From: Robert J. Bradbury (bradbury@aeiveos.com)
Date: Sat Jul 10 1999 - 16:53:00 MDT
> hal@finney.org wrote:
>
> The first is the claim that superintelligences (SIs) would not want
> to talk to us because we don't want to talk to nemotodes. I argued
> previously in another context that this is a false analogy (I used the
> example of insects). We don't talk to insects because they can't talk.
> They don't talk to each other, either.
Nonsense. Insects do a great job talking to each other for the
purposes of their survival. Moths, I believe they do some of
the best pheramone signalling known (a few molecules hundreds of yards,
perhaps even miles away). Crickets, certainly make themselves known
to nearby mates.
We may observe insects (for curiosity) or study them (for our education
as to how/why they work), but we don't talk to them because we aren't
at the same level on the scale. Imagine the conversation --
Hello...... long pause Hello
How are you today.... I'm hungry
Would you like to play a game... I have to go lay some eggs, bye...
The problem isn't that they can't talk, its that the conversation
would be extremely short due to their limited intelligence. They
don't have the intelligence or memory capacity to make the conversation
interesting. [When was the last time you talked at length with a
*human* who bored you to tears?] Have you ever interacted with someone
with Down's syndrome? You may care for them deeply and motivate yourself
to make conversation and try to figure out what is going on in their
mind but it isn't very interesting for very long (at least in my
experience). Greg's example with the lemurs is a case in point -
"they concoct simple plans in pursuit of their constant desire
to stir up mischief."
They are interesting because they are creative. Greg can predict
how they may behave some of the time, but not perhaps all of the time.
I would argue that an M-Brain could predict with relatively high
accuracy what humans would do. They have the observational capacity
and storage capacity to build a huge database of statistical
responses to given situations (just as Greg does with the lemurs).
They also have the computational capacity to simulate a variety
of possible futures for any human situation and would likely be
able to predict exactly how things will play out (within the limits
of chaos theory).
Species communicate about stuff necessary for their survival or
simply about stuff that interests them (if they are more intelligent).
We are intelligent enough to be interested in an M-brain, but
I really doubt we are intelligent enough to interest it for long.
On the other hand, post-singularity we have the capacity to be
very "educatable" and an M-brain might take an interest in us
as we do in human children.
>
> To make the analogy valid, we have to look at the modes of communications
> that lower animals use, and ask whether we interact with them at that
> level.
I don't speak moth pheromones or cricket chirps. Scientists may
"study" these things but only communicate with them as part of an
experiment.
> This would correspond to SIs who talk to us at our level, even
> though they communicate amongst themselves in more sophisticated ways.
Ok, so no we are back to aliens "talking" to us as part of an "experiment".
>
> I suggest that we do interact with lower animals at the level that they
> interact with themselves. An insect has a simple model of the world,
> adequate to find food, avoid predators, mate and survive. We do interact
> with insects at this level.
The only time I feed an insect is by accident (being bitten) or
by intent (to poison). They are used as food by some people and
they certainly prey on us. We don't mate with them due to the
scale differences. We are at war with them, just as the plants are.
The only communication there is between us/plants & insects are
attacks and defenses.
> Insects are able to survive and even thrive around humans.
Of course, due to the scale differences our environmental niches
do not overlap very much so we aren't in direct competition.
> We are part of the same physical universe and from the insects'
> point of view we are interacting and communicating with them
> just like other parts of the universe.
Ah, but I don't call this communication. My original statement
was "we don't talk to nematodes and SIs don't talk to us". When
I say "talk" I mean the exchange of a significant number of
bits of information for which there is mutual understanding
on both sides of the communications link. When one party
doesn't understand, it says so and the other party tries
to explain it better in terms that it does understand.
I call that "communication".
>
> As our technologies improve and we learn more about insects, we will
> become even better at communicating with them in their own terms. We will
> learn to use chemical messengers and other methods to interact with them.
> This would be closely analogous to an SI speaking to us in our own
> terms.
>
Yes, since we have some intelligence we can ask it a question
and we may get a response. But the problem is much worse than
that of a parent, who bored with a child says, "Ok, now
you go outside and play with the other children...".
Can communcation occur, yes, but the quantity and quality
is fundamentally limited by the bandwidth of the channel
and capacity of one of the parties.
> Even though we are not capable of understanding all of what SIs say to
> each other, it is still plausible that they might choose to communicate
> with us at our level. Our own experiences with lower animals suggest
> that there are many motivations for higher animals to communicate with
> lower ones.
Only because we don't understand them completely yet! We are studying
them. We stop studying them once we undertand them fully (unless we
choose to experiment on them). You aren't going to sequence the yeast
genome 23 times, you only do that once. You aren't going to understand
aerodynamics of bee wings 5 times, you only do it once.
If we happen to represent a truely new species or novel developmental
path to an SI, then yes they might study us and maybe even choose
to talk to us. But if you've got billions of years of observations
and the evolutionary paths of trillions of species under your belt,
I think it is going to look like same old thing to you.
>
> The second point is that it does not seem productive to imagine that
> the universe is heavily engineered by active SIs. Throughout history,
> we have succeeded by assuming the opposite, that there is no higher
> intelligence or higher power behind the observations that we make.
Is it more logical or productive to imagine that *all* technological
civilizations suffer the Universal Fatality Equation?
I would say that we have succeeded by using the simplest explanation
that fits the observations. I'll I'm doing is pointing out some
unexplained observations and providing what to me is a simple and
logical explanation.
You may be right about "heavily engineered". SIs might never
do stellar engineering (it could be very hazardous to take stars
apart) and might never colonize (for the reasons of increasing
diversity as I've mentioned or perhaps their own moral principles).
Even so, the universe is more than old enough to have evolved
many SIs in many different locations along many different
evolutionary paths. The limits of physics may funnel intelligence
into a small set of optimal architectures.
>
> Why should we decide, now, that when there is a mystery in our
> observations, that intelligence is probably behind it?
The difference between previous "explanations by the shaman" and
modern explanations is that they are based on the principles of physics
(e.g. natural laws). Is it not possible that the evolution of intelligence
is controlled by these "natural laws"?
- elements consist of certain particles organized in certain ways.
- molecules connect according to fundamental rules
- gravity distorts space and causes molecules to form stars
- the principles of natural selection allow the development
of intelligence and drive it to its limits.
> It's never been true before. We face a solar neutrino discrepancy now.
> Maybe SIs are engineering the sun? We have a missing mass problem.
> Maybe SIs are hiding mass? We don't have a good model for how life
> started. Maybe SIs brought life to earth?
I think you have to add the criteria that drives evolution --
Is it in your self-interest? If it were in the interest(s) of
an SI to do any of those things, then yes, I would say that
SIs might be responsible for it.
>
> Such "answers" are the end to inquiry. Sufficiently advanced technologies
> are magical and there is little we can say about them.
Not true, nanotechnology seems like magic, but Drexler & Freitas
seem to be able to tell us what is probable, possible and impossible.
Several authors have argued that star dismantlement seems possible,
but just like interstellar travel, the benefits may not be worth
the costs. Are attaining personal immortality and more intelligence
in my self-interest? If the answer is yes (for most people), then
civilizations will develop these technologies.
>
> The weakest part of this theory is the assumption that although the
> universe is full of SIs they are not here in the solar system. It seems
> very implausible that they would have left our solar system untouched for
> all time.
I too find this aspect uncomfortable. But I've offered explanations
for this -- that they may be here, that we are an experiment in
increasing diversity, that they are simply disinterested in us, etc.
It is important to keep in mind that SIs gain relatively nothing
by colonizing new space. The progagation delays between stars
mean that you can't do effective parallel computations (even over
millions of years). On the other hand discovering a new computational
architecture, a new algorithm, a new material structure, etc. that
lets you think smaller/faster/cheaper, etc. has very big payoffs.
To paraphrase Feynman -- "There is *more* room at the bottom".
> This means that we must reconsider the possibility of divine
> intervention throughout history.
I know, don't you just love it... But you don't have to invoke
SIs for that requirement. Nanotechnology could easily explain
the miracles of Jesus. So a single non-SI alien species that
wanted to put a little spin on things be responsible for Christianity.
To throw away "divine intervention", you have to put a substantial
limit on the evolution of life or the evolution of intelligence.
You are back to arguing that "we" must be a special case.
> I suggest reading Robin Hanson's article which takes at face value
> the absence of stellar engineering, and draws conclusions from it:
> http://hanson.berkeley.edu/greatfilter.html.
I read it ages ago
>
ps -f
ps -f
This archive was generated by hypermail 2.1.5 : Fri Nov 01 2002 - 15:04:27 MST