> Create a machine that is like a human in all respects that you cite, tool use, language use, abstract thought, logical reasoning, permanent memory, and if it is just comprised of that, we will call it a very amazing robot. You could even add self-preservation routines, and we will call it an amazing robot that knows its own value.
I disagree. Much science fiction (for example Asimov) deals with the social positions of intelligent robots. I would assume, based on the number of people that find the arguments these works make compelling or at least interesting enough to buy them, that if such robots did exist, they would quickly gain at least a small faction of human political supporters. Eventually, through normal human social and political processes, our society and culture would hopefully reach some kind of position on what exactly makes something morally equivalent to humans, in a way that assigns some definite and logically consistent moral status to these robots.
> Similarly, I believe that emotional response is the reason why humans agree that it is not correct to destroy babies, but that some find it fine to destroy embryos.
It's probably true that emotional responses are a big part of the reason for much of the support behind most laws and policies, from the least controversial (murder is illegal) to the most (abortion, Obamacare, death penalty, affirmative action, welfare...). But that doesn't mean we can't or shouldn't come up with abstract intellectual justifications for them.
> Divorcing my emotions from every decision making process would just be disgusting...
Emotions are basically heuristics which the human brain has evolved. They quickly give you guidelines for how you should act in a given situation without wasting valuable time and intellectual effort creating an elaborate argument as a justification of the obvious.
Emotions are usually right, but sometimes they lead us astray. And we're at a complete loss when we have to make a law that goes either one way or the other on something, but different people in our society have different emotional responses that lead them to different positions on the issue.
And besides, emotions can lead to things like racism or Nazism if followed blindly. (Godwin's Law, I know.)
That's where debate and argument come into play.
In a good debate, your empathy should let you experience the emotions on both sides of the issue, freeing your brain to decide which argument has a good logical structure and is adequately supported by facts.
Once you've decided, you should try to partially (but not totally! -- your brain can be wrong too!) suppress your empathy for the losing side and encourage your empathy for the winning side. You can actually train your emotions -- basically calibrating your heuristics to agree with your intellectual decisions, so in the future you can remember your position on the issue in your gut, without having to remember and fully reconstruct the argument you made. Yeah, I've found that hacking my own brain is awesome :)
By the way, using emotional heuristics as a shortcut is something I relied on in my reductio point against using emotions as your sole means of political decision-making -- your emotions immediately tell you that racism and Nazism are bad things, because you've been trained to think of them that way. While you can probably come up with a sound argument that racism and Nazism are bad, your emotions tell you those are bad things without you having to consciously construct or remember a justification of why they're bad. So you can immediately sense that the structure of my argument -- "relying blindly on emotions without consulting reason is bad because it leads to bad things" -- is sound, without having to go into an entire sub-argument about whether racism and Nazism are good or bad, and why, and whether the logical structure of the sub-argument is okay, and mentally fact-checking the facts cited by the sub-argument against what you know of the history of Hitler and World War II and slavery and Jim Crow and Martin Luther King...and then more sub-sub-arguments about fundamental moral questions like whether genocide is morally okay, or whether slavery is morally okay, or whether it's morally okay to treat people badly based on a certain set of genetic traits without actually enslaving them...without those emotional shortcuts to help us out, we'd never be able to communicate moral ideas efficiently or make moral decisions fast enough for them to matter in a fast-moving real-life situation.
(FYI, I suspect this feedback mechanism is what Terence Tao was talking about when he discussed how mathematicians train themselves to quickly decide whether something is true or false without constructing an airtight proof [1].)
[1] http://www.quora.com/Mathematics/What-is-it-like-to-have-an-...