"Being sane in insane places", the 1973 Rosenhan experiment, concluded "we cannot distinguish the sane from the insane in psychiatric hospitals" http://en.wikipedia.org/wiki/Rosenhan_experiment
Your argument is highly misleading, First, Rosenhan was commenting on the state of psychiatry at the time rather than making a universal statement. Second, the Diagnostic and Statistical Manual of Mental Disorders (DSM)--today regarded as the Bible of psychiatry--was not written yet, meaning that psychiatry was at a state where repeatable metrics had not been defined and ratified for everyday use by medical professionals. In fact, Rosenhan's experiment was one of the catalysts that led to the creation of the DSM. Your comment supports a sentiment that is vastly outdated and no longer supported by the current state of psychiatric research.
"Your comment supports a sentiment that is vastly outdated and no longer supported by the current state of psychiatric research."
[Citation needed]
I am no psychiatrist, but from what I've read, the current consensus indicates that psychiatric orders can be detected using reasonable criteria. The widespread use of the DSM demonstrates this idea. The DSM is occasionally updated, and each revision reflects the changing ideas about psychiatry.
Source: http://www.nytimes.com/2012/05/12/us/dsm-revisions-may-sharp...
40 years ago psychiatry was very different than it is today.
>It has a long history of shifting its positions, and a long history of debunked and discredited bodies of theory
This is not a bad thing on its own. It just means the search space is large and complex with many potholes and convergence will take a while yet.
> Don't you think that, with a few weeks of study, I could apply any subset of mental disorders from the DSM-IV to any person I wished?
No I don't think so. Not in any substantial way that couldn't just as accurately be replicated by a markov chain with the specificity of a fortune cookie. You can't just trivialize an entire branch of study like that. Certainly there is a lot of room for improvement and terms like disorder, psychopath, multiple personality, schizophrenic are abused and misused but there are behavioural patterns and characteristics by which people can be clustered. The mistake most people make is to think these clusters are static and disjoint.
Your ideas on Psychology are outdated. What Freud believed is nonsense and without experimental feedback he could do little better. Things are better these days, there is cross-talk between areas like Machine Learning, Psychology and neurobiology. Here is a better example of what the future of Psychology will be like: http://videolectures.net/icml09_niv_tnorl/
Also: http://www.med.wisc.edu/news-events/news/psychopaths-brains-...