Somewhere Else Entirely
by Mike Combs
Copyright © 2004
In the 1970's, Princeton physicist Gerard O'Neill began a line of research which
came to an unexpected conclusion. Alarmed by the then-prevalent talk about "limits to
growth", he started looking at the prospects for the expansion of human society beyond
the Earth. Contrary to the expectations left to us by science fiction, O'Neill concluded
that the most significant part of our future beyond the homeworld did not lay on the
surfaces of other planets. He asserted that conditions on planetary surfaces were less-than-ideal
for continued expansion of our technological civilization, and felt that artificial
habitats in free orbit would have
numerous advantages over the more conventional,
planet-bound concepts for extraterrestrial colonization.
But in the years since the stir in the space community caused by O'Neill's counter-intuitive
conclusions, we've seen the planetary paradigm largely regain its former
monopoly on our thinking regarding the future of the human race beyond Earth. At
present, the planet Mars holds sway over most space advocate’s dreams of a second home
for humanity. Is it true (as many seem to take for granted) that in the years since
publication of O’Neill’s "The High Frontier", newer information has rendered this
particular vision less practical, or the Mars colonization dream more practical? Or are
there no reasons for this shift in view other than planetary chauvinism?
"Planetary chauvinism" was a phrase coined by Isaac Asimov. He used it to refer
to our natural tendency to assume that most activities are better done on planetary (or
lunar) surfaces than in orbital space.
I'm active on several space-oriented newsgroups, forums, and maillists. I'm not
terribly qualified to throw out opinions on single-stage vs. two-stage-to-orbit, air
breathing engines vs. rockets, or NASA vs. private industry. Those discussions I follow,
but I don't have much to contribute. But I do have one specialized interest, and it's
planetary chauvinism. I tend to speak up when I think I'm detecting a whiff of it in a
particular proposal. It's made for some interesting, frustrating, and occasionally funny
On two different occasions, I've heard someone advocate a particular kind of
manufacturing facility on the moon, creating end-products for use elsewhere. Being
well-versed in the NASA studies on space manufacturing from lunar resources, I was
aware that their recommendation was to concentrate industrial activity in High Earth
Orbit (HEO) where there was continuous access to solar power. Those studies recommended
lunar surface manufacturing for only one class of products: those whose end use was on
the lunar surface.
So I would ask, "Why not place the manufacturing facility in HEO?" to be told
that this "just wasn't practical".
"Because you need a stable platform for major industrial operations."
"What does this mean, 'stable platform'?"
"Well… suppose you had this big stamping press, continuously hammering away
in the same direction day in and out. After a while it would disturb the orbit of the space
These kinds of assertions from people make me wonder how much of their
childhood was spent pulling themselves around in a wagon. I asked one of them why we
hadn't retired those old-fashioned rocket engines to the junk heap in favor of electric-powered
pile drivers hammering away at the interior noses of inter-orbital space craft.
While I never got an answer to that particular question, the arguer asserted he had seen a
TV documentary which said it was possible for astronauts to disturb the orbit of the
Space Shuttle by bumping into its interior cabin walls.
The shuttle has sometimes engaged in experiments where even the minutest
disturbances in the orientation of the orbiter can be detrimental to results (such as
Earth-scanning radars or astronomical observations). I suspected a confusion between
"orientation" and "orbit", either on the part of the debater or the documentary itself. But
to my horror, another debater (by all accounts a physics PhD) agreed that astronauts inside
a space craft could indeed change its orbit by bouncing off its walls.
That same PhD had earlier (and somewhat enigmatically) declared, "The moon is
the port of entry into Earth". When we asked what exactly he meant by that, he said he'd
have to think about it a while and get back to us (there could hardly be a clearer signal of
the ad-hoc reasoning process at work).
When his explanation came, it was this: In the future there will be asteroid
mining. Both raw materials and semi-finished goods will be sent back to Earth. (High
Frontier thinkers tend to be skeptical of the economics of bringing space resources down
to Earth, but let's let him have this point.) Before final delivery, these cargoes will need
to be stored. This will happen in vast warehouses on the moon.
I wasn't even the first on the forum to ask him, "Why spend all the money needed
to descend and then ascend the lunar gravity well when instead we could just skip the
moon and house all this material in HEO?"
We couldn't do this, he explained, because unloading this much material into an
orbital facility would disturb its orbit.
I explained that orbital rendezvous was synonymous with matching orbits, that
final closing maneuvers could be as gentle as we had patience for, and asserted that the
only way payloads entering or leaving this facility would significantly disturb its orbit
was if we used some kind of electromagnetic system to decelerate rapidly incoming cargo
or rapidly accelerate outgoing. I then suggested we simply not use our storage warehouse
as either a catcher’s mitt or cannon. But this wasn't what he was talking about. He was
talking about the fact that placing this much cargo inside the orbital facility would change
its center of gravity, and thus its orbit. And he had any number of bewildering equations
to show us he was right.
Now this wasn't so much a wrong answer as it was a case of straining at a gnat
while swallowing a camel. The center of gravity issue he raised might amount to a few
meters in an orbit ranging over many thousands of kilometers. The amount of propellant
needed to brake several thousand tons of raw materials or semi-finished goods into lunar
orbit, de-orbit, hover, soft-land, and then later reverse the entire process he did not
consider an especially-great problem. The amount for a brief, minor station-keeping burn
on an orbital facility: a show-stopper. Only the moon could provide a sufficiently stable
platform to take on several thousand tons of cargo.
I finally decided that all of his equations, arguments, and dire concerns was
simply the more acceptable alternative to saying, “An orbital storage facility… Gee, I
hadn’t thought of that.”
On sci.space.policy, one person said that Mercury would be a good place to build
solar power stations for beaming energy to all parts of the Solar System. I asked him
why build the solar power plants in a place which was dark half the time, and suggested
that even if we assume Mercury was the best source for the raw materials needed (there
may be better asteroids, even this close to the sun), mass-drivers could be used lift them.
After trying to suggest that maybe the surface of Mercury would make a good heat sink,
he later said that no, he didn't have any kind of bias, but that I had "stupid blinkers on".
To judge from appearances, I’d utterly failed to get him to see that a power station which
operated continuously was worth twice as much as one only operating half the time.
We might say that up until now we've only been discussing amateurs, and that
surely space professionals wouldn't let any kind of bias trip them up. But then we have
Dr. David Criswell who has also proposed solar power stations in a place dark half the
time (for 14 days at a time), in this case on the lunar surface. This was puzzling, because
not only was Criswell aware of Gerard O’Neill’s proposal for orbital manufacture of
Solar Power Satellites (SPS) from lunar resources, he had even helped work on it. Why
then was he now proposing solar power plants on the lunar surface? There could only be
one reason for supposing the lunar surface proposal competitive with the orbital one: The
expenses of lifting lunar resources into space were so overwhelming that avoiding it
made even solar power stations less than half as effective more economical.
Criswell took note of some critical comments I’d made in my
Space Settlement FAQ,
and kindly sent me many papers he’d written on his Lunar Power Station (LPS)
proposal. And I did make some modifications to
what I had written. I’d first like to say
that if Dr. Criswell has calculated LPS could turn a profit, I’ve every reason to expect
he's right. But I turned through the papers looking for the comparison with the O'Neill
proposal which might have steered him in this new direction. All the papers had
comparisons with ground-based solar and with Earth-launched SPS. None had detailed
numerical comparisons with lunar-derived orbital SPS. The earliest one or two papers
mentioned this option briefly only to dismiss it as impractical with little or no detailed
discussion. Past that point, the High Frontier scenario was never again mentioned.
(But I may now need to retract at least this point.)
It's true the O'Neill plan very much turns on
the practicality and economics of
mass-driver launch from the lunar surface. But if Criswell had discovered any fatal flaw
in the mass-driver concept, he kept it to himself.
There are professional proposals for radio telescopes on the lunar farside. An
aluminum noise shield less than a millimeter thick can do the same job as is proposed for
the entire moon, leaving an orbital dish freely pointable in any direction, as well as easier
to access from Earth. I’ve heard it argued that the stability of the lunar surface is
indispensable where interferometric observations are concerned. But note that NASA is
moving forward on an orbital mission which will indeed do optical interferometry in
space. It could be that if one’s primary concern is for staying within a fixed budget, as
opposed to justifying lunar development desired for its own sake, more economical
alternatives present themselves.
Planetary chauvinism may also influence most thinking regarding permanent
human settlements beyond the Earth. It seems to be a near-universally-held opinion that
setting up such communities on Mars will obviously be less expensive than in HEO. It's
not considered close enough to even be worthy of question. Why? Because Martian
communities will import only a bare minimum from elsewhere, whereas a habitat in free
orbit will obviously import everything. This fact is considered to trump any additional
expenses involved with operating at a much greater distance from the Earth.
The problem is that there simply aren't any studies of 10,000-person permanent
habitats for Mars, at least not in anywhere near the level of numerical detail of the NASA
space settlement studies. So in truth, direct comparisons are impossible to make. We
would need an "Island One for Mars" study before both locations could be compared on
the basis of similar aims and returns. The existing attitudes on the relative difficulties or
expenses involved are based on what seems obvious to everyone, rather than on peer-reviewed studies.
Unable to assail the logic behind any of the half-dozen or so
points O’Neill made
regarding the advantages of an orbital location over a planetary surface, most Mars
advocates simply argue that these points are moot since orbital habitats are impractically
difficult to set up in the first place. Is there anything to the widely-accepted truism that
Mars is “easier” than orbital space? To answer this question, we must ask what Mars
provides naturally which must be provided artificially in orbit.
One thing is gravity (though it must be pointed out that Mars provides the wrong
level of gravity for maintaining what we presently consider normal muscle tone). Orbital
habitats can provide any desired level of “gravity” by rotation. Is there any reason to
expect this to be any significant initial or ongoing expense?
O’Neill calculated that the electric motor needed to spin up even his enormous
Island Three habitat need only be about the size of a car engine. The vast cylinders could
be spun up on a timescale of months using only the electricity generated by the sunlight
falling on the end caps. One only has to overcome inertia. Ongoing power requirements
are apt to be far smaller, given reasonably-well-designed bearings. Certainly there are
reasons to expect most any space habitat to have a non-rotating section, but in a pinch
this section could be dispensed with, and the entire habitat could rotate without friction as
a monolithic unit. Incoming spacecraft could dock in the same manner we saw in the
The only other thing Mars gives us for free is a natural day/night cycle. In a space
habitat, this would be accomplished by shifting large mirrors made of aluminized Mylar.
Given the flimsiness and simplicity of the mirrors, there’s no reason to expect any major
expense here, either.
The more significant expenses involved with setting up human communities
beyond the Earth will be: creating new industrial infrastructure totally from scratch,
engineering large, pressurized volumes, and setting up and maintaining closed, balanced
ecologies. Obviously these requirements will be much the same whether discussing Mars
or orbital space.
It’s difficult to avoid thinking that surely there’s some savings in not having to
build the very ground beneath one’s feet. It surprises many to find that in an O’Neill
habitat, the loading on the structure from the centrifugal force acting on the interior soil
and furnishing is significantly less than the loading from the internal air pressure. Once
you’ve engineered the required pressure vessel hull (which we must do in either case),
you don’t have that much further to go to engineer something safe to landscape and build
But discussions of which is easier may not be as relevant as which is most likely
to come about in the first place. O’Neill felt that his orbital habitats would forever
remain dreams unless he could propose some economic opportunity which they alone
could exploit; one which would pay for the initial investment with profit and would
balance the trade with Earth going forward. This would have to involve some service
marketable to Earth (the only customer currently in existence), which nonetheless could
be better performed in orbit than on the Earth’s surface. He proposed the construction of
SPS, built from space materials in HEO and then delivered to GEO, as this service.
It should be mentioned that past a point, O'Neill freely conceded that construction
of SPS from space resources would precede large, Earthlike space habitats,
and not the other way around. And in a similar vein, the use of the phrase
"High Frontier" throughout this article should be interpreted as "significant
manufacturing capability in HEO fed by space resources" rather than as "Island
Three". Still, it remains that once we've set up the infrastructure needed
to manufacture SPS from space materials, pretty much everything we need is
in place to later build pleasant orbital habitats. Surely by the
time SPS construction has ramped up to the point that orbital workers
number in the thousands, something on the scale of Island One will have been
built, if only to reduce worker turnover.
Mars advocates have said orbital habitats will never happen because SPS is a
flawed business plan, but then propose habitats on Mars seemingly in the absence of any
business plan at all. They also tend to minimize the importance of exports for the
balancing of trade, perhaps only because Mars would seem to have little to sell Earth. Or
alternately, may posit intellectual property as the export for the red planet. But it seems
that if Mars settlements can balance their trade with sale of intellectual property to Earth,
then space settlements should as well, even if SPS is a bust.
Bearing in mind that in the long run some humans may need no reason for living
on Mars other than a love of the planet, High Frontier advocates see much that their
scenario can offer to the enterprise of settling Mars. But Mars advocates tend to be
dismissive of suggestions there are any contributions which the space islands could
possibly make to their goals. This puzzled me, until I realized that there were two places
the mind of the Mars enthusiast enjoys going to.
The first is to a near-term future, one in which the US government has gotten back
into an Apollo frame of mind, and is willing to send scientific expeditions of four persons
apiece to the planet Mars. Orbital habitats are certainly not needed for missions of this
scope. Indeed, Robert Zubrin has successfully demonstrated that not even so much as the
presently-existing LEO space station is needed for sensible execution of this plan.
The other place is the much more distant future of an independent Martian
civilization. High Frontier has nothing to offer here either, since the Martian civilization
is defined as independent.
Obviously much must happen between the one vision and the other. But the mind
of the Mars advocate tends to skip lightly over this place. Narratives typically leap from
small bases built up from the latter Mars Direct missions to Bradbury City with scarcely a
paragraph or two on the vast intervening eras.
But these are the eras in which major manufacturing capability in HEO supplied
with space materials has the most to offer Mars colonization efforts. Consider the first
10,000 tons of mining equipment, the first 10,000 tons of ore refineries, the first 10,000
tons of manufacturing facilities of every kind, and the spacecraft for transporting the first
10,000 colonists (along with all the above). Consider that before this can be sent to the
red planet, it all must first be built by people who were born on this planet. But even if
built by Earth people, these seeds of the new Martian civilization need not necessarily be
assembled on the surface of the Earth. If all this can be produced in HEO from materials
already in space, then this is a tremendous leg up on any attempts to colonize Mars. But
since this era is not a place the mind of the Mars advocate tends to go to, they tend not to
see or acknowledge this.
Gerard O’Neill’s unexpected results prompted him to make the following remarks
in “The High Frontier”:
We should ask, critically and with appeal to the numbers, whether the best
site for a growing advancing industrial society is Earth, the Moon, Mars,
some other planet, or somewhere else entirely. Surprisingly, the answer
will be inescapable - the best site is "somewhere else entirely."
But O’Neill’s surprising answer seems largely forgotten in the 21st Century space
advocacy community. Is there a reason other than that it is surprising; than that it’s
ill-fitting to imaginations conditioned more by science fiction than by any real-life attempts
to live permanently beyond the Earth?
Sometimes when space thinkers propose the building of a manufacturing plant, a
solar power station, a radio telescope, or a permanent human settlement, I suspect they
frequently proceed from an Earth-bound assumption. As we all know, the first step in
building anything is finding a plot of ground upon which to build it. In my debates with
many, I’ve seen little to convince me that they began with no preconceptions, gave due
consideration to orbital locations as well as planetary ones, and only ruled out orbit after
seeing major advantages to a planetary site. When called on this, the reasoning offered
up is after-the-fact, as they struggle to justify a previously unexamined starting
assumption they now feel committed to.
The problem with unspoken assumptions is that they remain unquestioned
assumptions. We must train ourselves to think outside the planetary box when considering the best
locations for all of our future activities beyond our current terrestrial home.
Return to Articles