As a depressive, I can tell you the experience is certainly very real, even if much of psychology is on shaky ground, and it's certainly not a pleasant or healt

As a depressive, I can tell you the experience is certainly very real, even if much of psychology is on shaky ground, and it's certainly not a pleasant or healthy state of mind. Are there not relatively good quality studies demonstrating certain treatments such as CBT are effective in treating depression + anxiety? And surely a schizophrenic who hears voices and is detached from reality can be said to be mentally unwell? And aren't their studies demonstrating anti-psychotic pharmaceuticals are effective in those cases? Though I am all for data-based scientific reasoning in psychology, I do think there's a place for practical, empirical techniques to treat things we don't quite understand yet (and perhaps don't hope to understand for some time, i.e. a detailed understanding of the operation of the brain). When people have mental health issues they need help now, not in X years when we have sufficient understanding to make it a harder science. I often wonder about the disdain those of the harder sciences have for the 'softer' ones. Physics, chemistry + friends have the wonderful advantage of being able to examine phenomena in the small, once you examine things in the large (such as medical studies, or sociological behaviour, etc.) there are often sufficient factors to make it analytically intractable + require fuzzy uncertain statistical techniques. Does that approach make a science 'soft' in your opinion? I am not for the big fuzzy Freudian unfalsifiable theories that thus can't be tested by the way. Rather I think a lot of big problems require imperfect statistical techniques now which can be improved with a deeper understand later.