It's great when someone replies as thoughtfully as this, and I love your phrase "gently disagree". Would that more disagreements were of this kind.
> Well, for something like bipolar disorder this feels like it's true.
I agree. The open question is on what level these things are best to be understood. In some sense everything about us is biology, just as in some sense everything is physics. But we don't think of, say, heart disease as best addressed by physics. Similarly, that there is a biological stratum to our actions and feelings does not automatically mean that chronic emotional suffering is best understood as biology. One might as well conclude from the tongue and larynx that language is biology too. And indeed it is, sort of—yet its meaning lies elsewhere. Or you might as well conclude from the importance of neurons to learning that education is brain biology. Why bother with teachers or schools? We should just 'learn' the neurons directly. Such examples are obviously silly, at least given our current knowledge.
To know that chronic emotional suffering is a biological disorder requires more than the involvement of biological phenomena in it; it requires an experimentally verified model. My understanding is that we don't have anything close to that. The one that entered the public imagination, the serotonin-deficiency theory, is widely dismissed by experts [1,2,3,4]. The only argument seems to be whether they knew it was false from the beginning or discovered that it was false decades ago.
Yet we insist that modern science has discovered that mental illness is biological just the way that cancer is and so on. What do you call a conclusion like that which goes far beyond what we actually know?
It's worth realizing that psychiatry has always made this claim. The grounds for it shift every 20 years or so, and the previous grounds are always dismissed as ludicrous if not harmful (think lobotomies etc.). Nevertheless we're perpetually certain that we're beyond all that now.
> But I know that people might get benefit from meds
I don't think anyone questions that; the question is whether the benefit is that of a placebo or not. My understanding is that SSRIs are no more effective a treatment for depression than tricyclics were a generation earlier; their advantage is rather that they have fewer side effects [5]. So whatever explanation there is for their efficacy must plausibly explain how both of those (presumably very different?) biochemical mechanisms could do it. Given that even inert placebos produce most of the same effect (Kirsch's finding), the placebo explanation is pretty clearly a major candidate. If there's another, I'd like to know about it.
(I realize you were talking primarily about bipolar disorder, but I haven't read about that, so I've continued to talk about depression instead. That may lessen the relevance of the above.)
One of the problems with describing an illness as psychological is that people think we are dismissing their illness, that we are saying the illness is not real. They think we are either saying that it's all in their heads, or that it's not serious, or that it's just a matter of them needing to "pull their socks up".
I think that might be why people cling to a biological model for mental health problems.
Obviously, we're not saying any of that. I say that there are powerful talking therapies; that these are evidence based and effective for many people; that medication may help although it's probably over-prescribed and it can have unpleasant side-effects; and that some people won't respond to any of that and may need electro-convulsive therapy or other severe interventions.
I agree that people tend to overstate the biological model, and that is a problem. It's a problem because, as you say, we don't know if it's true, and it's a problem because it steers people away from talking therapies.
> Well, for something like bipolar disorder this feels like it's true.
I agree. The open question is on what level these things are best to be understood. In some sense everything about us is biology, just as in some sense everything is physics. But we don't think of, say, heart disease as best addressed by physics. Similarly, that there is a biological stratum to our actions and feelings does not automatically mean that chronic emotional suffering is best understood as biology. One might as well conclude from the tongue and larynx that language is biology too. And indeed it is, sort of—yet its meaning lies elsewhere. Or you might as well conclude from the importance of neurons to learning that education is brain biology. Why bother with teachers or schools? We should just 'learn' the neurons directly. Such examples are obviously silly, at least given our current knowledge.
To know that chronic emotional suffering is a biological disorder requires more than the involvement of biological phenomena in it; it requires an experimentally verified model. My understanding is that we don't have anything close to that. The one that entered the public imagination, the serotonin-deficiency theory, is widely dismissed by experts [1,2,3,4]. The only argument seems to be whether they knew it was false from the beginning or discovered that it was false decades ago.
Yet we insist that modern science has discovered that mental illness is biological just the way that cancer is and so on. What do you call a conclusion like that which goes far beyond what we actually know?
It's worth realizing that psychiatry has always made this claim. The grounds for it shift every 20 years or so, and the previous grounds are always dismissed as ludicrous if not harmful (think lobotomies etc.). Nevertheless we're perpetually certain that we're beyond all that now.
> But I know that people might get benefit from meds
I don't think anyone questions that; the question is whether the benefit is that of a placebo or not. My understanding is that SSRIs are no more effective a treatment for depression than tricyclics were a generation earlier; their advantage is rather that they have fewer side effects [5]. So whatever explanation there is for their efficacy must plausibly explain how both of those (presumably very different?) biochemical mechanisms could do it. Given that even inert placebos produce most of the same effect (Kirsch's finding), the placebo explanation is pretty clearly a major candidate. If there's another, I'd like to know about it.
(I realize you were talking primarily about bipolar disorder, but I haven't read about that, so I've continued to talk about depression instead. That may lessen the relevance of the above.)
[1] http://www.npr.org/blogs/health/2012/01/23/145525853/when-it...
[2] http://www.plosmedicine.org/article/info:doi/10.1371/journal..., via http://www.plosmedicine.org/article/info:doi/10.1371/journal...
[3] http://www.psychiatrictimes.com/blogs/couch-crisis/psychiatr..., paywall bypassable by clicking link at http://goo.gl/3LxUW
[4] http://www.madinamerica.com/2012/01/revising-the-history-of-...
[5] http://www.ncbi.nlm.nih.gov/pubmed/17636689. This looks like it was withdrawn because it was to be superseded by a larger study, but I couldn't find that one.