> This is a reasonable hypothesis held I believe by some but not all psychiatrists/neuroscientists. But there is no real or dependable knowledge on any form of mental illness. In the past, and perhaps still, some psychiatrists suggested that there is no such thing as depression, there are only bipolar patients who have never had a manic episode. In any case, the current typology of mental illness almost certainly will change. No current diagnosis is much more than an educated guess.
I suspect you're thinking of Thomas Szasz' "The Myth of Mental Illness," or others likely influenced by his thought or that book. His position is often misunderstood. He never disputed the reality of any of those experiences of misery, maladjustment or disorientation that are commonly identified as "mental illness." His point was rather that "mental illness" is a category mistake: an unjustified extension (and not an innocent extension; more on this below) of a category that makes perfect sense in the context of (physical) medicine: a context in which the existence of a pathogenic agent is assumed, whether it has yet been successfully identified or not.
> Until something bettger comes along, SSRIs, vicious as they are in some ways, are probably a useful placebo that helps many to trudge along with their lives.
SSRIs fucking rock. But then so does heroin. At least for a while. It's been a while since I looked at the research, but I think SSRIs have been shown pretty conclusively in many credible, double-blind studies to have real effects at the chemical level - i.e., not just placebo. Long term effects? Fuck knows. Having worked in a psych clinic years ago, I can also assure you their prescription - to anyone where a plausible case for them can be made - is amply incentivised.
More broadly, although there's convincing evidence for various genetic predispositions to various kinds of suffering, dysfunction or struggle, it often gets elided that those events of suffering, dysfunction or struggle - the extreme versions of which used to be called "nervous breakdowns" - occur in history, and the people who suffer them almost *always* associate them with material facts in the world, if given a chance and a sympathetic ear. There are exceptions: e.g., where there is demonstrable injury to the brain, the adrenal system, etc. But those are precisely the cases where Szasz would say (correctly) that it *is* a question of disease - disease that turns out to be *somatic*. There is bodily illness, and there is mental torment. "Mental illness" is simply a category mistake.
The *political* edge of Szasz' thought is that it's not an *innocent* category mistake, but one that lends itself to exploitation by - and is exploited by - the bureaucratic institutions of psychiatry, and allied institutions of behaviour management and incarceration, all of which serve capital. If I'm a psychiatrist, it's a category mistake I can use to deprive you of your agency and rights. With the very best of intentions, of course. "Trust me, this will be good for you. I'm an *eeexpert*."
For anyone interested in the political dimension of Szasz' thought, here's the blurb from another of his books, "Law, Property and Psychiatry":
"Psychiatric interventions in the administration of civil law regulating the individual's right to manage his person and property serve two often contradictory aims: to protect the ward and maximize his material benefits, as defined by individuals other than the ward himself, and to protect the potential ward's civil rights and to minimize abridgments of his freedom and powers to make independent decisions as defined by himself. It is the view of the authors that the potential ward's freedom and right to self-determination should be recognized as more important values than the protection, even against his will, of his person and property."
<http://onlinelibrary.wiley.com/doi/10.1111/j.1939-0025.1972.tb02527.x/abstract>
Okay, maybe not classically Marxist, but....