It's an interesting issue: they did have a depression sub-type in DSM-IV called 'melancholic depression': https://en.wikipedia.org/wiki/Melancholic_depression h

It's an interesting issue: they did have a depression sub-type in DSM-IV called 'melancholic depression': https://en.wikipedia.org/wiki/Melancholic_depression https://en.wikipedia.org/wiki/Melancholic_depression At any rate, at least it still seems to be of interests to 'clinical academics' in some prominent mental health research institutes (e.g. http://www.blackdoginstitute.org.au/healthprofessionals/depression/ourmodelofdepression/melancholicdepression.cfm http://www.blackdoginstitute.org.au/healthprofessionals/depr...). The entire premise of the DSM is a bit of a crock: mental illness simply doesn't fit well into the disease model of illness. Mental dysfunction is not binary, it is measurable on a continuous scale. And although some symptoms do present in certain 'groups' more frequently than others, symptoms of mental illness are more 'mix and match' than the DSM would lead one to believe. This sometimes means that people are either diagnosed with 2-3 'co-morbid' disorders or none at all.