On Thursday, August 30, 2012 7:38:27 PM UTC-4, Brent wrote:
>  On 8/30/2012 2:19 PM, Craig Weinberg wrote:
>>  Computational analogies can only provide us with a toy model of 
> morality.  Should I eat my children, or should I order a pizza? It depends 
> on the anticipation of statistical probabilities, etc...no different than 
> how the equilibrium of oxygen and CO2 in my blood determines whether I 
> inhale or exhale.
> It also depends on what you want.  No decision problem can be solved with 
> values.  The values that evolved biologically are common and don't change 
> very fast; so it's a good bet you love your children more than yourself.

It's a good bet to metabolize carbohydrates before protein too, but that 
doesn't imply that a moral dimension could or would be conjured out of 
nowhere to somehow assist that decision.

> This kind of modeling may indeed offer some predictive strategies and 
> instrumental knowledge of morality, but if we had to build a person or a 
> universe based on this description, what would we get? Where is the 
> revulsion, disgust, and blame - the stigma and shaming...the deep and 
> violent prejudices? Surely they are not found in the banal evils of game 
> theory. 
> They're found in your the banal neurons of your brain, 

Not necessarily. All that we find in neurons so far is molecules. No sign 
of any disgust. We have only our own word to take for the fact that such a 
thing as disgust even exists. TV shows aren't in the banal pixels of a TV 
screen. The internet isn't in my computer.

> so they could be part of the morals of a robot if we chose to build it 
> that way.  From our perspective as citizens in a very diverse and 
> interconnected world of billions of people, we can see ways in which we 
> might give a robot better, more adaptive, values than biology has given us.
> Brent

If morals didn't exist, why would we choose to invent them? What possible 
purpose could be served by some additional qualitative layer of experience 
on top of the perfectly efficient and simple execution of neurochemical 
scripts? Don't you see that the proposed usefulness of such a thing is only 
conceivable in hindsight - after the fact of its existence?


You received this message because you are subscribed to the Google Groups 
"Everything List" group.
To view this discussion on the web visit 
To post to this group, send email to everything-list@googlegroups.com.
To unsubscribe from this group, send email to 
For more options, visit this group at 

Reply via email to