Output Stream of Consciousness

I have a small passel of graphology books. They were fun, especially when I was younger and wanted to see whether my peculiarities of handwriting were "good" or "bad", and was indeed happy with the bevy of "good" signs.

Revisiting the craft, though, the claims seem puffed up, and worse, in a way, is that it is deeply steeped (the material I have, at least) in simplistic Freudian terms: the upper zone is the superego, the middle zone the ego, the lower zone the id, left-leaning is maternal influences, and lots and lots of sexual issues.

When I went poking around looking for an evidence-based graphology, I got by and large two things: places (e.g. here) that purported graphology to be evidence-based, but that looked quacky (e.g. it's evidence-based, but just for fun you could find out if your wife is pregnant with a boy or a girl?!), and those (e.g. here) that skeptical of the whole enterprise with links to some studies of existing graphology methods that showed them in rather a bad light.

I'm inclined to think that there are bucketloads of chaff in graphological methods. There seem to be some general correspondences, but they are the sort of thing that the untrained eye can certainly see.

If current graphology methods are bogus, it would be nice to be able to start from the ground up and get some evidence-based correlation checks going on.

Not just taking current methods and seeing whether they stack up. Start from scratch.

Correlating questionnaires and measurable writing characteristics may find something, little or nothing.

Either we fix graphology or consign it to the entertainment heap along with the other forms of fortune-telling.