RE: computer pain

From: Stathis Papaioannou <>
Date: Mon, 25 Dec 2006 14:36:23 +1100

Jef Allbright writes:

[Stathis Papaioannou]
>> If slavery could be scientifically shown to promote the
>> well-being of the species as a whole does that mean we
>> should have slavery? Does it mean that slavery is good?
> Teaching that slavery is "bad" is similar to teaching that lying is
> "bad". In each case it's a narrow over-simplification of a more general
> principle of what works. Children are taught simplified modes of moral
> reasoning to match their smaller context of understanding. At one end of
> a moral scale are the moral instincts (experienced as pride, disgust,
> etc.) that are an even more condensed form of "knowledge" of what worked
> in the environment of evolutionary adaptation. Further up the scale are
> cultural--including religious--laws and even the patterns of our
> language that further codify and reinforce patterns of interaction that
> worked well enough and broadly enough to be taken as principles of
> "right" action.
> Relatively few of us take the leap beyond the morality that was
> inherited or given to us, to grasp the broader and more extensible
> understanding of morality as patterns of behavior assessed as promoting
> increasingly shared values over increasing scope. Society discourages
> individual thinking about what is and what is not moral; indeed, it is a
> defining characteristic that moral principles subsume both narrow self
> interest and narrow situational awareness. For this reason, one can not
> assess the absolute morality of an action in isolation, but we can
> legitimately speak of the relative morality of a class of behavior
> within context.
> Just as lying can clearly be the right action within a specific context
> (imagine having one's home invaded and being unable, on moral grounds,
> to lie to the invaders about where the children are hiding!), the moral
> issue of slavery can be effectively understood only within a larger
> context.
> The practice of slavery (within a specific context) can be beneficial to
> society; numerous examples exist of slavery contributing to the economic
> good of a locale, and on a grander scale, the development of western
> philosophy (including democracy!) as a result of freeing some from the
> drudgery of manual labor and creating an environment conducive to deeper
> thought. And as we seek to elucidate a general principle regarding
> slavery, we come face-to-face with other instances of this class of
> problem, including rights of women to vote, the moral standing of
> sentient beings of various degrees of awareness (farm animals, the great
> apes, artificial intelligences), and even the idea that all "men", of
> disparate mental or emotional capability, are "created equal"? Could
> there be a principle constituting a coherent positive-sum stance toward
> issues of moral interaction between agents of inherently different
> awareness and capabilities?
> Are we as a society yet ready to adopt a higher level of social
> decision-making, "moral" to the extent that it effectively promotes
> increasingly shared values over increasing scope, one that provides an
> increasingly clear vision of effective interaction between agents of
> diverse and varying capabilities, or are going to hold tightly to the
> previous best model, one that comfortingly but childishly insists on the
> fiction of some form of strict equality between agents? Are we mature
> enough to see that just at the point in human progress where
> technological development (biotech, nanotech, AI) threatens to
> drastically disrupt that which we value, we are gaining the necessary
> tools to organize at a higher level--effectively a higher level of
> wisdom?

Well, I think slavery is bad, even if it does help society - unless we were actually
in danger of extiction without it or something. So yes, the moral rules must bend
in the face of changing circumstances, but the point at which they bend will be
different for each individual, and there is no objective way to define what this
point would or should be.

Slightly off topic, I don't see why we would design AI's to experience emotions
such as resentment, anger, fear, pain etc. In fact, if we could reprogram our
own minds at will, it would be a very different world. Suppose you were upset
because you lost your job. You might decide to stay upset to the degree that it
remains a motivating factor to look for other work, but not affect your sleep,
ability to experience pleasure, etc. If you can't find work you might decide to
downgrade your expectations, so that you are just as content having less money
or a menial job, or just as content for the next six months but then have the
motivation to look for interesting work kick in again, but without the confidence-
and enthusiasm-sapping disappointment that comes from repeated failure to
find work. Or any variation on the above you can imagine.

Stathis Papaioannou
Be one of the first to try Windows Live Mail.
 You received this message because you are subscribed to the Google Groups "Everything List" group.
To post to this group, send email to
To unsubscribe from this group, send email to
For more options, visit this group at
Received on Sun Dec 24 2006 - 22:36:40 PST

This archive was generated by hypermail 2.3.0 : Fri Feb 16 2018 - 13:20:12 PST