From: Joscha Bach Sent: Friday, November 1, 2013 6:31 AM To: Ari Gesher Cc: Greg Borenstein; Sebastian Seung; Joi Ito; takashi ikegami; Kevin Slavin; Martin Nowak; Jeffrey Epstein Subject: Re: MDF Ari, sorry for the slight delay until I got around to answering your latest =ail. The discussion is very interesting and inspiring to me! With respect to our original discussion of intelligence with respect to =ooperation, competition and deception, it is mostly tangential (all =ecipients, please be warned, we spun off towards the metaphysics of =1). > While I think the notion of functionalism stands as a thought > =xperiment, building equivalent systems that "perform exactly the same =unction" as the original is pretty elusive. I remember hearing from a =esearcher at UW trying to build a mechanical finger to study human =ovement for cybernetic purposes. (...) The value I see in the essentialist approach is that natural, evolved =ystems use all the subtlety, the complexity of their medium. The neat =pproach keeps trying to add complexity until epsilon hits zero on its =ntegral. But isn't a zero epsilon actually asymptotical outside of the =lean confines of math? Or at least elusive (in this arena) until we =ctually understand what level of physical reality that neurons are =ssentially operating in? I do not think of functionalism as a recipe to build something, but as =n epistemological position. Functionalism recognizes that we construct =ur concepts (including the concept of mind and intelligence) based on =hat things do, not on what they 'essentially' are. A mind is not an =ntrinsic power animated by a soul with no empirical properties, but a =ausal arrangement that processes information in such a way that it is =ble to participate in discourse, control a body, reflect upon itself, =ind creative solutions, imagine and dream, and so on. I do not suggest =o reduce any of these properties away, but to focus on the right level. This level is not