planetary terrorism

2 views
Skip to first unread message

Brandon J. Van Every

unread,
Aug 16, 2001, 10:46:05 PM8/16/01
to
Imagine a planetary colonization scenario 500 years in the future. There
are no aliens, just human beings expanding outwards from Earth. The rate of
expansion is slow enough that it takes a long time to travel between star
systems and humanity simply hasn't run into any aliens yet. "Slow" can be
whatever velocity and whatever technology you're comfortable with positing
in 500 years' time. If you say "Well there shoulda been aliens by now!"
then I'll just say the evolution of sapient life is statistically rarer than
that. Humans are alone, for now.

However, humans are not entirely peaceful and not without their differences.
The vast bureaucracy of stellar expansion allows for a wide range of
political opinion. Some people believe planets are resources for the
taking, and do industrial manufacture with oceans of cyanide. Others
believe at least certain planets are natural treasures to be protected.
Many religions exist as to the ultimate purpose of humanity. The United
Planets operates as a kind of giant welfare office, but it can't control
everything, and organized crime is a popular means of employment. The
occasional isolationist, dictatorial regime seizes power in a far corner of
the United Planets and declares its sovereigny. Usually such planets are
summarily nuked.

Basically, you can project any modern problem into a future one. The
difference is, in the future most people have ready access to industrial
equipment that can easily destroy a planet's surface.

Let's say the United Planets likes to establish colonies to spread the seed
of the human race. Any habitable, Earth-class planet is used for this
purpose. And in the spirit of democratic tolerance, of course different
factions with different ideologies are allowed to settle there. Now, when
political difficulties arise between factions, what prevents them from
summarily nukeing one another? Consider:

- the residents of the planet can easily manufacture the means of their own
destruction
- an offworld ship can easily show up with the means of destruction, dump
its payload, and leave

This seems to be the inevitable consequence of ideologically silly people
having access to too much technological power. It seems that colonies, and
maybe not even the human species, could survive such power. But short of
"magical" technological solutions, like force fields that freeze nuclear
explosions like some Bugs Bunny cartoon, how might the human race actually
survive such high-tech, ideological conflicts? Some possibilities, please
feel free to add and to pick apart:

- humanity just spreads out. Some planets have ideological conflict and
become scorched earth. Others don't. Natural selection is carried to the
stars. Those that can find ways to get along without nukeing each other,
however tenuous and strained, survive.

- a lot of people live in orbital and subspace habitats. Why put all the
eggs in one basket? Problem: it only takes 1 terrorist and 1 nuke to blow
up an orbital habitat. It's not really any different than spreading out to
lotsa planets, it's just that you have more eggs if everyone's spread out in
orbital habitats. Without a means of hiding your habitat, however, this
amounts to ducks in a shooting gallery....

- The United Planets enforces strict separation between residential and
industrial planets. All the heavy gear is used on industrial planets, so
people's access to doomsday devices is limited. Manufacture on residential
planets has to be done "the old fashioned way," with fairly limited
equipment. Problem: cargo ships still arrive at residential planets.
Fissionable materials can still be exploited in secret on a residential
planet.

- The United Planets is a police state. A citizen's brain is surgically
altered at birth to minimize the chance of destructive behaviors. The
communications of the citizenry are monitored constantly by computers.
Violent acts are met with swift reprisal and immediate execution. Most of
the populace is conditioned to pacifism - or else. Problem: the police have
to be violent. How does violence ultimately prevent violence? Problem: the
UP is big. How can they hope to enforce everywhere?

- Conventional warfare is allowed as a legal means of settling differences
between parties. All conventional war is legal, there is no such thing as
starting a conventional war illegally. If you don't like the result of a
lawsuit, you can attack an enemy corporation. Litigation is usually cheaper
than warfare, however, so most businesses accept the results of a lawsuit.
The only weapons banned in warfare are those of mass destruction.
Collateral damage within the territory of the combatants is perfectly legal.
Problem: if one side is clearly going to lose, what's to stop them from
nukeing the other side? And every other stinkin' person on the planet,
while they're at it? Problem: how can economies exist when the strong can
always legally take what they want by force?


--
Cheers, www.3DProgrammer.com
Brandon Van Every Seattle, WA

20% of the world is real.
80% is gobbledygook we make up inside our own heads.

Joseph Hertzlinger

unread,
Aug 17, 2001, 12:01:40 AM8/17/01
to
On Fri, 17 Aug 2001 02:46:05 GMT, Brandon J. Van Every
<vane...@3DProgrammer.com> wrote:

>Basically, you can project any modern problem into a future one. The
>difference is, in the future most people have ready access to
>industrial equipment that can easily destroy a planet's surface.

We should not take a superficial view of planets. Even if the top mile
is incinerated, that still leaves thousands more still to go.

We can look at this in terms of the ratio of the energy one person can
control to the mass one person can control. If the ratio is the same a
kiloyear in the future as now, the increased destructive ability is
matched by the fact that there is more to absorb it.

Brandon J. Van Every

unread,
Aug 17, 2001, 12:30:25 AM8/17/01
to

"Joseph Hertzlinger" <jher...@ix.netcom.com> wrote in message
news:9li4v4$m2e$3...@nntp9.atl.mindspring.net...

> On Fri, 17 Aug 2001 02:46:05 GMT, Brandon J. Van Every
> <vane...@3DProgrammer.com> wrote:
>
> >Basically, you can project any modern problem into a future one. The
> >difference is, in the future most people have ready access to
> >industrial equipment that can easily destroy a planet's surface.
>
> We should not take a superficial view of planets. Even if the top mile
> is incinerated, that still leaves thousands more still to go.

I'm not convinced that most residents would want to live underground. Nor
am I convinced that residents below ground are much safer, as they will
depend upon circulatory systems that reach the surface. A military
installation will be safer underground, and maybe under some circumstances
they could protract a nuclear war, but if all of the civvies are dead then
what are they fighting for? Finally, even a military installation can't
hold out indefinitely. It would be "groundhog" war. First one to pop his
head up, gets whacked!

> We can look at this in terms of the ratio of the energy one person can
> control to the mass one person can control. If the ratio is the same a
> kiloyear in the future as now, the increased destructive ability is
> matched by the fact that there is more to absorb it.

I suppose you could take a dim, mechanistic view of humanity, that they all
move underground as a matter of survival, and adopt a more mechanistic,
expressionless culture due to their confines. Saw a Star Trek TOS episode
about that recently. I guess that is the sort of non-magical survival
excuse that I asked for it. I just don't like the aesthetic consequences.
:-)

But then again, how does this in any way erase the factions/religious
terrorist problem? I didn't posit a homogenous culture with uniform moral
codes. I posited a culture of dissent, where some wackos really are crazy
enough to strap a nuke to their chests and take out a city in a suicide run.
The question is, what can you do to prevent this sort of thing, if the
industrial materials for nuclear fission are commonplace? Or if you can't
prevent it, how does humanity survive a concerted war of extermination by
some nuthatch faction? Seems like any individual egg can always be nuked,
so it's only a question of having lotsa eggs, and hiding them.

Maybe that's an answer. Forgetting where the hell you came from. Thorowing
out your technological knowledge. Hiding on a distant spiral arm of the
galaxy, away from all the trouble. Consorting with apes.

Joseph Hertzlinger

unread,
Aug 17, 2001, 1:35:28 AM8/17/01
to
On Fri, 17 Aug 2001 04:30:25 GMT, Brandon J. Van Every
<vane...@3DProgrammer.com> wrote:

>I'm not convinced that most residents would want to live underground.
>Nor am I convinced that residents below ground are much safer, as
>they will depend upon circulatory systems that reach the surface.

Since energy supplies have to be delivered via the surface, anything
that interferes with access to the surface will also limit the ability
to construct bombs.

>I suppose you could take a dim, mechanistic view of humanity, that
>they all move underground as a matter of survival, and adopt a more
>mechanistic, expressionless culture due to their confines. Saw a
>Star Trek TOS episode about that recently. I guess that is the sort
>of non-magical survival excuse that I asked for it. I just don't
>like the aesthetic consequences. :-)

Why must cavemen have a "mechanistic, expressionless culture"?

>But then again, how does this in any way erase the factions/religious
>terrorist problem? I didn't posit a homogenous culture with uniform
>moral codes. I posited a culture of dissent, where some wackos
>really are crazy enough to strap a nuke to their chests and take out
>a city in a suicide run.

If the society is a Dyson sphere with a population a billion times
that of today, a million people getting blown up each day will result
in approximately the same death rate as being struck by lightning in
our society.

Since the point of terrorism is to inspire fear, in such a society the
terrorists may just give up.

>The question is, what can you do to prevent this sort of thing, if
>the industrial materials for nuclear fission are commonplace?

Fission? That long-obsolete system? A kiloyear from today, a nuclear
terrorist will have to steal the nuke from a museum.

Let us calculate how much energy can be obtained from radioactive
materials inside the Earth. The Earth's heat (derived from
radioactivity) is about 40 terawatts. If the mean lifetime of the
radioactive material is 10 billion years (much of it is thorium) and
if the amount of energy from fission is 4 times the amount from
radioactivity, that amounts to 5x10^31 Joules.

The Sun gives off that much energy in less than two days. In the Dyson
sphere, nukes are popguns. If you want to write a terrorism story
write one about a madman who gets control over one the main power
beams or maybe a madman who sends space colonies colliding with each
other.

>Or if you can't prevent it, how does humanity survive a concerted war
>of extermination by some nuthatch faction?

In a transparent society, we will know where they live.

<digression>
I've been following some of the debates on intellectual property
rights and "information wants to be free." Maybe the information that
wants to be free includes the names, addresses, and credit card
numbers of unauthorized users.
</digression>

Brett Evill

unread,
Aug 17, 2001, 1:43:02 AM8/17/01
to
In article <Nz%e7.16219$ZM2.1...@newsread2.prod.itd.earthlink.net> ,

"Brandon J. Van Every" <vane...@3DProgrammer.com> wrote:

>Imagine a planetary colonization scenario 500 years in the future. There
>are no aliens, just human beings expanding outwards from Earth. The rate of
>expansion is slow enough that it takes a long time to travel between star
>systems and humanity simply hasn't run into any aliens yet. "Slow" can be
>whatever velocity and whatever technology you're comfortable with positing
>in 500 years' time. If you say "Well there shoulda been aliens by now!"
>then I'll just say the evolution of sapient life is statistically rarer than
>that. Humans are alone, for now.
>
>However, humans are not entirely peaceful and not without their differences.
>The vast bureaucracy of stellar expansion allows for a wide range of
>political opinion.

Why is there a vast bureaucracy? How is it practicable given lightspeed
communications and slow travel? How can it enforce its decrees on
distant planets?

--
Regards,

Brett Evill
<ev...@NO.netspeed.JUNK.com.MAIL.au>

Jordan S. Bassior

unread,
Aug 17, 2001, 6:48:03 AM8/17/01
to
Brandon J. Van Every said:

>This seems to be the inevitable consequence of ideologically silly people
>having access to too much technological power. It seems that colonies, and
>maybe not even the human species, could survive such power. But short of
>"magical" technological solutions, like force fields that freeze nuclear
>explosions like some Bugs Bunny cartoon, how might the human race actually
>survive such high-tech, ideological conflicts? Some possibilities, please
>feel free to add and to pick apart:

The most obvious solutions are:

1) Dispersion - Don't put most of your populations on planetary surfaces. Put
them in separate habs, equivalent to villages, towns, or cities in terms of
size, limiting the maximum destruction from any one weapon to one hab.

2) Mobility - Make sure that the habs have at least emergency maneuvering
capability, so that they can bug out of war zones.

3) Security - Be very careful who you let into your habs, and with what
equipment, and what you let them do with dangerous equipment when in there.

4) Active Defense - Deploy area and point defense systems to deal with attacks
by spacecraft, KKV's, RKV's etc, so that "one shot" does _not_ equal "one
kill."

5) Hardening - Make sure that your populations, whether on planetary surfaces
or in habs, live in "hardened" homes. That means that the homes are of sturdy
construction to resist blast, and can at least in emergencies isolate
themselves from their environments for some time so as to resist NBC weapons
effects, and at least slow military nanotech attacks (they could have their own
active nanotech defenses, too, and probably would have to in order to deal with
nuisance nano-hacking attacks).

6) Stealth - Don't let your enemies know where you are. Pellegrino likes this
one, but in reality it would be inconsistent with most of your other
objectives.

--
Sincerely Yours,
Jordan
--
"To urge the preparation of defence is not to assert the imminence of war. On
the contrary, if war were imminent, preparations for defense would be too
late." (Churchill, 1934)
--

Jordan S. Bassior

unread,
Aug 17, 2001, 6:58:53 AM8/17/01
to
Brandon J. Van Every said:

>I'm not convinced that most residents would want to live underground.

Why not?

Keep in mind, living underground wouldn't preclude going up to the surface for
recreational activity, and "living underground" in the context of a life
support and entertainment technology five centuries in advance of our own is
not going to feel very different from livng on the surface. You could even have
virtual sunlight and landscapes. (This is, btw, a _very_ old sf-nal concept,
dating back to the 1960's at least).

>Nor am I convinced that residents below ground are much safer, as they will
>depend upon circulatory systems that reach the surface.

Why will they?

This is supposed to be 500 years in the future, remember? Their houses could be
equipped not only with filtration systems and active nanotech defenses, but
also with at least short- to moderate-duration full-cycle life support systems.

Heck, we could do the "filtration" part today, and could easily build
short-duration partial-cycle life support systems (nuclear submarines have
these).

>A military
>installation will be safer underground, and maybe under some circumstances
>they could protract a nuclear war, but if all of the civvies are dead then
>what are they fighting for?

Why are all the civvies dead?

Oh, and "revenge" is the obvious answer to your question.

>Finally, even a military installation can't
>hold out indefinitely. It would be "groundhog" war. First one to pop his
>head up, gets whacked!

Well, they might just need to hold out long enough for relief to arrive. It
would be a "siege," basically.

>I suppose you could take a dim, mechanistic view of humanity, that they all
>move underground as a matter of survival, and adopt a more mechanistic,
>expressionless culture due to their confines.

Why is living underground necessarily a "dim" lifestyle, given the tech of 500
years hence? Why are they particularly "confined?" (consider what
fusion-powered plasma drills could excavate). Why must their culture be
"expressionless?"

You said "500 years in the future," but you seem to be assuming that the only
technological progress has been in the fields of space travel and _offensive_
weaponry. This makes no sense -- extensive space travel implies advanced life
support technology; command of greater energies for offense implies that these
same energies may also be used for defense.

>Saw a Star Trek TOS episode
>about that recently.

Has it occurred to you that you've partially exposed your own fallacy here? The
ST: TOS was making technological assumptions based on the limitations of
_1960's_ life support and entertainment technology. These are _already_
outdated.

>I guess that is the sort of non-magical survival
>excuse that I asked for it. I just don't like the aesthetic consequences.

They are only aesthetic consequences of your pessimistic assumptions about
excavation, life support, and entertainment technology. Assuming that these
other technologies advance to the degree that you postulate transportation and
offensive weapons technologies have advanced, there are no such consequences.

>But then again, how does this in any way erase the factions/religious
>terrorist problem? I didn't posit a homogenous culture with uniform moral
>codes. I posited a culture of dissent, where some wackos really are crazy
>enough to strap a nuke to their chests and take out a city in a suicide run.
>The question is, what can you do to prevent this sort of thing, if the
>industrial materials for nuclear fission are commonplace?

Scan incoming persons and cargoes for nuclear weapons. Have active internal
surveillance inside your habs.

> Or if you can't
>prevent it, how does humanity survive a concerted war of extermination by
>some nuthatch faction? Seems like any individual egg can always be nuked,
>so it's only a question of having lotsa eggs, and hiding them.

Put active defenses on your habs.

Brandon J. Van Every

unread,
Aug 17, 2001, 7:04:48 AM8/17/01
to

"Brett Evill" <ev...@netspeed.com.au> wrote in message
news:3b7c...@news.netspeed.com.au...

> >
> >However, humans are not entirely peaceful and not without their
differences.
> >The vast bureaucracy of stellar expansion allows for a wide range of
> >political opinion.
>
> Why is there a vast bureaucracy?

Because the United Planets is a bureaucracy, and it attempts to cover a vast
distance.

> How is it practicable given lightspeed
> communications and slow travel? How can it enforce its decrees on
> distant planets?

Who said it was practical or terribly enforceable? This is a Wild West
scenario. People somewhat obey laws because that's what they were
conditioned to obey on another world, and it helps keep most people alive.
But now they're out on the frontier, and lotsa people threaten to get ugly.
However, the United Planets does have a few detachments of marines in orbit
who are perfectly willing to make scorched earth of the whole place if
anyone gets too out of hand....

Brandon J. Van Every

unread,
Aug 17, 2001, 7:11:20 AM8/17/01
to

"Joseph Hertzlinger" <jher...@ix.netcom.com> wrote in message
news:9liaf0$cc1$1...@slb7.atl.mindspring.net...

> >
> >I suppose you could take a dim, mechanistic view of humanity, that
> >they all move underground as a matter of survival, and adopt a more
> >mechanistic, expressionless culture due to their confines. Saw a
> >Star Trek TOS episode about that recently. I guess that is the sort
> >of non-magical survival excuse that I asked for it. I just don't
> >like the aesthetic consequences. :-)
>
> Why must cavemen have a "mechanistic, expressionless culture"?

'Cuz it's friggin' gloomy down there. Humanity is a creature of the light
and open spaces.

> If the society is a Dyson sphere with a population a billion times
> that of today, a million people getting blown up each day will result
> in approximately the same death rate as being struck by lightning in
> our society.

Why would only a million people get blown up a day? Certainly if you can
build a Dyson sphere, you can blow up the whole thing. Also it should be
noted that if you set off a mere H-bomb in any metropolitan area of the USA,
you'd kill way more than 1 million people. Imagine the catastrophe of
future weapons and population densities.

> Since the point of terrorism is to inspire fear, in such a society the
> terrorists may just give up.

Actually, the more I think about it, the more I see that they won't give up,
and the society will have to adopt elaborate covert operations technologies.
Measure for countermeasure. It may not quite be a police state, but if you
find yourself under suspicion you're going to be in big trouble.

> The Sun gives off that much energy in less than two days. In the Dyson
> sphere, nukes are popguns. If you want to write a terrorism story
> write one about a madman who gets control over one the main power
> beams or maybe a madman who sends space colonies colliding with each
> other.

Bear in mind I said 500 years hence. I think cannibalizing our entire solar
system in only 500 years is wildly optimistic. Or pessimistic, if you're an
environmentalist.

Incidentally, why would such people be "madmen?" If you could flick a
switch and end 1 billion people's lives at no consequence to yourself, it's
not an insane thing to do. It's just a larger threshold of "collateral
damage."

The whole problem is, people really shouldn't be given that switch because
they're such assholes. Yet it seems that they're going to get that switch,
sooner or later. How does the human race survive that kind of individual
perogative of power?

Jordan S. Bassior

unread,
Aug 17, 2001, 7:22:28 AM8/17/01
to
Brandon J. Van Every said:

>'Cuz it's friggin' gloomy down there. Humanity is a creature of the light
>and open spaces.

The technology is up to interstellar travel, but it can't simulate sunlight? Or
excavate a large enough cavern to avoid claustrophobia? Or _simulate_ a broad
vista?

Karl M. Syring

unread,
Aug 17, 2001, 8:25:53 AM8/17/01
to
"Jordan S. Bassior" <jsba...@aol.com> schrieb im Newsbeitrag
news:20010817072228...@mb-mn.aol.com...

> Brandon J. Van Every said:
>
> >'Cuz it's friggin' gloomy down there. Humanity is a creature of the
light
> >and open spaces.
>
> The technology is up to interstellar travel, but it can't simulate
sunlight? Or
> excavate a large enough cavern to avoid claustrophobia? Or _simulate_ a
broad
> vista?

Perhaps Brandon was often confined to tight spaces as a child ?
I myself think, there should be a possibility to go out into the wilderness,
even if this is very time-consuming and expensive to do so. May be, you
could invent some high risk sports that would help to get rid of potential
troublemakers.

Karl M. Syring

Jordan S. Bassior

unread,
Aug 17, 2001, 8:57:13 AM8/17/01
to
Karl M. Syring said:

>I myself think, there should be a possibility to go out into the wilderness,
>even if this is very time-consuming and expensive to do so.

Living underground doesn't mean spending all one's time underground, any more
than living under roofs means that one can never walk under the sky.

jeff suzuki

unread,
Aug 17, 2001, 10:31:39 AM8/17/01
to
"Brandon J. Van Every" wrote:
>
> Now, when
> political difficulties arise between factions, what prevents them from
> summarily nukeing one another?

What kept the SU and the US from nuking each other?

> Problem: it only takes 1 terrorist and 1 nuke to blow
> up an orbital habitat.

Now, if you're talking acts of terrorism, the point to keep in mind
is that terrorism is about terror; it's not about killing people
(although this is the usual means by which it is achieved). A
terrorist action can be as simple as dropping a coke can into a turbine
generator that powers half a continent.

There is a fine line for terrorists: they must be damaging enough
to get noticed, but not so damaging that they get noticed by
The Big Boys. (Has Qaddafi done anything big since bombing the
Berlin nightclub?) Nuking a planet or orbital habitat would probably
qualify.

The goal of a terrorist is to change their society. They choose to
do so by "proving" that the society as it exists cannot provide the
citizen with the basic needs. Protection is one of the basic needs,
but so is (in our modern society) power (remember the various threats
to take out the power grid during the 70s?), water (LSD in the
water supply?) and entertainment (those poor baseball players, who
only make more per game than I do per year...yeah, I feel _really_
sorry for them...).

Jeffs

Karl M. Syring

unread,
Aug 17, 2001, 10:19:00 AM8/17/01
to
"Jordan S. Bassior" <jsba...@aol.com> schrieb
> Karl M. Syring said:
>
> >I myself think, there should be a possibility to go out into the
wilderness,
> >even if this is very time-consuming and expensive to do so.
>
> Living underground doesn't mean spending all one's time underground, any
more
> than living under roofs means that one can never walk under the sky.

Of course, you are right, but the "no escape" situation was exactly the
point of my posting. Think of a settlement on of the Jupiter moons where you
are kill after 10 minutes by the cosmic death rays if you go outside.

Karl M. Syring

jeff suzuki

unread,
Aug 17, 2001, 10:37:54 AM8/17/01
to
"Brandon J. Van Every" wrote:
>
> I'm not convinced that most residents would want to live underground.

Depends on how attractively packaged it is.

Lots of people live in high rise condos, where they pay more for less
space and less privacy than someone in a detached single family home.
Why do they do it? It's a status symbol, for one. There is a
semblance of security, for two.

And it's not that far from living underground. "High rise" doesn't
mean much when you're surrounded by buildings equally high. Having
lived in buildings like that, I know that you don't get direct
sunlight for very much of the year (maybe not at all, if you're in the
part of the building facing the interior courtyard).

> I suppose you could take a dim, mechanistic view of humanity, that they all
> move underground as a matter of survival, and adopt a more mechanistic,
> expressionless culture due to their confines.

Why would underground living be any more mechanistic or expressionless
than living aboveground?

Jeffs

James Nicoll

unread,
Aug 17, 2001, 10:46:00 AM8/17/01
to
In article <3B7D2B60...@bu.edu>, jeff suzuki <je...@bu.edu> wrote:
>
>There is a fine line for terrorists: they must be damaging enough
>to get noticed, but not so damaging that they get noticed by
>The Big Boys. (Has Qaddafi done anything big since bombing the
>Berlin nightclub?) Nuking a planet or orbital habitat would probably
>qualify.

You may have heard of the Lockerbie plane bomb? That involved
Libyans well after the retaliation for the La Belle Disco bomb.

Although I could have sworn reading the Syrians were actually
the prime suspects (Off-limits to the US during the Cold War, though)
for the nightclub, I can't find support for this online.

James Nicoll

jeff suzuki

unread,
Aug 17, 2001, 10:50:39 AM8/17/01
to
"Brandon J. Van Every" wrote:
>
> I posited a culture of dissent, where some wackos really are crazy
> enough to strap a nuke to their chests and take out a city in a suicide run.
> The question is, what can you do to prevent this sort of thing

Bottom line: you can't. If someone is truly wacko enough to do this,
there is _nothing_ you can do to stop them. (It's a Secret Service
truism: you cannot prevent a truly determined assassin from killing
the President. All you can do is make sure that aforesaid truly
determined assassin is dead afterwards. Most people aren't that
determined, so the knowledge that they won't survive the attempt is
deterrence enough.)

But the wackos aside...the "rational" terrorists have to think in
terms of what they gain from such an act. In most cases, what they
gain will be far, far outweighed by what they lose: killing millions
of innocent civilians will surely allow The Guvmint to take _whatever_
action it finds necessary to exterminate (note the verb) the
terrorists, and no one will cry one tear over them.

For example, let's consider the Middle East situation. Every
terrorist claims that "it" (whatever "it" is) is the fault of the
evil Americans. However, if you look at the targets, the targets
are generally 1) military personnel, 2) embassies located in ME
countries. #1 knows the risks when they sign up. #2, well, the
political equation is that if we are getting bombed every week,
maybe it's _not_ worth the effort...which is exactly what the
terrorists are after.

Bombing the WTC was, in Talleyrand's phrase, worse than a mistake:
it was a blunder. The one thing it would _not_ have done is get
the Americans out of the Middle East. (For one thing, it's too
distant: it's easy to say "Gee, our people are getting killed in
Saudiaqia, maybe we should get out of Saudiaqia..." It's a lot
harder to rationalize, "Gee, our people are getting killed in NYC
because of our presence in Saudiaqia, maybe we should get out of
Saudiaqia...")

Jeffs

Mike Williams

unread,
Aug 17, 2001, 4:28:52 AM8/17/01
to
Wasn't it Brandon J. Van Every who wrote:

>Imagine a planetary colonization scenario 500 years in the future. There
>are no aliens, just human beings expanding outwards from Earth. The rate of
>expansion is slow enough that it takes a long time to travel between star
>systems and humanity simply hasn't run into any aliens yet. "Slow" can be
>whatever velocity and whatever technology you're comfortable with positing
>in 500 years' time.

With long travel times, you end up with a situation rather like the Old
West. Each planet ends up being politically isolated for all practical
purposes. If you've only got light speed communications and sub light
speed transport, it's going to take decades to respond to things that
happen in other star systems. By the time the United Planets battle
cruiser arrives at Epsilon Eridani III the conflict that they went to
pacify will be ancient history.

I'd guess that the majority of star systems wouldn't be suitable for
colonisation, so the distance from one colony to its nearest neighbour
colony might well be 10 or more light years.

The only possible way for a non-FTL galactic enforcement agency to work
would be to post UP forces at each colony planet. However, you've then
got the problem of finding people you can trust to be incorruptible and
ideologically pure, and who wants the boring job of manning the UP
presence for the years when their intervention is not required. Such
garrison forces would have to be pretty well equipped, because the
nearest relief won't even know there's a problem for over a decade.

I can't believe that such garrisons would be practical, particularly if
their ideology is democratic, so there's little alternative but to let
each planet sort out its own ideological conflicts in whatever ways they
see fit.

--
Mike Williams
Gentleman of Leisure

Isaac Kuo

unread,
Aug 17, 2001, 1:33:58 PM8/17/01
to
"Brandon J. Van Every" <vane...@3DProgrammer.com> wrote in message news:<B51f7.1968$D4.1...@newsread1.prod.itd.earthlink.net>...

>But then again, how does this in any way erase the factions/religious
>terrorist problem? I didn't posit a homogenous culture with uniform moral
>codes. I posited a culture of dissent, where some wackos really are crazy
>enough to strap a nuke to their chests and take out a city in a suicide run.
>The question is, what can you do to prevent this sort of thing, if the
>industrial materials for nuclear fission are commonplace?

If I ever write a SF novel, this is one of the two themes I'll write
about. I've mulled over lots of possible scifi/fantasy story ideas,
and "the nuclear terrorism problem" is the one which I've developed
the most.

I didn't go so far into the future as to worry about interstellar
travel, just some colonization as far as Mars.

Basically, what IS the solution to the "nuclear terrorist" problem?
In the past, there have been only two ways to prevent terrorism--by
reducing oppression to eliminate the cause behind terrorism and
by absolute no-holds-barred totalitarianism.

The problem is that the former solution merely reduces terrorism.
There's still wackos out there and contradictory fanatic groups
which will never all be satisfied. If the terrorists in question
have ready access to nuclear bombs (or worse), then even this
reduced number of terrorist acts is unacceptable.

The latter solution, while horrific to Western liberal ideals,
may be the only option. We've always sacrificed some degree
of freedom for the sake of safety--this is just taking that
principle to an extreme.

Where does a science fiction story fit into this? Well, you
start off with a multi-national world with one particularly
powerful superpower. After a few terrorist acts, the citizens
of this superpower (including their leadership) realizes that
the only way for the police state to eliminate nuclear terrorism
is for ALL of humanity to be under its regime. Voila! The
motivation for a global takeover--one with an argueably logical
and justifiable purpose. Other countries will naturally resist,
of course (freedom-loving and otherwise). In this war, both
sides are justified.

In the story-line I've been mulling about, this superpower
acheives global domination in a relatively short time--ironically
by completely nuking a few particularly defiant countries
to make an example of them (the President only cares about
saving American lives, not foreigners). That just leaves
the measly Martian colony, which has a chance mainly because
that superpower has no space warships (nor does anyone else).
This gives me lots of excuses to bring in my endless
speculative ideas on near-future interplanetary warfare
technology and tactics.

>Or if you can't
>prevent it, how does humanity survive a concerted war of extermination by
>some nuthatch faction? Seems like any individual egg can always be nuked,
>so it's only a question of having lotsa eggs, and hiding them.

>Maybe that's an answer. Forgetting where the hell you came from. Thorowing
>out your technological knowledge. Hiding on a distant spiral arm of the
>galaxy, away from all the trouble. Consorting with apes.

With your assumption of interstellar travel, there is the
possibility of "running away". These ideas were explored in
the later Dune books, in a way I really liked.

Isaac Kuo

Tapio Erola

unread,
Aug 17, 2001, 4:15:59 PM8/17/01
to
jeff suzuki <je...@bu.edu> writes:

> Bottom line: you can't. If someone is truly wacko enough to do this,
> there is _nothing_ you can do to stop them. (It's a Secret Service
> truism: you cannot prevent a truly determined assassin from killing
> the President. All you can do is make sure that aforesaid truly
> determined assassin is dead afterwards. Most people aren't that
> determined, so the knowledge that they won't survive the attempt is
> deterrence enough.)

Which gets to one of the more yucky possibilities of using humans as
weapons.

Consider a scenario where some faction develops a 100% efficient method
of laundering someone's brain, producing troops loyal to death to
the said faction (or even single person).

--
Tapio Erola (t...@rieska.oulu.fi) No mail to t...@rak061.oulu.fi please!

"Say what you will about the sweet miracle of unquestioning faith.
I consider the capacity for it terrifying." --Kurt Vonnegut, Jr.

Hop David

unread,
Aug 17, 2001, 4:38:21 PM8/17/01
to

"Brandon J. Van Every" wrote:

> "Brett Evill" <ev...@netspeed.com.au> wrote in message
> news:3b7c...@news.netspeed.com.au...
> > >
> > >However, humans are not entirely peaceful and not without their
> differences.
> > >The vast bureaucracy of stellar expansion allows for a wide range of
> > >political opinion.
> >
> > Why is there a vast bureaucracy?
>
> Because the United Planets is a bureaucracy, and it attempts to cover a vast
> distance.
>
> > How is it practicable given lightspeed
> > communications and slow travel? How can it enforce its decrees on
> > distant planets?
>
> Who said it was practical or terribly enforceable? This is a Wild West
> scenario. People somewhat obey laws because that's what they were
> conditioned to obey on another world, and it helps keep most people alive.
> But now they're out on the frontier, and lotsa people threaten to get ugly.
> However, the United Planets does have a few detachments of marines in orbit
> who are perfectly willing to make scorched earth of the whole place if
> anyone gets too out of hand....

The Romans had trouble holding onto their vast empire because of limited
communication and transportation.

Maybe a global empire is possible with modern communication and transportation
(don't see it happening soon, though). But an empire across the nine planets
would be a lot harder to maintain. When detachments of marines are months or
years away, it'd be much easier for rebels to evade the troops or launch
counteroffensives.

I believe it much more likely that several entities would be enforcing laws.

If transportation improved drastically, maybe there could be a system wide
empire. For example in _Martian Rainbow_ by Dr. Forward, earth forces are able
to reach Mars in a short time using an anti-matter drive.

-- Hop
http://clowder.net/hop/index.html
http://www.tabletoptelephone.com/~hopspage/HopsTiles.html

Brandon J. Van Every

unread,
Aug 17, 2001, 5:41:25 PM8/17/01
to

"Jordan S. Bassior" <jsba...@aol.com> wrote in message
news:20010817065853...@mb-mn.aol.com...

>
> This is supposed to be 500 years in the future, remember? Their houses
could be
> equipped not only with filtration systems and active nanotech defenses,
but
> also with at least short- to moderate-duration full-cycle life support
systems.

Hmm, I suppose geothermal tapping is an alternative to underground nuclear
power plants. Either could work. Maybe also long coils that pick up the
magnetic field of the planet.

Seems like to be defensible, it would have to rely on a lot of elevators. I
guess elevators are always going to be breaking down. It would be like mass
transit, only vertical. Isn't bombing just a matter of dropping nukes down
the appropriate shafts? If you make the shafts constricted, then you can't
move as many people up and down. You're not going to be able to close heavy
duty blast shields on *all* shafts, there's always going to be that one
shaft like on the Death Star that's left open. It was hard to remember to
lock all the doors on English castle mansions too. It was a big job, an all
day job.

> >A military
> >installation will be safer underground, and maybe under some
circumstances
> >they could protract a nuclear war, but if all of the civvies are dead
then
> >what are they fighting for?
>
> Why are all the civvies dead?

Because people spend inherently less money on civilian defense. Civvies are
not expected to be in combat, soldiers are. If you believe that civvies
*will* be in combat, then you have no division between civilian and military
in your society. Everyone's armed and fortified.

> Oh, and "revenge" is the obvious answer to your question.

Revenge becomes meaningless in a Mutual Assured Destruction scenario.
Almost everyone on both sides is already dead.

> You said "500 years in the future," but you seem to be assuming that the
only
> technological progress has been in the fields of space travel and
_offensive_
> weaponry. This makes no sense -- extensive space travel implies advanced
life
> support technology;

Ok, fine, granted. People can live in spectacular style underground.

> command of greater energies for offense implies that these
> same energies may also be used for defense.

Why? It's easy to destroy. It is not so easy to contain.

> >Saw a Star Trek TOS episode about that recently.
>
> Has it occurred to you that you've partially exposed your own fallacy
here?

Has it occurred to you that I'm not married to anything at present? I *did*
think it was a reasonable cultural projection in that particular episode's
fiction, even if it isn't an inevitable one.

> Scan incoming persons and cargoes for nuclear weapons. Have active
internal
> surveillance inside your habs.

So, they fake out the sensors, and the planet goes BOOM. You'd really need
a hermetic police state to keep this sort of thing from happening.

> Put active defenses on your habs.

Again, an exceedingly paranoid society, with no division between civilian
and military life.

Maybe the paranoia could be delegated to the robots, leaving humans free to
have their usual existence? Robots are the hardasses with all the stress,
which doesn't happen to bother them. Every once in awhile, a human is
wrongly accused and is torn to shreds. But that is deemed an acceptable
cost for the safety of human society from Islamic terrorists.

Brandon J. Van Every

unread,
Aug 17, 2001, 6:15:10 PM8/17/01
to

"jeff suzuki" <je...@bu.edu> wrote in message
news:3B7D2B60...@bu.edu...

> "Brandon J. Van Every" wrote:
> >
> > Now, when
> > political difficulties arise between factions, what prevents them from
> > summarily nukeing one another?
>
> What kept the SU and the US from nuking each other?

During the period of US nuclear monopoly, it's mainly because the US govt.
was (1) sane (2) democratic. The people wouldn't stand for it. There *was*
a Preventive War movement, but it was a minority movement.

Once the Soviets got the A-bomb, there were practical matters. The US
really didn't have enough A-bombs to completely polish off the Soviets in 1
strike. Much less vice versa. This wasn't known to the public, but it was
known to the military planners who actually had the authority to carry out
such a first strike.

The USA got the H-bomb first. Meanwhile I believe both sides had developed
second-strike capabilities, thus cementing MAD policy. Also there was a lot
of uncertainty about who really had what capability. The USA was again
constrained by being a democratic society. If the Soviets had gotten the
H-bomb first, they may not have been. Still, the reason MAD worked is
because both sides were (1) sane.

MAD goes out the window if one side actually desires their own martyred
death.

> > Problem: it only takes 1 terrorist and 1 nuke to blow
> > up an orbital habitat.
>
> Now, if you're talking acts of terrorism, the point to keep in mind
> is that terrorism is about terror; it's not about killing people
> (although this is the usual means by which it is achieved). A
> terrorist action can be as simple as dropping a coke can into a turbine
> generator that powers half a continent.

Hmm, the media value of terrorism, as opposed to its genocidal value.
That's an interesting angle to explore.

> There is a fine line for terrorists: they must be damaging enough
> to get noticed, but not so damaging that they get noticed by
> The Big Boys. (Has Qaddafi done anything big since bombing the
> Berlin nightclub?) Nuking a planet or orbital habitat would probably
> qualify.

Well, what if your organization is big enough that you *are* one of the Big
Boys? You're just trying to eliminate the competition. Then again, as one
of the Big Boys you have a lot more to lose. I guess the growth of
organizations and agendas do put a throttle on conspirational action. The
attacks have to meet the goals of the organization, which generally aren't
doomsday, but mind control.

Still, there is the problem of *individual* crackpots. The Timothy McVeighs
of the 25th Century.

Brandon J. Van Every

unread,
Aug 17, 2001, 6:22:13 PM8/17/01
to

"Tapio Erola" <t...@localhost.localdomain> wrote in message
news:m3u1z6c...@localhost.localdomain...

>
> Which gets to one of the more yucky possibilities of using humans as
> weapons.
>
> Consider a scenario where some faction develops a 100% efficient method
> of laundering someone's brain, producing troops loyal to death to
> the said faction (or even single person).

You don't need a human for that, a cosmetically acceptable Terminator will
suffice.

To revisit the Secret Service truism, maybe in 500 years humanity will have
recognized the reality that the President's life *must* be protected. By
technology. Humans may not have the attention spand and reaction time to
deal with ongoing threats, but robots could.

Balancing this is the necessary interface of vulnerability to other
locations and other members of society. You can't live a "normal" life if
everything must be safe, and to date, humanity has always eschewed safety.
Danger isn't just a part of life, it's sexual.

Another question is how often in 500 years did a planetary terrorist nukeing
prove to be a problem, and what was the body count.

Jordan S. Bassior

unread,
Aug 17, 2001, 6:35:48 PM8/17/01
to
Brandon J. Van Every said:

>Hmm, I suppose geothermal tapping is an alternative to underground nuclear
>power plants. Either could work. Maybe also long coils that pick up the
>magnetic field of the planet.

Don't know if the last-named would work, but either of the first two sound
reasonable. Um, this is 500 years in the future, they probably aren't limted to
nuclear _fission_ either -- I'd be very surprised if _no_ version of nuclear
fusion worked well enough to be used in massive stationary power plants. Heck,
for _that_ application there are (dangerous) ways we know to obtain fusion
energy _now_.

>Seems like to be defensible, it would have to rely on a lot of elevators. I
>guess elevators are always going to be breaking down.

Um, and I think the capabilty to repair elevators will still exist 500 years in
the future. In fact, they'll probably have uber-elevators beyond our current
engineering technology :)

>It would be like mass
>transit, only vertical.

They'd probably have horizontal mass transit, too.

>Isn't bombing just a matter of dropping nukes down
>the appropriate shafts?

If you design the shafts _very_ badly. The logical way to design them would be
with no direct straight line path leading from the shelters to the outside
world. There are a number of ways this can be done. The basic technology of
avoiding a direct straight path gateway was invented in several millennia BCE
when the first fortresses were constructed. The technology of applying it to
elevators in fortresses was invented by the Belgians or French in the early
20th century.

>You're not going to be able to close heavy
>duty blast shields on *all* shafts, there's always going to be that one
>shaft like on the Death Star that's left open.

Why?

> It was hard to remember to
>lock all the doors on English castle mansions too. It was a big job, an all
>day job.

My word! You'd need to invent some sort of "thinking machine" to remember to do
that. And some sort of powered system to close the doors. Do you think they'll
have that technology 500 years from now?

(Hints: What are you typing your posts on? How do the doors in a supermarket
open and close?)

>> Why are all the civvies dead?
>
>Because people spend inherently less money on civilian defense. Civvies are
>not expected to be in combat, soldiers are. If you believe that civvies
>*will* be in combat, then you have no division between civilian and military
>in your society. Everyone's armed and fortified.

Nonsense. Building civil defense shelters, or simply subterranean houses, does
not require that the civilians be armed and fortified to the degree miltary
installations (in that future world) are. Nor does this mean that the civilians
will be preferentially targetted in warfare, unless you assume that "stratregy"
and "tactics" were forgotten in the intervening half-millennium.

Merely giving the civilians shelters good enough to save them from anything but
hits right above them, and life support systems good enough to let them survive
if the outside environment is NBC-contaminated, in particular does not require
any such absurd degrees of fortifications.

>Revenge becomes meaningless in a Mutual Assured Destruction scenario.
>Almost everyone on both sides is already dead.

Revenge is why the victim launches, too.

And who said "Mutual" Assured Destruction? Planet A may have listened to your
advice and not sheltered their civilian population; Planet B may have listened
to mine and constructed orbital defenses and civil defense shelters. In that
situation, Planet A may be a dead world, but Planet B is still alive.

>> command of greater energies for offense implies that these
>> same energies may also be used for defense.
>
>Why? It's easy to destroy. It is not so easy to contain.

Because the "destruction" must be done at a distance. The advantage of the
attacker is initiative; that of the defender is logistics.

>> Scan incoming persons and cargoes for nuclear weapons. Have active internal
>> surveillance inside your habs.
>
>So, they fake out the sensors, and the planet goes BOOM.

A single bomb, gotten past the sensors, is powerful enough to destroy the
entire planet?

>> Put active defenses on your habs.
>
>Again, an exceedingly paranoid society, with no division between civilian
>and military life.

Why is that "exceedingly paranoid?"

Are you aware that, for all forms of attack other than ballistic missiles,
America, and other Great Powers, _already_ have "active defenses" deployed?

Jordan S. Bassior

unread,
Aug 17, 2001, 6:43:39 PM8/17/01
to
Jeff Suzuki said:

>Bombing the WTC was, in Talleyrand's phrase, worse than a mistake:
>it was a blunder. The one thing it would _not_ have done is get
>the Americans out of the Middle East. (For one thing, it's too
>distant: it's easy to say "Gee, our people are getting killed in
>Saudiaqia, maybe we should get out of Saudiaqia..." It's a lot
>harder to rationalize, "Gee, our people are getting killed in NYC
>because of our presence in Saudiaqia, maybe we should get out of
>Saudiaqia...")

Indeed -- if it does _anything_, it might make us decide "Gee, people are being
killed in NYC by those bastards in Inimica, maybe we'd better take out
Inimica."

Brandon J. Van Every

unread,
Aug 17, 2001, 6:44:16 PM8/17/01
to

"Isaac Kuo" <mec...@yahoo.com> wrote in message
news:acc26c07.01081...@posting.google.com...

>
> Where does a science fiction story fit into this? Well, you
> start off with a multi-national world with one particularly
> powerful superpower. After a few terrorist acts, the citizens
> of this superpower (including their leadership) realizes that
> the only way for the police state to eliminate nuclear terrorism
> is for ALL of humanity to be under its regime. Voila! The
> motivation for a global takeover--one with an argueably logical
> and justifiable purpose. Other countries will naturally resist,
> of course (freedom-loving and otherwise). In this war, both
> sides are justified.

Questions: is the nuclear terrorist aligned with a nation? i.e. does the
nuke have a clear return address? If it does, that country ceases to exist.
Highly unlikely that a rogue nation has any treaties that would cause the
rest of the world to cease to exist, at least at this point in history. One
might hope that by the time they have such treaties, they're also a more
responsible nation.

If the nuclear terrorist is not clearly aligned with a nation, then the USA
(for instance) threatens to invade/nuke all suspect nations. That could be
a really nasty showdown, like, the USA against the entire Middle East. It
could result in the world being annihilated, as uninvolved nations would
rightfully fear for their own soverignty. But clearly, the USA is going to
do something about having been nuked. Someone must pay.

Probably the USA says the terrorist *is* from some particular nation,
regardless of whether he is or isn't. The clear return address is
established simply because it's not politically feasible to wage war against
the entire Middle East.

What if the terrorist blows up his bomb in a country that's not as tough as
the USA? What if it's Germany? What if it's a non-nuclear country? It
would be a good terrorist strategy to bomb someone that isn't likely to
retalliate, and isn't likely to marshal the political will to get a UN
retalliation. But again, if there's a provable return address, the UN is
going to stomp that country.

> In the story-line I've been mulling about, this superpower
> acheives global domination in a relatively short time--ironically
> by completely nuking a few particularly defiant countries
> to make an example of them (the President only cares about
> saving American lives, not foreigners).

Presumably these are countries without nuclear defense treaties. I don't
see how the USA can dominate Europe, Russia, China, or pretty soon even
India. Do you assume a common UN co-prosperity sphere? Nuclear club
incumbents vs. terrorist nukeing rogue nations?

Jordan S. Bassior

unread,
Aug 17, 2001, 6:45:18 PM8/17/01
to
Issac Kuo said:

>Basically, what IS the solution to the "nuclear terrorist" problem?
>In the past, there have been only two ways to prevent terrorism--by
>reducing oppression to eliminate the cause behind terrorism and
>by absolute no-holds-barred totalitarianism.

Assuming you control the territory the terrorists base from, police methods
suffice. If you don't, you tell the power that does control said territory to
cooperate to suppress the terrorists, and if they don't, you go to war with
them, conquer them, and install a new government that does.

Brandon J. Van Every

unread,
Aug 17, 2001, 9:09:15 PM8/17/01
to

"Jordan S. Bassior" <jsba...@aol.com> wrote in message
news:20010817183548...@mb-cr.aol.com...

>
> >Isn't bombing just a matter of dropping nukes down
> >the appropriate shafts?
>
> If you design the shafts _very_ badly. The logical way to design them
would be
> with no direct straight line path leading from the shelters to the outside
> world. There are a number of ways this can be done. The basic technology
of
> avoiding a direct straight path gateway was invented in several millennia
BCE
> when the first fortresses were constructed. The technology of applying it
to
> elevators in fortresses was invented by the Belgians or French in the
early
> 20th century.

Yeah but you don't really want to increase transit times for all the
citizens that use them regularly and are nuked rarely.

> >You're not going to be able to close heavy
> >duty blast shields on *all* shafts, there's always going to be that one
> >shaft like on the Death Star that's left open.
>
> Why?

Because there will always be some shaft somewhere that someone forgot to
close, or didn't quite build properly, or has a weaker security protocol,
etc. A chain is only as strong as its weakest link.

Brandon J. Van Every

unread,
Aug 17, 2001, 9:11:34 PM8/17/01
to

"Jordan S. Bassior" <jsba...@aol.com> wrote in message
news:20010817184518...@mb-cr.aol.com...

> Issac Kuo said:
>
> >Basically, what IS the solution to the "nuclear terrorist" problem?
> >In the past, there have been only two ways to prevent terrorism--by
> >reducing oppression to eliminate the cause behind terrorism and
> >by absolute no-holds-barred totalitarianism.
>
> Assuming you control the territory the terrorists base from, police
methods
> suffice. If you don't, you tell the power that does control said territory
to
> cooperate to suppress the terrorists, and if they don't, you go to war
with
> them, conquer them, and install a new government that does.

They deny that they harbor terrorists and the world press is taking their
side. What do you do now?

Brandon J. Van Every

unread,
Aug 17, 2001, 9:36:23 PM8/17/01
to

"Jordan S. Bassior" <jsba...@aol.com> wrote in message
news:20010817183548...@mb-cr.aol.com...

>
> > It was hard to remember to
> >lock all the doors on English castle mansions too. It was a big job, an
all
> >day job.
>
> My word! You'd need to invent some sort of "thinking machine" to remember
to do
> that. And some sort of powered system to close the doors. Do you think
they'll
> have that technology 500 years from now?

Do you think it will be possible to sever a segment of the communications
net, undetected? Like when Bell is digging for new telephone cables or
something?

> (Hints: What are you typing your posts on?

The Internet was designed to survive in tatters, not to perform reliably all
the time.

> How do the doors in a supermarket open and close?)

Ever seen 'em break? Ever thrown a firebomb through 'em? You aren't
seriously proposing that they open and close for everyone who passes
through, are you?

> >> Why are all the civvies dead?
> >
> >Because people spend inherently less money on civilian defense. Civvies
are
> >not expected to be in combat, soldiers are. If you believe that civvies
> >*will* be in combat, then you have no division between civilian and
military
> >in your society. Everyone's armed and fortified.
>
> Nonsense. Building civil defense shelters, or simply subterranean houses,
does
> not require that the civilians be armed and fortified to the degree
miltary
> installations (in that future world) are.

You aren't going to be in your civil defense shelter when the terrorist bomb
goes off. You're going to be walking down your subterranean street minding
your own business, or sitting in your home. *All* facilities that civilians
use would have to be specially constructed. Roads, offices, pubs,
everything. And... here's the kicker... it's still not going to stop a
determined terrorist from placing the nuke in such a way as to blow somebody
to kingdom come. A lot of people, in fact. The basic problem is that human
circulation is too "open." Humans would have to adopt an "access tube"
fortification mentality to everything, checkpoints at every intersection,
blast doors constantly opening and closing, to contain nuclear terrorists.
Then it would be a question of how many blast doors a nuke can take out.

> Nor does this mean that the civilians
> will be preferentially targetted in warfare, unless you assume that
"stratregy"
> and "tactics" were forgotten in the intervening half-millennium.

Citizens were a legitimate target of the military industrial complex in WW
II. The thing is, there are enough nukes to wipe out everything around many
times over. If there's a full-blown war, as opposed to some terrorist
attack, all the civvies are dead.

> Merely giving the civilians shelters good enough to save them from
anything but
> hits right above them,

In a full-blown war with the guidance systems of 500 years from now, the
hits will be right above 'em. There's no reason for 'em to miss, only a
possibility of 'em being intercepted.

> >Revenge becomes meaningless in a Mutual Assured Destruction scenario.
> >Almost everyone on both sides is already dead.
>
> Revenge is why the victim launches, too.

Right, the revenge is already over with. If there are squabbling military
factions left after both sides blow their wads, that's lucky.

> And who said "Mutual" Assured Destruction? Planet A may have listened to
your
> advice and not sheltered their civilian population; Planet B may have
listened
> to mine and constructed orbital defenses and civil defense shelters. In
that
> situation, Planet A may be a dead world, but Planet B is still alive.

Your measures are inadequate. The terrorists can strike at will on both
worlds.

> >> command of greater energies for offense implies that these
> >> same energies may also be used for defense.
> >
> >Why? It's easy to destroy. It is not so easy to contain.
>
> Because the "destruction" must be done at a distance.

Distance from what? A robot terminator android? The tip of a cruise
missile?

> The advantage of the attacker is initiative; that of the defender is
logistics.
>
> >> Scan incoming persons and cargoes for nuclear weapons. Have active
internal
> >> surveillance inside your habs.
> >
> >So, they fake out the sensors, and the planet goes BOOM.
>
> A single bomb, gotten past the sensors, is powerful enough to destroy the
> entire planet?

Why not? You're talking 500 years hence, about an interstellar civilization
with a vested interest in bigger and better star drives. Also, why does it
have to be a single bomb? It could be 50 bombs, and only 20 of 'em have to
get through to do the job.

I guess that's an interesting physics question. What's the smallest density
of matter that could destroy the largest expanse of a planet's surface? How
small can your Doomsday device be? What are the limits of undetectability?

What liberties do I have to shape my device like something else? Should all
ships with large spherical components be quarantined?

> Are you aware that, for all forms of attack other than ballistic missiles,
> America, and other Great Powers, _already_ have "active defenses"
deployed?

Remember Pearl Harbor.

Conrad Hodson

unread,
Aug 18, 2001, 3:17:10 AM8/18/01
to
On 17 Aug 2001, Joseph Hertzlinger wrote:

> On Fri, 17 Aug 2001 04:30:25 GMT, Brandon J. Van Every
> <vane...@3DProgrammer.com> wrote:
>
> >I suppose you could take a dim, mechanistic view of humanity, that
> >they all move underground as a matter of survival, and adopt a more

> >mechanistic, expressionless culture due to their confines. Saw a
> >Star Trek TOS episode about that recently. I guess that is the sort
> >of non-magical survival excuse that I asked for it. I just don't
> >like the aesthetic consequences. :-)
>
> Why must cavemen have a "mechanistic, expressionless culture"?

Caves are only one protective option. Let's assume, in such a situation,
that there would be several patterns of response:

One would be active defense--a substantial military/security force to
protect the planet. This might include civil defense type measures to
reduce casualties from partially successful or small-scale attacks.

One would be passive defense--and this could be done by individuals or
small groups in the absence of public policy, too. Think survivalists.
People choosing to live away from major target areas, storing supplies,
air filtration, stuff like that.

Dispersal: People who've had a close call from terrorism decide to get out
of Dodge, as such survivors have so often in Earth's history. The
geometry of an expanding sphere being what it is, if those people are
willing to just travel longer than the usual pioneering trip, they won't
be found by friend or foe for generations. (If they're using relativistic
sublight travel, this may be fairly simple to do with a high tau.)

Camouflage: combined with the dispersal. With a sufficiently advanced
technology, you could have quite a civilization that wouldn't need to show
from space at all. If you combine present-day permaculture techniques
with a bit of genetic engineering, your agriculture can look like an
ordinary ecosystem. If your energy use is efficient and your population
small, I can think of several ways to generate electricity that would be
invisible from space with _today's_ technology---from turbines submerged
in rivers to underground geothermal sites (how's an outside observer to
know that hot spring used to run hotter?) to nuclear fission or fusion
plants in anchored submarines. There are also some plausible ways to
camouflage solar panels and solar heaters, especially power ponds. If you
use air and water transport a lot, you can do without a road network
too. Insulate buildings really well, and they need little energy and have
IR signatures small enough to handle along with the visual
camouflage. Have all communication by buried cable or
tightbeam. Millions of people could live at a higher standard of living
than modern Americans (given little more than present technology) and be
undetectable from low orbit--and how long before a colony has more people
than that?


>
> If the society is a Dyson sphere with a population a billion times
> that of today, a million people getting blown up each day will result
> in approximately the same death rate as being struck by lightning in
> our society.
>

> Since the point of terrorism is to inspire fear, in such a society the
> terrorists may just give up.

Unless they have more local targets and objectives. In other words, if
the terrorists are religious fanatics trying to put the fear of God into
the horrible sinners of Habitat 23335, they may not care about opinion 40
million miles away.

>
> >The question is, what can you do to prevent this sort of thing, if
> >the industrial materials for nuclear fission are commonplace?
>

> Fission? That long-obsolete system? A kiloyear from today, a nuclear
> terrorist will have to steal the nuke from a museum.
>
> Let us calculate how much energy can be obtained from radioactive
> materials inside the Earth. The Earth's heat (derived from
> radioactivity) is about 40 terawatts. If the mean lifetime of the
> radioactive material is 10 billion years (much of it is thorium) and
> if the amount of energy from fission is 4 times the amount from
> radioactivity, that amounts to 5x10^31 Joules.


>
> The Sun gives off that much energy in less than two days. In the Dyson
> sphere, nukes are popguns. If you want to write a terrorism story
> write one about a madman who gets control over one the main power
> beams or maybe a madman who sends space colonies colliding with each
> other.

Or, consider how much space travel--just sheer shlepping of people and
stuff--it takes to settle our end of the Galaxy, or to convert the
Asteroid Belt into a Dyson-type cloud of habitats. Millions of ships
moving at very high relative speeds will be routine--that many potential
kinetic energy weapons makes it unlikely that any terrorist would _bother_
stealing that fission bomb out of the museum.

Conrad Hodson

Guns don't kill people, meteors kill people

Jordan S. Bassior

unread,
Aug 18, 2001, 4:52:14 AM8/18/01
to
Brandon J. Van Every said:

>They deny that they harbor terrorists and the world press is taking their
>side. What do you do now?

Go to war, probably. You've just described a situation that's going to lead to
at least local war, because the victim of massive terrorist attack isn't going
to tolerate it for very long. If "world opinion" is sufficiently on the side of
the aggressor state, this may lead to general war.

Jordan S. Bassior

unread,
Aug 18, 2001, 4:57:41 AM8/18/01
to
Brandon J. Van Every said:

>Yeah but you don't really want to increase transit times for all the
>citizens that use them regularly and are nuked rarely.

At that level of technology your "elevators" are probably vehicles running on
tracks rather than weights being hauled up and down shafts anyway. So it's not
that difficult to put a curve in the shaft somewhere, or set things up so that
there is at least one horizontal stretch between vertical stretches.

>> >You're not going to be able to close heavy
>> >duty blast shields on *all* shafts, there's always going to be that one
>> >shaft like on the Death Star that's left open.
>>
>> Why?
>
>Because there will always be some shaft somewhere that someone forgot to
>close, or didn't quite build properly, or has a weaker security protocol,
>etc. A chain is only as strong as its weakest link.

You're assuming that the subterranean hab is all one unit, with no
compartmentation or dispersal. But it would be rather easy to build both into
the design. Especially compartmentation, and (given the likely capabilities of
fusion powered plasma drills) even a high degree of dispersal.

And you're also assuming that the attacker has complete plans of the hab _and
knows the security protocols_. That's a fairly high degree of intelligence (in
the military sense) and protocols are changed when the perceived threat level
increases (such as in wartime).

Now, admittedly, an attack could be very highly effective against a poorly
designed hab given complete suprise and lax security -- _but how is that any
different from the real world?_

Jordan S. Bassior

unread,
Aug 18, 2001, 5:12:17 AM8/18/01
to
Brandon J. Van Every said:

>Do you think it will be possible to sever a segment of the communications
>net, undetected? Like when Bell is digging for new telephone cables or
>something?

Hard to say. It'll probably be even harder in the future than it is today to
sever _any_ of it (since the more modern nets are built with multiple
redundancy), and there will probably be sapient agents monitoring the nets for
problems like that.

>> How do the doors in a supermarket open and close?)
>
>Ever seen 'em break? Ever thrown a firebomb through 'em? You aren't
>seriously proposing that they open and close for everyone who passes
>through, are you?

No, my point is that the powered opening and closing of doors is a problem that
has been solved for the last 50-100 years.

>You aren't going to be in your civil defense shelter when the terrorist bomb
>goes off. You're going to be walking down your subterranean street minding
>your own business, or sitting in your home.

Simply being "subterranean" shelters you to a higher degree than being on the
surface would.

>*All* facilities that civilians
>use would have to be specially constructed. Roads, offices, pubs,
>everything.

Why? You don't need to prevent _all_ death and destruction for your civil
defense system to "work." All you need to do is _reduce_ the death and
destruction, and instead of your whole civilization being wiped out, you
instead take some percentage casualties.

And the survivors are very, very, VERY pissed off. Which is where deterrence
enters the picture.

>And... here's the kicker... it's still not going to stop a
>determined terrorist from placing the nuke in such a way as to blow somebody
>to kingdom come.

Well, sure, if merely killing "somebody" is one's objective. But so what?
That's been true since Mankind evolved.

>A lot of people, in fact.

The maximum is the number of people in any one hab.

>The basic problem is that human
>circulation is too "open." Humans would have to adopt an "access tube"
>fortification mentality to everything, checkpoints at every intersection,
>blast doors constantly opening and closing, to contain nuclear terrorists.

Not really. What one does is to scan persons entering or leaving a hab for
bombs. That's not necessarily all that hard, especially if you assume that
sensor technology advances in the next 500 years (we could do it right now with
real-world weapons, easily, but I'm assuming cold-triggered bombs).

Remember that, 500 years from now, most work-related "commuting" will probably
be cyber-commuting anyway. That means that you _don't_ have a _huge_ stream of
people entering and leaving habs constantly.

>Then it would be a question of how many blast doors a nuke can take out.

*sigh*

Please, do us a favor. Study how fortresses are actually constructed. Read up
on Cheyenne Mountain, in particular, because _that_ fort was built to take a
_direct_ nuclear hit and survive.

For one thing, you don't put all your blast doors _in line!!!_

>In a full-blown war with the guidance systems of 500 years from now, the
>hits will be right above 'em.

The ones which were targetted on them. Being subterranean means that they
_aren't_ killed by the ones targetted on the ones a few miles over.

> There's no reason for 'em to miss, only a
>possibility of 'em being intercepted.

You don't _primarily_ target an attack on the civilian populace. That's
_stupid_ -- that leads to a situation where you kill the wives and children of
the enemy military, which now has no reason not to spend their lives killing
_you_. You attack the enemy military _first_, then if there is leftover strike
capacity you _may_ attack civilian targets, to wear down the enemy economy and
society.

>Your measures are inadequate. The terrorists can strike at will on both
>worlds.

Where are these terrorists basing from? Why can't the police there handle them?
And if the police there aren't trying, then it's not "terrorism", it's
disguised direct warfare, and the solution is to close one's borders,
annihilate or subdue the aggressor state, and then open your borders again when
the war's over.

>Distance from what? A robot terminator android?

Your "robot terminator androids" can of course get past the scanners at the
entries to the enemy's habs? Or the anti-personnel defenses?

>The tip of a cruise missile?

How did said "cruise missile" reach its target?

>> A single bomb, gotten past the sensors, is powerful enough to destroy the
>> entire planet?
>
>Why not? You're talking 500 years hence, about an interstellar civilization
>with a vested interest in bigger and better star drives. Also, why does it
>have to be a single bomb? It could be 50 bombs, and only 20 of 'em have to
>get through to do the job.

If _that's_ the case, then any physical transport to and from the planet (or
hab, or whatever) is going to be _very_ carefully monitored, and lots of active
defenses will be mounted on and around anything worth protecting.

>What liberties do I have to shape my device like something else? Should all
>ships with large spherical components be quarantined?

I think they'd be scanned in minute detail before being permitted to approach.

>> Are you aware that, for all forms of attack other than ballistic missiles,
>> America, and other Great Powers, _already_ have "active defenses"
>deployed?
>
>Remember Pearl Harbor.

Remember the sequel.

Brandon J. Van Every

unread,
Aug 18, 2001, 12:25:33 PM8/18/01
to

"Jordan S. Bassior" <jsba...@aol.com> wrote in message
news:20010818051217...@mb-cu.aol.com...

>
> >The basic problem is that human
> >circulation is too "open." Humans would have to adopt an "access tube"
> >fortification mentality to everything, checkpoints at every intersection,
> >blast doors constantly opening and closing, to contain nuclear
terrorists.
>
> Not really. What one does is to scan persons entering or leaving a hab for
> bombs. That's not necessarily all that hard, especially if you assume that
> sensor technology advances in the next 500 years (we could do it right now
with
> real-world weapons, easily, but I'm assuming cold-triggered bombs).

And we assume that counter-sensor technology advances right along with it.
You're back to needing an exceedingly controlled fortress everywhere.

> Remember that, 500 years from now, most work-related "commuting" will
probably
> be cyber-commuting anyway. That means that you _don't_ have a _huge_
stream of
> people entering and leaving habs constantly.

False. They'll be engaged in recreational tourism.

> Please, do us a favor. Study how fortresses are actually constructed. Read
up
> on Cheyenne Mountain, in particular, because _that_ fort was built to take
a
> _direct_ nuclear hit and survive.

How badass are your nukes of 500 years hence? Do you believe in offensive
or defensive technology?

> The ones which were targetted on them. Being subterranean means that they
> _aren't_ killed by the ones targetted on the ones a few miles over.

Long as you've got enough ammo to do all the jobs....

> You don't _primarily_ target an attack on the civilian populace. That's
> _stupid_ -- that leads to a situation where you kill the wives and
children of
> the enemy military, which now has no reason not to spend their lives
killing
> _you_. You attack the enemy military _first_, then if there is leftover
strike
> capacity you _may_ attack civilian targets, to wear down the enemy economy
and
> society.

Dude, you attack everyone, with enough ammo to blow everyone up 100 times
over.

> Where are these terrorists basing from? Why can't the police there handle
them?

'Cuz they're sneaky.

> >The tip of a cruise missile?
>
> How did said "cruise missile" reach its target?

Hovering waiting for blast doors to open. How about bore-bombs?

> >Why not? You're talking 500 years hence, about an interstellar
civilization
> >with a vested interest in bigger and better star drives. Also, why does
it
> >have to be a single bomb? It could be 50 bombs, and only 20 of 'em have
to
> >get through to do the job.
>
> If _that's_ the case, then any physical transport to and from the planet
(or
> hab, or whatever) is going to be _very_ carefully monitored, and lots of
active
> defenses will be mounted on and around anything worth protecting.

Yeah right. Very carefully monitor everything. Dude, it's not possible to
check everything. Not unless you only allow 1 ship in at a time along an
immense solar system corridor. The whole solar system would have to be
designed like a prison.

Jordan S. Bassior

unread,
Aug 18, 2001, 12:33:41 PM8/18/01
to
Branon J. Van Every said:

>And we assume that counter-sensor technology advances right along with it.

Where the distance to be scanned does not change, it historically hasn't. And
even if the distance didn't change, _right now_ you could not smuggle an atomic
bomb past (for instance) an airport security check.

>You're back to needing an exceedingly controlled fortress everywhere.

No, you just have to check people and goods coming onto and off your hab or
world. That's not really that hard to do, assuming you invest in such a system.

>> Remember that, 500 years from now, most work-related "commuting" will
probably
>> be cyber-commuting anyway. That means that you _don't_ have a _huge_ stream
of
>> people entering and leaving habs constantly.
>
>False. They'll be engaged in recreational tourism.

Most travelers, today, are commuters. And cyber-tourism to some degree can
replace physical tourism, too.

>How badass are your nukes of 500 years hence? Do you believe in offensive
>or defensive technology?

Both.

>Long as you've got enough ammo to do all the jobs....

... enough ammo _surviving interception_. That's an important caveat.

>Dude, you attack everyone, with enough ammo to blow everyone up 100 times
over.

Dude, you defend everyone, with enough ammo to blow up all the attackers 100
times over.

Cuts both ways.

>> Where are these terrorists basing from? Why can't the police there handle
them?
>
>'Cuz they're sneaky.

Oh, well, I bet _real_ terrorist groups would become much more powerful if they
learned your easy solution. Just be "sneaky."

>> How did said "cruise missile" reach its target?
>
>Hovering waiting for blast doors to open.

1) It was produced, from the factory, right outside the shelter? The enemy
didn't have to launch it from their own territory to get it to that point?

2) Why aren't the point defense systems blowing it to crap while it's waiting?
Heck, if it's "hovering", why aren't _policemen with pistols_ blowing it to
crap while it's waiting?

>How about bore-bombs?

"Bore-bombs," in your reality, also have the magic power to teleport across
interplanetary distances?

>Yeah right. Very carefully monitor everything. Dude, it's not possible to
>check everything. Not unless you only allow 1 ship in at a time along an
>immense solar system corridor. The whole solar system would have to be
>designed like a prison.

??? - Why can't you have multiple teams doing the checking?

Isaac Kuo

unread,
Aug 18, 2001, 4:41:18 PM8/18/01
to
"Brandon J. Van Every" <vane...@3DProgrammer.com> wrote in message news:<47hf7.18159$ZM2.1...@newsread2.prod.itd.earthlink.net>...

>"Isaac Kuo" <mec...@yahoo.com> wrote in message
>news:acc26c07.01081...@posting.google.com...

>>Where does a science fiction story fit into this? Well, you
>>start off with a multi-national world with one particularly
>>powerful superpower. After a few terrorist acts, the citizens
>>of this superpower (including their leadership) realizes that
>>the only way for the police state to eliminate nuclear terrorism
>>is for ALL of humanity to be under its regime. Voila! The
>>motivation for a global takeover--one with an argueably logical
>>and justifiable purpose. Other countries will naturally resist,
>>of course (freedom-loving and otherwise). In this war, both
>>sides are justified.

>Questions: is the nuclear terrorist aligned with a nation? i.e. does the
>nuke have a clear return address? If it does, that country ceases to exist.

Doesn't work if the nuclear terrorist lives too close to home.

Who are the British going to nuke in retaliation to an IRA
nuclear bombing? Who are the Israelis going to nuke in
retaliation to a PLO nuclear bombing? Who will we nuke in
retaliation to the next Tim McVeigh?

It's the problem of the Tim McVeighs which I'm most concerned
about. In a free society, there's still going to be a small
fraction of wackos who decide to just kill as many people as
they can even if their cause is not well thought out. All
that's really required is enough anger and/or desperation.
That'll cloud any judgement over the true repercussions of
one's actions.

The question is--how much risk to your own life do you deem
acceptable? Let's say an average of one guy goes "wacko"
in your city per year. If he goes on a school rampage with
an assault rifle or steals an M60 tank and ends up killing
less than a dozen people, none of whom you probably even
know personally, that's a risk you can live with. (In preference
to living 24/7 under the looking glass of a police state, I
assume.)

What if instead he goes to local Radio Shack and buys "Mr.
Fusion"--which is easily modified using instructions on
Internet II into a city killing device? You, your family,
your friends, are all killed. If not this year, next year or
the one after that. Is this an acceptable risk? Clearly
not. Even if it's a miserable life living 24/7 under the
microscope in a police state, it's preferable to an expected
life span under a decade.


>But clearly, the USA is going to
>do something about having been nuked. Someone must pay.

What did we do in retaliation to the Oklahoma City bombing?
We didn't retaliate to any other nations--and it's a good
thing we didn't.

What will we do in retaliation to the next Oklahoma City,
which actually destroys the entire city? Whether or not
the criminal is found and brought to justice is almost
besides the point. No amount of punishment on one person
can make up for the loss. If it turns into a regular
occurance, the ones who pay in the end may be everyone.

>>In the story-line I've been mulling about, this superpower
>>acheives global domination in a relatively short time--ironically
>>by completely nuking a few particularly defiant countries
>>to make an example of them (the President only cares about
>>saving American lives, not foreigners).

>Presumably these are countries without nuclear defense treaties. I don't
>see how the USA can dominate Europe, Russia, China, or pretty soon even
>India. Do you assume a common UN co-prosperity sphere? Nuclear club
>incumbents vs. terrorist nukeing rogue nations?

Nuclear defense treaties? Do you mean ones with the U.S.?
We can always just ignore them (like right now).

Or do you mean the sort of treaty where one country agrees to
retaliate if another country is attacked? That brings up some
interesting questions. Would such a treaty REALLY be upheld
under all circumstances? Even if that retaliation would
be doomed to fail, because the U.S. has a sophisticated
anti-ballistic/anti-cruise missile system?

When it comes right down to it, what would be preferable:
unconditional surrender involving the acceptance of a
police state puppet government, or total annihilation?
In my story idea, some countries DO prefer the latter.
But they're the first to go, for obvious reasons...

Isaac Kuo

Jordan S. Bassior

unread,
Aug 18, 2001, 6:25:51 PM8/18/01
to
Isaac Kuo said:

>What if instead he goes to local Radio Shack and buys "Mr.
>Fusion"--which is easily modified using instructions on
>Internet II into a city killing device? You, your family,
>your friends, are all killed. If not this year, next year or
>the one after that. Is this an acceptable risk? Clearly
>not. Even if it's a miserable life living 24/7 under the
>microscope in a police state, it's preferable to an expected
>life span under a decade.

The solution is not to live in cites. IMO, nuclear terrorism is one of the
reasons why cities, as we currently know them, are going to be obsolete within
100-200, let alone 500, years.

Terje A. Bergesen

unread,
Aug 18, 2001, 6:52:53 PM8/18/01
to
Isaac Kuo wrote:

...


> Or do you mean the sort of treaty where one country agrees to
> retaliate if another country is attacked? That brings up some
> interesting questions. Would such a treaty REALLY be upheld
> under all circumstances? Even if that retaliation would
> be doomed to fail, because the U.S. has a sophisticated
> anti-ballistic/anti-cruise missile system?

Sorry for commenting somewhat off-topic here... I read this
news group, but typically do not contribute... Anyway...

How do such treaties work? Will they be upheld, and at what
cost? Case in point, NATO 1980s.

Small country in pact, situated strategically well, but with
a small population. NATO is a large number of countries who
originally were primarily working on a defence against the
then large USSR.

Assumption: USSR attack on small strategic country will always
be part of a large scale attack. The small country is in NATO,
and attacking it alone has no purpose.

USSR attacks Europe. Small strategic country included. Since
the country is difficult to defend, and the USSR can under no
circumstance be allowed a strong-hold there, something must
be done, but what?

Solution: Nuke the small, but strategically important ally?

Far fetched? Nope, was rehearsed a number of times in the 80s.

Completely irrelevant perhaps...


--
+------------------------+-------------------------------------------+
| Terje A. Bergesen | ter...@yahoo.com |
| Santa Monica, CA | http://terje.bivrost.com/ |
| Living it, loving it | http://ingrid.bivrost.com/ |

John Schilling

unread,
Aug 18, 2001, 7:00:15 PM8/18/01
to
mec...@yahoo.com (Isaac Kuo) writes:

>"Brandon J. Van Every" <vane...@3DProgrammer.com> wrote in message news:<47hf7.18159$ZM2.1...@newsread2.prod.itd.earthlink.net>...
>>"Isaac Kuo" <mec...@yahoo.com> wrote in message
>>news:acc26c07.01081...@posting.google.com...

>>>Where does a science fiction story fit into this? Well, you
>>>start off with a multi-national world with one particularly
>>>powerful superpower. After a few terrorist acts, the citizens
>>>of this superpower (including their leadership) realizes that
>>>the only way for the police state to eliminate nuclear terrorism
>>>is for ALL of humanity to be under its regime. Voila! The
>>>motivation for a global takeover--one with an argueably logical
>>>and justifiable purpose. Other countries will naturally resist,
>>>of course (freedom-loving and otherwise). In this war, both
>>>sides are justified.

>>Questions: is the nuclear terrorist aligned with a nation? i.e. does the
>>nuke have a clear return address? If it does, that country ceases to exist.

>Doesn't work if the nuclear terrorist lives too close to home.

>Who are the British going to nuke in retaliation to an IRA
>nuclear bombing?

Nobody; they'll just put all the Catholics in Northern Ireland in
concentration camps. Which will be renamed something nice and fuzzy,
for all that's worth.


>Who are the Israelis going to nuke in retaliation to a PLO nuclear

>Bombing?

See above, except now it's the Palestinians.


>Who will we nuke in retaliation to the next Tim McVeigh?

The idea of a Timothy McVeigh with a nuclear weapon is ludicrous.


>It's the problem of the Tim McVeighs which I'm most concerned
>about. In a free society, there's still going to be a small
>fraction of wackos who decide to just kill as many people as
>they can even if their cause is not well thought out. All
>that's really required is enough anger and/or desperation.

And a nuclear weapon. Nuclear weapons created with any foreseeable
technology require industrial-scale alchemy, either isotope enrichment
or actual transmutation, and are going to cost a million dollars or
more. No, you *can't* switch a few wires on your Mr. Fusion home
energy center and get a city-busting KaBoom, not for *any* of the
imagined controlled fusion power schemes.

And desperate, angry nutcase whackos do not command millions of dollars.
Not only does being a dysfunctional whacko rather argue against acquiring
(or even just holding on to) megabucks, but the options available to
anyone who *does* command such resources tends to defuse anger and
especially desperation.


Individual nuclear terrorism is not in the cards this side of the
Singularity. Nuclear terrorism by PLO/IRA level groups is perhaps
barely possible even now, but the agendas of such are not actually
served by nuclear terrorism, and besides they have too much to lose.
It isn't actually impossible, but it doesn't seem terribly likely.


--
*John Schilling * "Anything worth doing, *
*Member:AIAA,NRA,ACLU,SAS,LP * is worth doing for money" *
*Chief Scientist & General Partner * -13th Rule of Acquisition *
*White Elephant Research, LLC * "There is no substitute *
*schi...@spock.usc.edu * for success" *
*661-951-9107 or 661-275-6795 * -58th Rule of Acquisition *

Don Middendorf

unread,
Aug 18, 2001, 7:02:08 PM8/18/01
to
"Jordan S. Bassior" wrote:
>
> Isaac Kuo said:
>
> >What if instead he goes to local Radio Shack and buys "Mr.
> >Fusion"--which is easily modified using instructions on
> >Internet II into a city killing device? You, your family,
> >your friends, are all killed. If not this year, next year or
> >the one after that. Is this an acceptable risk? Clearly
> >not. Even if it's a miserable life living 24/7 under the
> >microscope in a police state, it's preferable to an expected
> >life span under a decade.
>
> The solution is not to live in cites. IMO, nuclear terrorism is one of the
> reasons why cities, as we currently know them, are going to be obsolete within
> 100-200, let alone 500, years.

Well I think we might see a decentralization of populations in the
future. On the other hand I think that people in the future, if there is
an increase in population density, and a continuing increases in
technology, the citizens of that future are more than likely going to
have live within some contraints on their access to that technology,
just as they're trying to build copy protection into our hard drives
already, to protect the financial interests of the motion picture
industry, what I'm fumbling around trying to get at is that I just don't
think Mr. Fusion is going to be available, in fact privacy as we know it
may not exist in many forms. Before you gasp and groan and wail about
the tyranny of the future though. Remember that privacy is mostly an
artifact of mobile anonymous societies of the 19th and 20th centuries,
in the past when you lived in a village, or small town everyone knew
you, and you had very little to no privacy.

Don Middendorf

Brandon J. Van Every

unread,
Aug 18, 2001, 10:11:09 PM8/18/01
to

"Jordan S. Bassior" <jsba...@aol.com> wrote in message
news:20010818123341...@mb-mj.aol.com...

> Branon J. Van Every said:
>
> >And we assume that counter-sensor technology advances right along with
it.
>
> Where the distance to be scanned does not change, it historically hasn't.

Without referants I'm afraid I don't understand your point.

> And
> even if the distance didn't change, _right now_ you could not smuggle an
atomic
> bomb past (for instance) an airport security check.

Sure you could. The thoroughness of airport security in our contemporary
world is woeful. You'd just pick the security checkpoint at the airport
that's being slack. Athens International Airport was famous for this for
many years, don't know if anything has changed. The real limiting factor on
atomic terrorism today is getting the materials and government agencies
pursuing the terrorists in the field, not the airport checkpoints.

> >You're back to needing an exceedingly controlled fortress everywhere.
>
> No, you just have to check people and goods coming onto and off your hab
or
> world. That's not really that hard to do, assuming you invest in such a
system.

Well, how exactly do you check ships coming to a world that's being
colonized? Insist on laying down a fortress of space mines before anyone
sets foot on the place? I'm assuming that 500 years from now, expense and
hassle is still an issue to some people. Far more likely that people just
get started, then you have this grey period where people could get nuked and
don't have any protection. Then, who foots the bill for the planetary
defenses? What do people pay for safety?

Hmm, that tradeoff could be part of my game, actually....

This is very similar to issues at the beginning of the cold war. Remember,
there was a Preventive War movement in the USA.

> Most travelers, today, are commuters.

People will have a lot more time for leisure travel, with all the robots
doing the mundane chores.

> And cyber-tourism to some degree can replace physical tourism, too.

People like the old-fashioned stuff, it's what keeps them human.

> >Long as you've got enough ammo to do all the jobs....
>
> ... enough ammo _surviving interception_. That's an important caveat.

Well I guess I'll have to think about systemic interception technologies in
the face of surprise. On the one hand, surprise doesn't last forever. On
the other hand, carefully orchestrated surprises can stack the deck in a
grim way.

> >Dude, you attack everyone, with enough ammo to blow everyone up 100 times
> over.
>
> Dude, you defend everyone, with enough ammo to blow up all the attackers
100
> times over.
>
> Cuts both ways.

Sure, the result is MAD. But the problem with MAD is if one side is
perfectly happy to die. Being pleased with death seems to be a matter for
the Thought Police.

> >> Where are these terrorists basing from? Why can't the police there
handle
> them?
> >
> >'Cuz they're sneaky.
>
> Oh, well, I bet _real_ terrorist groups would become much more powerful if
they
> learned your easy solution. Just be "sneaky."

Right now, Palestinian terrorists are walking into cafes in Israel and
blowing themselves up. Despite the fact that the resources of Israel are on
full alert. I don't think the damage is considered acceptably small, I
think the terrorists are getting away with it.

Now sure, a sufficiently big bomb or a sufficient number of bombs would
cause Israel to invade Palestine. But interestingly, that in and of itself
is not enough to stop the bombings, in fact it would dramatically increase
them. Then a sufficient number of bombs would cause Israel to
systematically exterminate anyone and anything resembling a terrorist
organization. That will either dramatically decrease or dramatically
increase the bombings again. Finally, a sufficient number of bombs will
cause the Israelis to indiscriminately kill every Palestinian man, woman,
and child. Now, past even *that* threshold, a sufficient number of bombs
would wipe out most of the Israelis. That's the MAD scenario.

Don't nukes take us straight to the MAD scenario? The Israelis don't have
security, they have retalliation.

[re the hovering cruise missile]

> 2) Why aren't the point defense systems blowing it to crap while it's
waiting?
> Heck, if it's "hovering", why aren't _policemen with pistols_ blowing it
to
> crap while it's waiting?

'Cuz it's sneaky. They haven't detected it. It's part of a surprise
strike. The goal of the surprise attacker is to get as many bombs into this
good position as possible before being detected.

> >How about bore-bombs?
>
> "Bore-bombs," in your reality, also have the magic power to teleport
across
> interplanetary distances?

It's a colony in its formative years. There's no fortress around the
planet, getting to orbit is easy. There are only so many smuggler ships
that the UP can check on. Corruption is also a dominant problem. A UP
robot ship could be made incorruptible, but to do so, it has to be made into
a relatively simple algorithm that doesn't exercise much judgement. If you
can think of a simple algorithm that blockades all forms of human
sneakiness, let me know.

Basically, the planetary border is like the War On Drugs. The druglords
have infinite resources, it's ridiculously permeable.

> >Yeah right. Very carefully monitor everything. Dude, it's not possible
to
> >check everything. Not unless you only allow 1 ship in at a time along an
> >immense solar system corridor. The whole solar system would have to be
> >designed like a prison.
>
> ??? - Why can't you have multiple teams doing the checking?

Why can't you have multiple teams doing the smuggling? If you believe the
movie "Traffic," the USA has never even been close to controlling the War On
Drugs.

Brandon J. Van Every

unread,
Aug 18, 2001, 10:25:44 PM8/18/01
to

"Isaac Kuo" <mec...@yahoo.com> wrote in message
news:acc26c07.01081...@posting.google.com...
>
> It's the problem of the Tim McVeighs which I'm most concerned
> about. In a free society, there's still going to be a small
> fraction of wackos who decide to just kill as many people as
> they can even if their cause is not well thought out. All
> that's really required is enough anger and/or desperation.
> That'll cloud any judgement over the true repercussions of
> one's actions.

I suppose I could make a play mechanic out of this. How deeply are you
willing to curtail civil liberties for safety? In a future of nuclear power
tools, democracy is a gamble! I guess misery would have to be another play
factor, because systemic misery is ultimately what causes the Tim McVeighs
of the world.

> Or do you mean the sort of treaty where one country agrees to
> retaliate if another country is attacked? That brings up some
> interesting questions. Would such a treaty REALLY be upheld
> under all circumstances? Even if that retaliation would
> be doomed to fail, because the U.S. has a sophisticated
> anti-ballistic/anti-cruise missile system?

If the USA had such a sophisticated system, then other nations would have
developed Preventive War movements long before that system was in place. I
really think the only option is to share such technology with other nuclear
incumbents. The other possibility is the USA has a defense system, but it's
a white elephant and doesn't really work.

> When it comes right down to it, what would be preferable:
> unconditional surrender involving the acceptance of a
> police state puppet government, or total annihilation?
> In my story idea, some countries DO prefer the latter.
> But they're the first to go, for obvious reasons...

There's more at stake than the fate of an individual country. There's the
fate of humanity to consider. Will the totalitarian government be a
permanent historical condition, or something that wears out after 100 years?
Also, assuming some missiles are launched, will the fallout and dust wipe
everyone out anyways, regardless of whether they make a choice? That said,
I believe Nuclear Winter is oversold. There are critiques to it, same as
there are critiques to Global Warming. I would never implement a game with
Nuclear Winter as a knee-jerk certainty, it would have to be an unknown.

Brandon J. Van Every

unread,
Aug 18, 2001, 10:30:01 PM8/18/01
to

"Jordan S. Bassior" <jsba...@aol.com> wrote in message
news:20010818182551...@mb-cs.aol.com...

>
> The solution is not to live in cites. IMO, nuclear terrorism is one of the
> reasons why cities, as we currently know them, are going to be obsolete
within
> 100-200, let alone 500, years.

Um, human beings are both social and greedy. Where are you going to get the
land rights to a 100 mile front lawn? Maybe out in space somewhere but not
on the surface of a relatively rare Earth-class planet. The hunter-gatherer
lifestyle was easier for people, but it was permanently supplanted by the
agrarianists. Once you start turning wildland into farmland, the wildland
doesn't exist anymore and you can't just graze on it. Similarly, galactic
civilization will irrevocably change the conditions under which people live.

Joseph Hertzlinger

unread,
Aug 18, 2001, 10:28:58 PM8/18/01
to
On Fri, 17 Aug 2001 11:11:20 GMT, Brandon J. Van Every
<vane...@3DProgrammer.com> wrote:

>'Cuz it's friggin' gloomy down there. Humanity is a creature of the
>light and open spaces.

Okay. We can have large well-lit caves.

>Why would only a million people get blown up a day? Certainly if you
>can build a Dyson sphere, you can blow up the whole thing.

Why?

>Also it should be noted that if you set off a mere H-bomb in any
>metropolitan area of the USA, you'd kill way more than 1 million
>people.

How much does a nuke cost? If we are rich enough for everyone to
afford one, won't we also be rich enough to afford mass to absorb
explosive energies?

Caveat: I just got the latest issue of _Wired_ and they have
statistics on the decline of the mass involved in the U.S. GNP per
dollar. Maybe we won't have that mass available. OTOH, if there is a
terrorist threat, people will start investing in shields.

>Imagine the catastrophe of future weapons and population densities.

Aren't maximum population densities declining?

>Actually, the more I think about it, the more I see that they won't
>give up, and the society will have to adopt elaborate covert
>operations technologies.

Covert operations will be a very big mistake. The largest potentially-
destructive energies will be in power beams and the relative motion of
space colonies. The most best terrorist tactic will be to infiltrate
power and transport operations. Secrecy will enable them to arrange
disasters without interference. We will need a society in which the
average citizen can watch the Establishment. The motto will be:
"Little brother is watching you."

>Bear in mind I said 500 years hence. I think cannibalizing our
>entire solar system in only 500 years is wildly optimistic. Or
>pessimistic, if you're an environmentalist.

500, 5000, ... They only differ by an order of magnitude.

Brandon J. Van Every

unread,
Aug 18, 2001, 10:36:09 PM8/18/01
to

"Terje A. Bergesen" <ter...@yahoo.com> wrote in message
news:3B7EF1F1...@yahoo.com...

>
> USSR attacks Europe. Small strategic country included. Since
> the country is difficult to defend, and the USSR can under no
> circumstance be allowed a strong-hold there, something must
> be done, but what?

Nuke Russia.

> Solution: Nuke the small, but strategically important ally?

Not necessary if Russia glows bright enough. Supply line has been cut.

Of course, Russia will nuke back. It's the end of the world as we know it.
That's why Russia doesn't invade Europe in the first place. To worry about
the small country is to miss the forest for the sake of the trees.

More interesting would be if Russia invades *only* the small country. Do
NATO and Russia agree to have a gentleman's conventional war on the soil of
that 1 small country? Possible. Not really any different than the proxy
wars that were actually fought. It's just the consumer packaging that's
difficult.

Joseph Hertzlinger

unread,
Aug 18, 2001, 10:43:22 PM8/18/01
to
On Sat, 18 Aug 2001 01:36:23 GMT, Brandon J. Van Every
<vane...@3DProgrammer.com> wrote:

>I guess that's an interesting physics question. What's the smallest
>density of matter that could destroy the largest expanse of a
>planet's surface? How small can your Doomsday device be? What are
>the limits of undetectability?

I once calculated that it wiuld take 2 trillion tons of antimatter to
blow up the Earth.

That's hard to smuggle.

Timothy Little

unread,
Aug 18, 2001, 10:47:15 PM8/18/01
to
John Schilling <schi...@spock.usc.edu> wrote:
>>Who will we nuke in retaliation to the next Tim McVeigh?
>
>The idea of a Timothy McVeigh with a nuclear weapon is ludicrous.

How many people, 500 years ago, could individually commit as much
death and destruction as Timothy McVeigh today?


>And a nuclear weapon. Nuclear weapons created with any foreseeable
>technology require industrial-scale alchemy, either isotope enrichment
>or actual transmutation, and are going to cost a million dollars or
>more.

Isotope enrichment is not an extremely difficult process on a small
scale. Sure, I'd agree that it's *currently* beyond one wacko's
ability to carry out. But increasing technology usually has follow-on
effects in unrelated fields. In even this coming century maybe some
form of microchannel analysis device can be modified to do enrichment
of uranium compounds; even a rate of a few micrograms per second would
suffice.

Of course, not getting found out may be trickier, since I expect by
then most of the population of the world will be under constant
surveillance by someone or something able to distinguish bomb-making
activity from harmless tinkering.

For one thing, nuclear weapons require *much* rarer materials than
chemical explosives. High-energy chemicals are manufactured routinely
in huge quantities for any of millions of applications. Bulk
high-energy nuclear materials are currently only refined for one of
two purposes: power generation or weapons. There is a huge amount of
nuclear energy locked up in abundant light elements, but it is
currently infeasible to release it without rare ultraheavy elements.
I don't expect this state of affairs to last forever.


> Nuclear terrorism by PLO/IRA level groups is perhaps barely
>possible even now, but the agendas of such are not actually served by
>nuclear terrorism, and besides they have too much to lose. It isn't
>actually impossible, but it doesn't seem terribly likely.

I agree for now.


- Tim

Brandon J. Van Every

unread,
Aug 18, 2001, 10:54:29 PM8/18/01
to

"John Schilling" <schi...@spock.usc.edu> wrote in message
news:9lms1v$ebu$1...@spock.usc.edu...

>
> >Who will we nuke in retaliation to the next Tim McVeigh?
>
> The idea of a Timothy McVeigh with a nuclear weapon is ludicrous.

Today. What about 500 years from now in a spacefaring civilization? It's
just crashing a spaceship into the surface of a planet. Or stealing the
fuel that's used all the time. Or setting your geothermal drill to
overload.

> And a nuclear weapon. Nuclear weapons created with any foreseeable
> technology require industrial-scale alchemy, either isotope enrichment
> or actual transmutation, and are going to cost a million dollars or
> more. No, you *can't* switch a few wires on your Mr. Fusion home
> energy center and get a city-busting KaBoom, not for *any* of the
> imagined controlled fusion power schemes.

How do you solve the problem of the dockworker?

> And desperate, angry nutcase whackos do not command millions of dollars.
> Not only does being a dysfunctional whacko rather argue against acquiring
> (or even just holding on to) megabucks, but the options available to
> anyone who *does* command such resources tends to defuse anger and
> especially desperation.

True, you'd need a terrorist organization to systematically do the big
stuff. And such an organization has goals, like leaving people alive so
they can be converted to the One True Thought Process. Only a doomsday cult
would think otherwise.

Brandon J. Van Every

unread,
Aug 18, 2001, 11:47:54 PM8/18/01
to

"Joseph Hertzlinger" <jher...@ix.netcom.com> wrote in message
news:9ln94a$vja$1...@slb3.atl.mindspring.net...

Um, what kind of antimatter? Real world antimatter, or Star Trek
antimatter?

Brandon J. Van Every

unread,
Aug 18, 2001, 11:45:57 PM8/18/01
to

"Joseph Hertzlinger" <jher...@ix.netcom.com> wrote in message
news:9ln89a$vja$9...@slb3.atl.mindspring.net...

>
> Covert operations will be a very big mistake. The largest potentially-
> destructive energies will be in power beams and the relative motion of
> space colonies. The most best terrorist tactic will be to infiltrate
> power and transport operations. Secrecy will enable them to arrange
> disasters without interference. We will need a society in which the
> average citizen can watch the Establishment. The motto will be:
> "Little brother is watching you."

That won't work, any more than civic participation works today. Most people
don't get involved. A few activists try to get very involved, but they
don't have the authority to act, only to pester. Society doesn't give them
the authority to act because that would ruin the checks and balances of
power. Do you *really* want your cops charged with racism for *every*
shooting of a black man?

The reason I say covert is what's going to happen, is because most human
beings aren't comfortable with reality. Thus, governments have to hide
reality from the citizens.

Brandon J. Van Every

unread,
Aug 18, 2001, 11:53:47 PM8/18/01
to

"Timothy Little" <t...@freeman.little-possums.net> wrote in message
news:slrn9nua...@freeman.little-possums.net...

>
> Of course, not getting found out may be trickier, since I expect by
> then most of the population of the world will be under constant
> surveillance by someone or something able to distinguish bomb-making
> activity from harmless tinkering.

How do you draw that line? The whole controversy over bombing the Sudanese
pharmaceutical factory a few years ago is that you cannot distinguish such
civilian operations from manufacturing bioweapons. They use exactly the
same equipment. I read a book on the Soviet bioweapons program, they just
bought off-the-shelf industrial incubators. Also having read the book, I
believe it is highly plausible that the Sudanese were manufacturing
bioweapons at that facility. And even if it wasn't, there's the matter of
public trust. Whose inspectors do you believe? The average citizen
doesn't/can't know what's going on, you have to be there checking it out.
It becomes a war of sound bytes on the six o'clock news.

It's much the same problem that the Thought Police face. What kind of
thought will result in threatening action? The Thought Police don't take
chances, they just indiscriminately kill. Pretty much your options are a
permissive society or a totalitarian society, because the problem is
inherently a slippery slope.

Erik Max Francis

unread,
Aug 18, 2001, 11:52:35 PM8/18/01
to
Joseph Hertzlinger wrote:

> I once calculated that it wiuld take 2 trillion tons of antimatter to
> blow up the Earth.

It would take two trillion tonnes of mass energy, which would correspond
to one trillion tonnes of antimatter (the matter for it to react with
comes free in the form of the Earth).

--
Erik Max Francis / m...@alcyone.com / http://www.alcyone.com/max/
__ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/ \ A father is a thousand schoolmasters.
\__/ Louis Nizer
blackgirl international / http://www.blackgirl.org/
The Internet resource for black women.

Erik Max Francis

unread,
Aug 18, 2001, 11:53:00 PM8/18/01
to
"Brandon J. Van Every" wrote:

> Um, what kind of antimatter? Real world antimatter, or Star Trek
> antimatter?

What is the distinction, exactly? The figure he got (less a factor of
two) comes from a simple calculation.

--
Erik Max Francis / m...@alcyone.com / http://www.alcyone.com/max/
__ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE

/ \ They love too much that die for love.
\__/ (an English proverb)
Kepler's laws / http://www.alcyone.com/max/physics/kepler/
A proof of Kepler's laws.

Don Middendorf

unread,
Aug 18, 2001, 11:45:31 PM8/18/01
to
"Brandon J. Van Every" wrote:

> Why can't you have multiple teams doing the smuggling? If you believe the
> movie "Traffic," the USA has never even been close to controlling the War On
> Drugs.


Okay... But, what the hell do movies have to do with reality? The war on
drugs is a political construct, there is significant opposition to its'
very existence, as well as considerable political hurdles to it's
execution. That is persons are allowed to enter and leave the United
States of America with relative ease, because so far the worst they've
done is smuggle in the odd bit of illegal drugs, since so far there
hasn't been a serious influx of nuclear weapons. Were that to change and
a city or two be blasted I think you'd find that we could do a much,
much better job of actually patrolling our borders.

In the case you had mentioned I don't see where it's that big a deal at
_all_ to inspect every Starship coming into or out of a system, ditto
for interplanetary travel, already in our lax security model, I doubt if
much of anything comes into the USA on commercial aircraft, and I'm
thinking it's difficult to imagine a scenario where interstellar, or
interplanetary travel is more common than commercial air travel is
today.

Don Middendorf.

Timothy Little

unread,
Aug 19, 2001, 1:06:31 AM8/19/01