors to give up the religion of four humors and stop bleeding their patients. In medicine, this clinical syndrome approach has proven very successful scientifically in the last century, after Osler finally slayed the ghost of Galen, but in psychiatry, we have retrogressed.

        We now have a false diagnostic system, DSM-III to 5, based on an unconscious postmodernist, social constructionist ideology held by its creators in the 1970s.  The leadership of DSM-IV has been most explicit: they admit that the main basis for DSM criteria is not science, but “pragmatism,” meaning pure utilitarianism, what is good for doctors, or patients, or the profession – or at least what the American Psychiatric Association (APA) DSM committee leadership thinks  – is best for everyone else.  Why should God or nature so design genes or brain structures to correspond to the diagnostic wishes of APA committees? DSM is false and that’s why our biological, genetic and pharmacological research has failed so much since 1980. It was doomed to fail. It was guaranteed. 

        We come back to Leonhard. Yes, there was some research that supported his views, specifically three studies in the late 1960s as cited by Ban (Angst 1966; Perris 1966; Winokur, Clayton and Reich 1969) and a few more in the 1970s. This was the thin reed of science on which the radical rejection of Kraepelin’s manic-depressive illness (MDI) happened in 1980 with DSM-III.  The claim was that the genetics of bipolar and unipolar patients differed, as did their course, with bipolar illness happening earlier in life (mean age around 20 years) and unipolar later (mean age around 30 years).  But we now have four decades more research that has not confirmed those findings and mostly contradicts them. Modern genetic studies clearly do not find a difference between bipolar illness and so-called “major depressive disorder” (another DSM invention), if the latter is to be a stand-in for “unipolar” manic-depressive illness.  There clearly is great genetic overlap, which would argue in favor of Kraepelin’s MDI and strongly against DSM’s neo-Leonhardian bipolar/MDD dichotomy.  On course of illness, if neo-Leonhardianism is correct, we should not see unipolar depressed children. And yet the diagnosis of “MDD” in children is rampant.  In contrast, mainstream psychiatry today is very critical of diagnosing bipolar illness in children, not realizing that the whole concept of bipolar illness is based in part on the view that it should happen early in life, unlike unipolar depression.  If we accept that the profession is not now completely deluded, and that unipolar depression does happen frequently in children and young adults, then again, this course evidence argues against Leonhard and for Kraepelin.

        The strongest argument is perhaps in Jules Angst’s own work, the Zurich cohort, which he started in 1959.  At the time of his famous 1966 paper, he had about half a decade of data. These early data were used for the radical rejection of Kraepelin’s MDI in 1980 with DSM-III.  In the 1980s and 1990s, Angst continued to follow his patients and he found that the clear distinctions began to go away. The course and genetics of the bipolar versus unipolar subjects began to merge. As a good scientist, he began to walk back his theory and argued for a broader bipolar spectrum, and a unipolar spectrum, both of which merged with each other – a perspective that begins to get close to Kraepelin’s MDI concept.  Angst proposed these changes for DSM-IV and DSM-5, and I led the International Society for Bipolar Disorders (ISBD) 2008 Diagnostic Guidelines Task Force where a consensus of world researchers supported Angst’s proposal. It was never even seriously considered by DSM-5.  Here’s the lesson: DSM-III made radical Leonhardian changes based on limited data but refused to make smaller changes decades later with much larger data, using literally the same database!

        In sum, I agree with Donald Klein, but not for the same reasons. He argues against Leonhard’s complex schema because it is not useful, compared to Kraepelin’s simpler nosology. I argue against Leonhard’s tradition, and its baneful effect in DSM-III to 5, because it is unscientific, Galenic, speculative and mostly disproven.  In contrast, poor old Kraepelin, the king of the hill that everyone likes to knock down, is in the most successful scientific tradition of medicine and his work continues to be supported with our most recent research even in genetics. 

        The relevance of all this history of psychiatry would be if it would change our nosology to accord with our best research – and not every 100 years, nor every 20 years with each DSM revision. In fact, though, the radical neo-Leonhardian core of DSM nosology of mood illness has not changed in substance since 1980.  We have stood still, because the dictionary became a Bible, a book to be revered not revised, the modern ideology of American psychiatry, unconsciously worshipping Leonhard and his tradition. 



Angst J. Zur Ätiologie und Nosologie endogener depressiver Psychosen. Eine genetische, soziologische und klinische Studie. Berlin/Heidelberg/ New York: Karger; 1966.

Perris C. A study of bipolar (manic-depressive) and unipolar recurrent depressive psychoses. Acta Psychiatr Scand 1966; 42: 1-189.

Winokur G, Clayton P, Reich T. Manic Depressive Illness. Saint Louis; Mosby: 1969.


May 7, 2020