re: low profile

Forrest Bishop (
Sat, 12 Oct 1996 00:37:17 -0700

[[Robin Hanson writes:]]

Thanks to Forrest Bishop for putting some thought into what David Brin
calls the "Berserker" question.

[[I though Berserkers were stray military machines, like the Borg, that
simply destroyed because it was their programmming.]]

The possibility that there is lots of
life all around, all trying to play dead to avoid retribution, is very
popular among the very best hard science fiction authors, including
Brin, Benford, and Bear.
[[Oh. Haven't been reading much SF in recent years. ]]

It is well worth more attention than I gave
it in my Great Filter paper. (Btw, I extracted the bio part of that
article, made it more formal, and just submitted it to Science. See:
[[Congratulations! Think they're wired enough to print it?]]

In fact, it would be great fun, and worth doing intellectually, to run
a game-theoretic simulation of this problem.
[[I agree. I do hope the "low profile" scenario is incorrect. It
would be conducive to hope about the future to provide a studied
counter example. If the "low profile", or "Berserker" situation is true, one
should realize that our descendents _are_ the Berserkers.

Could get an Icarus
publication out of it. Any worthy coauthors interested? :-)
[[Interested, yes.]]

Let me try to sketch out the strategic situation here.

A new civilization like ours knows that a shell of our early radio
signals is expanding out at the speed of light, and that this may
trigger "berserker" probes flying in at near the speed of light to
destroy our solar system. So some folks may want to get the hell out
of here and try to colonize elsewhere.

[[Two cases here:
1) We've intercepted their radio signals.
2) We haven't, and are paranoid.

Case 1): We either launch a preemptive strike, or send "welcome space brother" noises
in their direction. We simultaneously beat feet in the opposite direction.

Case 2): For the case of not knowing the direction of an imagined attack, one direction
is as good as the next. We expand in a rough sphere, always preceeded by our past radio shell.

Some complications:
Interstellar media variations put speed limits on starships. We are currently in the
interesting position of being between two interstellar gas clouds. Our Solar System
will enter a gas cloud in about 50000 (?) years, at which time interstellar travel may
have to slow down. One open question: do we use big fat starships or wee little
Our current radio signals may only be perceptible for a few thousand light years, before
being lost in the background.]]

They'd want to go as fast as
possible, to escape a possible detonation wave of unknown size and

[[A putative speed of light spherical pulse of some kind is the nastiest
possibility. It would eventualy catch any material object.]]

(could someone nova our star, or galactic center?),

[[Yes. Consider a type of nova event, wherein a White Dwarf (a burned out
star with the mass of the Sun and diameter of the Earth) slowly accumulates
a hydrogen atmosphere on its surface. At some point, the temperature and
density of this atmosphere becomes high enough to sustain fusion. The
interesting part about this is that the fusioning is initiated at a _single region_
near the surface of the White Dwarf, then spreads horizontally to encircle the
globe, as well as vertically up through the hydrogen column as far as conditions
will permit hydrgen burning.
This is essentially a big hydrogen bomb, set off by a teensy trigger (a
catastrophy event).
Now consider an average, ho-hum star like the Sun. This system is in a delicate
balance between gravitational collapse (implosion), and runaway fusion (explosion)
(we've come to find out the Sun is not a perfectly stable system, BTW). The thermal
and radiation pressure that keeps the Sun from collapsing is generated by continuous
nuclear fusion at or near the core. This core is surrounded by a blanket of mostly
hydrogen, in a very high temperature and density state, in fact, just short of fusion
criticality. An anomalous increase in these parameters may be able to initiate a fusing
blast wave that encircles the core, similar to the White Dwarf scenario. The resulting
sudden energy release would then blow off the outer layers of the Sun, like a nova.
One method of providing this trigger is to fire an object at the Sun, that is large enough
and fast enough to penetrate the outer layers without burning up, and deposit its energy in
the pre-critical region.

The case of a supernova chain reaction at the core of the Milky Way Galaxy has been studied
in some detail, and may have already occured a number of times. There is apparently a big
(4 million+ Solar mass) Black Hole at the galactic core that occasionaly gulps down a
stray star or two. The resulting fireworks may have provided the energy input to nearby
(~ light year) densely clustered stars to cause them to nova or supernova, in a sort of chain
reaction. As a result, the Core, out to a radius of 10 or 20 thousand light years, is probably sterile.
It would behoove us to set up an Early Warning System within a few hundred light years of this

or a
followup wave of berserkers. And they'd also want to not send signals
back to Earth that such enemies might intercept
- signals would have
to look like cosmic background to someone without the right key (and
with the home base keeping your location info very localized, on a
hair trigger to destroy it.)

[[So you don’t buy the ‘wipe the memory idea’? The basis is that
one cannot be certain of a memory immune from infiltration.]]

They'd also want turn a corner
somewhere, so their path away from Earth (ionized regions) doesn't
lead right to them.

[[Dragging a cable behind the starship while passing near a star provides
a free (no mass or energy expense) course change, via Lorentz Force
interaction with the stellar magnetic field.

And not knowning how far they might be followed,
they'd want to go a good long way.

[[Hmmm. An argument for intergalatic travel. One can travel much
faster and with less wake noise between galaxies than through them. ]]

New colonies might try to hide themselves better, but even then some
of the locals may get itchy of being traced, or growing too big not to
be seen, and try to escape. Within any system, keeping the quiet is a
public good, so a strong central policing might be required to enforce

[[If the colony is a merged superinelligence, said policing would simply
be a form of self-discipline.]]

..Or considering the costs and
risks of strong central policing, they might decide to just run for it
from the very beginning.

[[Running early is an excellent policy, but suffers the “hare and the
tortoise” problem- that later, more technically advanced probes will
overtake the pioneers. Of course, that presumes there are later probes.]]

A civilization may also want to try to "test the waters", by sending
one probe to a distant system to try this "running for it" strategy,
and to then watch the results. They might also try a "loud mouth"
strategy in such a distant system, broadcasting as loud as they can to
see who responds how. Of course just how far away is "distant" is a
tricky question.

[[This reminds me of the strategy of some nesting birds. When threatened,
the hen will
move some distance from the nest and then call attention to herself.]]

The strategic situation of a potential berserker is also

It seems to me that the outcome of all this strategizing is that the
best strategy is just to expand as aggressively as possible. By the
time anyone comes to destroy any one system you have used it well,
converting it to a billion probes or much more within a few years.
That huge economy gives you all the better chance to learn new
technology, so some of your descendants become as advanced as your
advanced enemies.
To verify this intuition, an explicit simulation would be appropriate.

[[Indeed it would!

Forrest Bishop