[ggreer]

When Bostrom and others talk about humanity, they usually mean humanity and its extremely advanced descendants. They expect some value drift. Robin Hanson puts it rather succinctly in http://www.overcomingbias.com/2011/12/steps-up-the-ladder.ht...:

Of course we should also wonder what we may become as we rise. We are no longer the foragers who began this climb, nor the farmers who climbed just a few floors below, and those ancestors would probably not be pleased with everything we have become. We’ll probably also have misgivings about what our descendants become. But hopefully we will on net be proud of them, just as our ancestors would probably be proud of us.

[sdrothrock]

Thank you for the clarification!

[lutusp]

Okay, hold on a minute. When you say, "humanity and its extremely advanced descendants", when you link to an article containing the phrase "steps up the ladder", when you quote someone saying, "we should also wonder what we may become as we rise" [emphasis added], you're overlooking something basic about evolutionary theory that everyone needs to understand.

And that is that evolution is not necessarily a progression from less to more advanced, from less intelligent to more intelligent, or for that matter, from less anything to more anything.

Evolution is not a plan with a goal, it is a blind algorithm that chooses survivors, regardless of the survivors' traits, with the single requirement that they are the fittest for the environment in which they find themselves.

This talk about steps up the ladder and extremely advanced descendants is scientifically ignorant and a New Age fantasy. We are as likely to be replaced by cockroaches as by superbeings.

[ggreer]

1. I was explaining the position of Bostrom, Hanson, and others. I do not completely agree with them.

2. I think you have misinterpreted their position. Bostrom and Hanson know quite a bit about evolution. They know that evolution is undirected and would eventually result in an organism we wouldn't recognize, let alone value. But they both think that we are entering a time in which we will no longer be bound by evolution. They think that humanity will soon be able to engineer minds, allowing us to improve their raw abilities while having them retain many of our own values.

On this point, I do agree with them. Evolution hill-climbs, so it gets stuck in local maxima and can't search the entire solution space. We're already building lots of stuff that could never evolve: radio, wheels, impellers, turbines, lasers, etc. In billions of years, evolution hasn't figured out a way to send signals faster than 0.000001c (300m/sec). That's how fast sound waves and nerve signals travel. As optimization processes go, it really is quite terrible. If we want to make better minds in any reasonable time-frame, we'll need to engineer them ourselves.

[lutusp]

> I think you have misinterpreted their position.

I understand it perfectly -- they're either as ignorant as their followers, or they're exploiting public ignorance.

> Bostrom and Hanson know quite a bit about evolution.

They either know nothing about evolution, or they're deliberately misleading their readers. Contrary to their writing, natural selection is not a race to the top, because it's not a race to any particular objective.

> But they both think that we are entering a time in which we no longer be bound by evolution.

Apart from revealing their inability to grasp evolutionary theory, this is an ignorant New Age fantasy. We will always be bound by natural selection, even when we actively participate in the process.

> They think that humanity will soon be able to engineer minds, allowing us to improve their raw abilities while having them retain many of our own values.

But that's also evolution. To argue that people meddling with genetics isn't evolution is to misunderstand evolution's scope.

> Evolution hill-climbs ...

You really need to stop thinking about natural selection as though it's a race to the top of the hill. This idea contradicts both evolutionary theory and copious observational evidence.

> We're already building lots of stuff that could never evolve [emphasis added]: radio, wheels, impellers, turbines, lasers, etc.

All these things exist in nature, even including the lasers, all of which evolved in nature:

http://laserstars.org/summary.html

Bacteria use wheel- and axle-based electric motors to propel themselves through their environments:

http://www.ncbi.nlm.nih.gov/books/NBK22489/

All your examples have similar pre-existing embodiments in nature. And they all evolved.

> If we want to make better minds in any reasonable time-frame, we'll need to engineer them ourselves.

People doing engineering is an example of natural selection. There is nothing in the sequence of human events that isn't an example of evolution by natural selection.

All this talk about moving beyond evolution fails to grasp how evolution works in our lives. Even A/B testing of Web pages is an example of natural selection.

To summarize, these people you're quoting are simply pandering to low public taste -- they're either broadcasting their own ignorance or exploiting the ignorance of the public. Evolution doesn't work they way they claim, and their writing is a scientific laughingstock.

[edanm]

I don't know if you've read the writings of Bostrom/Hanson. You are very wrong in terms of both their message, and their purposes.

They do not misunderstand evolution. In fact, considering they're both respected professors, you might want to give them some amount of "benefit of the doubt". If you're basing your position about them based on one article, you really should at least consider the fact that you're misunderstanding them.

As for what you say about evolution, I have a hard time with what you call "evolution", because your definition seems to include literally everything that ever can or will happen on earth. So let's put aside the word "evolution" and talk instead of what we actually think is going to happen.

Hanson/Bostrom etc. talk about the fact that humanity will be able to quickly and significantly change what we are, as in rewriting our genetic code, rewriting our software, and so on. (If you want to call this "part of evolution", that's fair, but beside the point I'm making).

They consider this a "rise" in terms of what we, right now, consider to be better or worse. If you'd tell me that in 10 years, humanity will be replaced by cockroaches, you're right that it doesn't matter to "evolution", but it is certainly something that I, as a human, consider to be a step down.

In similar ways, rewriting our genetic code or making other changes to humanity can be considered an advancement from humanity's point of view.

That's the kinds of things they are talking about, and the reason they use phrases like "steps up the ladder".

[lutusp]

> You are very wrong in terms of both their message, and their purposes.

They repeatedly refer to evolution's goals, but evolution has no goals. They are wrong, and I am citing the standard scientific references to evolutionary theory.

http://evolution.berkeley.edu/evolibrary/misconceptions_faq....

Quote: "One important mechanism of evolution, natural selection, does result in the evolution of improved abilities to survive and reproduce; however, this does not mean that evolution is progressive"

The above flatly contradicts your sources, who argue that their version of evolution is progressive.

> In fact, considering they're both respected professors ...

While trying to engage in scientific debate, avoid this common logical error:

http://en.wikipedia.org/wiki/Argument_from_authority

> They consider this a "rise" in terms of what we, right now, consider to be better or worse.

But that's wrong -- we cannot possibly know what nature has in store for us in the future. And no evolutionary process, natural or unnatural, can or should be described as a "rise". When applied to human beings, this smacks of eugenics, of engineering the "defects" out of people. It assumes that we understand nature better than we do, or that we can outwit nature, or that we can implement eugenic projects without destroying society. We keep proving that we can't do that.

> In similar ways, rewriting our genetic code or making other changes to humanity can be considered an advancement from humanity's point of view.

So it is a eugenic proposal. All the worse for us. Eugenics suffers from many serious defects, one being that we can't outwit nature, another being that implementing eugenic plans inevitably falls apart for practical and political reasons.

Relating it to evolution is simply a way to give it a pseudoscientific gloss and put a respectable patina on a dangerous social idea.

[edanm]

"They repeatedly refer to evolution's goals, but evolution has no goals."

They're speaking metaphorically.

"The above flatly contradicts your sources, who argue that their version of evolution is progressive."

Like where? Where do they specifically say that?

"While trying to engage in scientific debate, avoid this common logical error:"

I wasn't arguing from authority. I did not say "they are authorities, believe what they say". I said that, since they are respected professors, you might want to give them the benefit of the doubt. Do you really believe that there is no chance that you are wrong and that you are misunderstanding them, and if the fact that they are both respected professors has zero relevance on whether you should at least try to see whether you might have misjudged them?

"But that's wrong -- we cannot possibly know what nature has in store for us in the future"

I'm just pointing out that you are now yourself talking as if "nature" had goals and purposes, exactly the behaviour you found troubling in their writings.

As for the rest of your post, I can't say I disagree with the facts, only with your opinions. Yes, we are discussing here things like changing people's genes, and yes, that means it is eugenics. No, I don't think this is necessarily a bad thing.

More importantly to Bostrom's point, it doesn't matter what we think - people will start doing things like this in the (near?) future, we'd better be prepared for it.

[ggreer]

I don't have time to address everything you've laid out here, but I must say that you seem to almost willfully misinterpret my statements. Moreover, many of your rebuttals are simply incorrect on matters of fact.

Please, in the future, try to remember the principle of charity. Also, you might foster more productive discussions by sprinkling a little tact onto your comments.

[lutusp]

> Moreover, many of your rebuttals are simply incorrect on matters of fact.

You're wrong, but post your evidence -- let the evidence decide. The authors you cite make a number of obviously false claims about evolution, claims falsified by citation in the standard references.

[kamaal]

>>Evolution is not a plan with a goal, it is a blind algorithm that chooses survivors, regardless of the survivors' traits, with the single requirement that they are the fittest for the environment in which they find themselves.

Well that does sound like a goal driven algorithm to me. Except that you can't actually say if a move made by the algorithm will actually produce a acceptable outcome to the algorithm itself. Due to the sheer number of variables, the only option the algorithm has is to do a kind of A/B testing. To make changes to a smaller subset of subjects and see if they survive, if they do they do. If they don't, well then you can kill that test and move on more modifications.

Evolution is an optimization algorithm, which makes changes, tests and makes more changes/corrects based on feed back.

[lutusp]

>> Evolution is not a plan with a goal, it is a blind algorithm that chooses survivors, regardless of the survivors' traits, with the single requirement that they are the fittest for the environment in which they find themselves.

> Well that does sound like a goal driven algorithm to me.

Not in the way most people mean. When people hear there's a scheme to evolution, many assume this means a gradual ascent in complexity or intelligence, but that's not necessarily so. There's no relationship between natural selection and any specific endpoint.

> Evolution is an optimization algorithm ...

No, it's an adaptation algorithm. Its outcome are never optimal, only the best approximate response to environmental changes, in an ever-changing environment.

[Houshalter]

1) That's not entirely true. Evolution is an optimization process and tends to produce more fit individuals over time. Intelligence is a great example of a trait that improved over time because of evolution.

2) No one is claiming that in the first place. The quote is about the evolution of culture and technology, not biological evolution.