The problem with writing a book about a subject that sits on the interface between science and the humanities is that you will likely either not satisfy either side, or you will satisfy one side and not the other. So it is with Obsession: A History, by Lennard J. Davis, University of Chicago Press, 2008.
In modern clinical psychiatry, the essential features of obsessive-compulsive disorder (OCD) are recurrent obsessions or compulsions sufficiently severe to cause marked distress to the individual.
Examine the detail: "...sufficiently severe to cause marked distress..."
Clearly, if your obsession (the novel you're writing or your current research project or your ballet practice) gives you pleasure directly or indirectly (by providing you with an income) you are of no interest to clinical psychiatry.
There is no useful point in blurring the distinction between obsession as a clinical diagnostic entity in psychiatry and obsession as a description for certain kinds of behaviors that can be extremely productive. The clinical entity, OCD, involves serious dysfunction, and in fact nearly 70 percent of OCD patients sooner or later suffer major depression and become even more dysfunctional.
Professor Davis, who teaches at the University of Illinois at Chicago and whose major specialty is apparently English Literature, navigates an intellectual minefield is his attempt to conflate clinical psychiatry and the humanities. He's unhappy with how clinicians and neuroscientists deal with obsessive-compulsive disorder. He wants more attention paid to history and literary allusions to obsessive behavior and obsessive characters. He wants less reductionism in attempts to understand OCD. In other words, he wants more humanities in science. That sounds lovely, but for the clinician and scientist, the problem is that scientists and clinicians hardly ever find the humanities a useful backdrop in their daily sweat to understand how the brain works when it's healthy and not healthy. As for medicine, it's true that psychiatric categories are merely labels for clusters of symptoms, but it's also true that psychiatric categories of some kind are absolutely necessary in the clinic as a guide for what sort of behavior to expect from a patient. A psychiatric label is simply a practical device to assist in treatment. Unfortunately, these labels are too often misunderstood, misused, and bandied about in public as buzz-words -- but usually not by clinicians and scientists.
Any book on this subject is a difficult project, and each reader will have a positive or negative response depending on their own intellectual context. This book is a literate look at a humanities-science interface -- but maybe from only one side of the fence between the two cultures. Yes, the fence is still there, and after all these years it may even be more formidable.