We prove the following descriptive set-theoretic analogue of a Theorem of R.O. Davies: Every Σ2 function f : R× R → R can be represented as a sum of rectangular Σ2 functions if and only if all reals are constructible.

has provoked a lot of interest and comment. For those who haven't read it, how would you summarise the argument? In a nutshell, I argue that psychiatry over the past 40 years, under the dominance of the medical or 'technological' model, has done a lot of harm in the name of helping vulnerable people. Not intentionally, I hasten to add, but as an outcome of taking the medical model too far. Your...