Liberian native physical medical treatments, according to Dr. George Harley, could be roughly divided into 3 groups: "it helps," "it does nothing" (a lot of magical stuff was here), and "it causes damage." I don't have the book any more (I read it 40 years ago), but I vaguely recall that he thought the number of treatments in the 3 groups were roughly equal. The snake society (to which most hunters belonged) had a few rituals and a magic preparation that involved powdered snake heads rubbed into scratches. Harley didn't attempt to test this, but as the link suggests it isn't obviously wrong.
We have quite a few people searching through primitive pharmacopeia for useful drugs.
Mental illness is a bit harder. I see things like "mental illness was thought to be caused by evil spirits," but that doesn't tell me much about treatments. "Magic(*)" OK, but suppose instead of trying to lump treatments we get more specific. Did they distinguish different classes of problems and do their categories correspond to ours at all? What exactly did they do to try to help, and did it help?
I assume they had the pragmatic triage categories: "You are a witch and dangerous to the rest of us. Die.", "We cannot live with you. Go away.", and "Nobody minds Ben Gunn." We, with more resources, institutionalize the first, expel the second (modulo homeless shelters), and try to help (sometimes stupidly) the third group.
Were there any other distinctions?
What did the witch doctors do to help the third group (and those in the other groups prior to the tribe's judgment)? How did they decide on treatments? And most importantly--did the treatments ever help? Figuring that out takes a lot longer than grabbing a handful of bark and flying back to the lab, unfortunately. It means somebody has to spend a long time in the village.
Dalrymple says that unhappiness is being medicalized into "depression," with the obvious consequences that unhappiness is not properly addressed and depression is trivialized. But the description from India of How to Treat Depression When Psychiatrists Are Scarce leaves me wondering if palliation for both can be similar (I am not a psychiatrist!), at least for mild depression. See AVI comment below
I can easily imagine treatments that only work within the psychological environment of the village with its network of obligations and meanings. Even something as simple as a cola nut has a set of uses in a Liberian village having to do with relaxation, welcome, and worship, that would not translate well to anything I can think of in the US--and I only know a little from the outside. So a treatment that works there might not work here.
But if something does work, it would be good to know how and why.
Probably there's already a treatise or three on the subject and I just haven't found it.
(*) Our categories are also apt to square-peg round-holers. Practitioners can give examples, but any framework trying to describe the borderlands of personal responsibility is going to have misfeatures.