Comments on: Morality of the Future

By: The Volokh Conspiracy

The Volokh Conspiracy — Tue, 27 Mar 2007 01:54:03 +0000

Assessing our Moral Beliefs in Light of Predicted Future Moral “Progress”:

At the excellent Overcoming Bias blog, Hal Finney makes an insightful point about our perceptions of past and future moral progress:

…

By: James Wetterau

James Wetterau — Sat, 24 Mar 2007 18:06:56 +0000

I see social standards of morality as, on average, a matter of attempting to address threats to an empowered group coming from individual behavior. (Of course some such attempts may go and have gone wildly awry.) I see the technological and scientific changes of preceding centuries as typically having broadened the sphere of those whom we are expected to consider with regard (increasingly as equals) and the precision and rationality with which we are expected to calculate the consequence of our actions. As our calculating powers become greater, I expect moral concerns to become more stringent and precise, as a reasonable person “ought to know better” in more and more circumstances.

By: Stuart Armstrong

Stuart Armstrong — Sat, 24 Mar 2007 06:26:36 +0000

Some future morality may not only depend on extra information, but mainly on extra technology. One you’ve got artificial meat, for example, of comparable taste and price to real meat, then the arguments against killing animals becomes much stronger.

If artificial reality becomes cheap, and it’s easy to make a lion think he’s killing a zebra when he isn’t really, then we’ll probably do so.

I think a lot of the moral arguments are already out there – it’s just a question of when/if it’ll become cheap and easy to follow them.

By: Eliezer Yudkowsky

Eliezer Yudkowsky — Fri, 23 Mar 2007 23:13:58 +0000

Robin, I think that may be an important heuristic – “Imagine a full range of possible future moralities, and ask yourself how likely you are to accept the future’s differing judgment.” It may give us important information about our own moral beliefs.

For example, I’d be very likely to accept the future’s differing judgment on the personhood of cats – this says both that I’m not absolutely certain on the issue and that I think that the moral judgment should change as a result of received information. (I currently think cats are not sentient and not persons.)

By: pseudonymous

pseudonymous — Fri, 23 Mar 2007 21:23:24 +0000

Could “animal rights” be extended to wild animals? If so, should we prevent lions from violating the rights of zebras?

If not, why not? Can you say you support a right, if you refuse to try to enforce it?

Could we extend “reproductive rights”? If we can allow abortion of an entity one day before birth, why not one day after birth?

By: Robin Hanson

Robin Hanson — Fri, 23 Mar 2007 20:23:49 +0000

The ability to substantially predict the future of morality would be a strong argument against morality changes being due information we learn, just as the ability to predict future stock prices would argue against stock price changes being due to information. So you have to imagine a full range of possible future moralities, in all the imaginable directions, and then ask yourself if you would on average accept the future’s differing judgment, which ever way it went. If not, you don’t really believe that moral changes are mainly due to information.

By: Matthew C

Matthew C — Fri, 23 Mar 2007 17:51:10 +0000

Before we worry about what will happen after the singularity, it would be helpful to actually have a singularity in the first place. . . Given that there has been no progress in understanding the hard problem of consciousness in the past century of neuroscience, that the promises of AI "10 years out" for the past 60+ years keep getting postponed, that no replicating nano-machines appear to be forthcoming, that, in fact, we don't even begin to have a clue how protein folding, organelle, and cellular morphologies are controlled (other than the vaguest notions of chemical gradients -- how is a concentration gradient supposed to generate this, for example) -- perhaps it might be premature to wistfully pine for the atheist version of the rapture and second coming which is the singularity. . .

By: Stuart Armstrong

Stuart Armstrong — Fri, 23 Mar 2007 16:22:15 +0000

Should we always care if the future disapproves of us?
A lot of our morality is very contigent on our circumstances, and will have to appear ridiculous to later generations. Certain forms of xenophobia and sexual restrictions make sense in very primitive, tribal societies; we disdain them rightly for ourselves, but was it automatically wrong for them? Our health system will be looked on with horror, nearly certainly, and probably also the way we treat our poor. It’s not axiomatic that with a different morality we could do better, in our current world.

On a lighter note, there will certainly be the crusty traditionalists, arguing that everything was better in the early twenty-first century “then, people really knew how to behave! What a great time it was to be alive back then!”

By: Stuart Armstrong

Stuart Armstrong — Fri, 23 Mar 2007 15:55:30 +0000

I feel post-singularity speculations are either entertaining stories, amplified ideas about the near future, or reflections about issues we worry about today. But they are in no way serious predictions! We seem to say:
1) The singularity will be a radical transformation, changing things in ways we can’t even begin to imagine.
2) Yet, it will have this characteristic, and this one, and this one, oh, and don’t forget that one.

Extrapolating from today is all we can do, combined with some “if – then” scenarios (if the post singularity world is hobbesian, then its morality will be hobbesian – if not, it won’t).

By: michael vassar

michael vassar — Fri, 23 Mar 2007 14:58:27 +0000

Anders: I wouldn’t recommend extrapolating from recent experience to the post-Singularity period. “Unknowable” was part of the original meaning, and while overly strong, “not usefully knowable via extrapolation, only via analytical methods” seems to remain a very good heuristic. Do you have any analytical disagreements with “The Future of Human Evolution”?

Surely some forms of morality, such as concern for the much less able resulting in sharing of resources with those who will predictably not return any, are disfavored in competition. Status-seeking with positive externalities, which is what most charity is, may not be disfavored (though they may, given more efficient ways of directly observing an agent’s goal system and resources) but optimizing, efficacy based forms of charity are unlikely to be optimal, and post singularity non-optimal may be very close to non-viable.

You agree with the latter statement with respect to dirt, don’t you? That post-singularity high entropy matter is in general unlikely to be left lying around uselessly?