nigelswift wrote: corroborate=confirm.....
But "confirm" is a problematic term isn't it? Does one result corroborate another? Or does it need ten? And can "similar" be cited as confirmation?
What about double blind testing? Why, since Mesmer's claims of animal magnetism were tested in that way in 1784, have many generations of dowsers maintained that self-confirmation or colleague-confirmation are enough and the only true corroboration?
All science has that same problem , which affects everyone who reasons inductively , i.e. generalising from the particular and the assumption that there is an underlying unity of nature .
There was a simplistic attempt at presenting that case here recently , by a defender of dowsing , sadly that type of defence does our cause more harm than good .
|