We’re back to Clement’s Game in the most direct sense, since I recently read Noise, one of Hal Clement’s novels.
This time, we’re going to echo back to the first post here. I’ll be talking about another of Hal Clement’s books. This one is called Still River. As you’ll see, the world-building here is sloppy as compared to Clement’s usual standards. But I still have a soft spot for the book, because it is a science fiction novel about planetary science graduate students.
The protagonists of the book are a bunch of students studying for the degree of “Respected Opinion” at a many-species institution of higher learning based on a number of planets around stars in the Carina Nebula. One wonders why a galaxy-spanning federation with relatively casual interstellar travel would put their university within 10 parsecs of a pending supernova, but at least the view is cool.
Although only one of the students (Molly) is human, the others are only referred to by the names given to them by Molly’s universal translator – just like “Respected Opinion” is an alias. The limitations of the translator are carefully explored, as are some of the quirks of the aliens’ psychologies. No two of them are identical, and while they are all Starfish Aliens, they aren’t incomprehensible. They all want to learn about various parts of planetary science and have written up proposals for field projects at one of the institute’s standard locations for field camps.
This small planet is called Enigma 88, and has been used for field exercises for many thousands of years (apparently, the federation has reached a state of technological near-stasis). This is where things get strange. Rather than having a standard set of things to explore while on the planet, with an expert guiding the students through and making sure they don’t miss something important, the students are not given access to the reports of the previous expeditions and are tossed down all by themselves to figure out why Enigma still has an atmosphere despite being so small and being bombarded by lots of high-energy photons from Eta Car. These aliens aren’t following anything like human training methods: the unnecessary lack of background information nearly leads to the deaths of half of the students. Doesn’t seem like the best strategy if the goal is to efficiently train planetary scientists. The thing about aliens is they’re alien?
The Pun In The Title
The students hypothesize that Enigma’s atmosphere is being sustained by outgassing from its interior, like how methane is replenished in Titan’s atmosphere. They approach and land on Engima; track the pattern of air flow and find outflows from a series of cave mouths near one pole; and start spelunking with armored environment suits, support robots powered by miniature fusion reactors, and kilometers of monofilament line. They are looking for ice or minerals holding lots of chemically-bound volatiles. In the caves they find a water-rich environment, supporting life. The life is just contamination: thousands of years of field camps dumping their waste has given a biosphere made of a mis-mash of imported microbes and assorted larger forms that were transported as spores. They find underground rivers, flowing downward to lower caves very slowly in the low gravity. As they go further down, the temperature rises and the different components of the fluid in the down-going rivers evaporate into the outgoing air streams, so what was once dominated by water is now dominated by such lovely things as hydrogen peroxide. The things are slowly moving hundred-kilometer-long distillation systems. Eventually, the students descend to a point where the evaporation rate is high enough that the river goes no further: the gas venting from still further down carries away the material that’s coming down as fast as it comes in. And the flow rate goes to zero.
And so we have an incredibly geeky pun. Given that Clement wrote “Mission of Gravity”, “Star Light”, “Close to Critical”, and “The Nitrogen Fix”, we should not be surprised.
But it wasn’t enough for Clement to have his fun distillation system be driven by outgassing from hot areas a few hundred kilometers inside Enigma. Instead, he had there be a planet-wide interior circulation system. The air flow out of the vents near one pole was balanced not by local down-welling, but by a series of vents at the opposite pole. The cave system went all the way through the planet, including a single lumpy several-hundred-kilometer-wide void right in the center. Enigma was supposedly a hollow planet. We can expect nonsense like that from George Lucas, but Clement should have known better.
Very obviously, it is a lower energy-state for dense material to be at the center of a planet and lower density material on the outside. The air inside the middle of Enigma is connected to the outside by the cave system. It is being pushed inwards only by the pressure from the air along the line of the vents. That pressure is hundreds to a thousand times less than the pressure from the rocks that are pushing the air in the central void outwards, since it is lower-energy for the rocks to occupy that space. The only way to maintain the central void would be if the rock shell itself could resist the pressure from its own weight. But as I mentioned in another earlier post, rocks aren’t that strong. The pressure on the rocks on the inside of Enigma’s shell would be much lower than that in the center of the Earth, but it would still be measured in gigapascals, way above the strength of a large mass of any geologic material. Enigma can’t exist; it would immediately and violently collapse in on itself.
There’s another part of the impossibility: not only did the students not know the place was hollow, apparently their teachers and all of the previous testing teams didn’t know either. When they get back with their reports and get out of hospital, the students are rewarded with their degrees and the news that Enigma is now being promoted to active research area rather than field camp. That makes no sense. Even without landing on the place and doing seismology, it would be obvious in Enigma’s moments of inertia that it was hollow (and also in the higher-order terms in the gravity field since the voids aren’t spherically symmetric). The big void space would also show up on very long-wave electromagnetic work, measuring the planet’s interaction with the surrounding interplanetary medium. That’s how we know about the liquid layers inside Io, Europa, Ganymede, and Titan.
I can only conclude that Clement let his liking for chemistry overwhelm his knowledge of astronomy for this one.
For the second alien species in the Fermi Problems setting, I decided to go with another take on the “Where are they?” line. Fermi’s argument for a spacefaring civilization spreading across the galaxy in only a few million years depends on a long period of nearly exponential growth and on a travel speed that is a significant fraction of the speed of light. My friend Jacob Haqq-Misra has worked out some limitations on the rapid growth of civilizations, but I decided to design aliens who could not travel at more than 0.001 c. This requires two things: a species that cannot handle high acceleration, and a technology base that does not allow them to either engineer themselves or robotic emissaries to accelerate faster or to build a rocket that can continuously accelerate for a very long time.
The latter is a rather severe limitation. Accelerating at 0.03 m/s^2, a spacecraft would still reach 0.003 c after 1 year and cover ~8 lightyears in a century if the acceleration was sustained. The later is unlikely for something internally powered: fusion rockets are limited to maybe 0.1 c peak speed, given the requirement to stop at the destination. Even a Bussard ramjet is limited to 0.119 c by the exhaust velocity of the fusion rocket. But I have set a speed limit of 0.001 c or 300 km/s. Even a fission-reactor-powered ion engine could in theory reach or somewhat exceed that speed. So, I needed to design a species that cannot handle high acceleration and can travel interstellar distances with nothing fancier than a chemical rocket. Taking some liberties with biochemistry, I propose a biosphere that evolved on small asteroids in the nominal habitable zone of a binary made of two M-dwarf stars. I picked CM Draconis because there is a lot of data available on it, it doesn’t have any large planets in the habitable zone, and it isn’t too close to Earth. Once again, I gave the fictional version of it a name in addition to the catalog listing: Druk. It happens that CM Draconis forms a hierarchical triple system with a white dwarf, GJ 630.1B, several hundred AU away. Continuing the theme, call it Wyvern.
I fully confess that the following is far more improbable than the ursians are, because the reservoirs for biochemistry to develop in are far smaller if for no other reason. With that caveat, I ask for any reasons why it is impossible, rather than merely incredibly improbable (levels of likelihood comparable to large-scale quantum tunneling may be counted as impossible). The improbable makes for good science fiction; the impossible is not allowed.
Consider a water- and carbon-rich asteroid, between a few hundred meters and a few tens of kilometers across. The particular object we are discussing doesn’t exist anymore, but it was like billions of others: a rubble-pile of variously-sized grains. Fine-grained material accumulates in dust ponds on such objects, and on a carbonaceous-chondrite object those grains will be rich in a slew of various relatively complex carbon compounds. This much is well-established.
The fictional assumption that I invoked was for there to be enough chemistry going on in those dust ponds (using monolayers of water molecules on the grain surfaces?) to produce small closed membranes durable enough to keep their contents separated from the surrounding grains and protected from being shaken off a grain into the surrounding vacuum while still being permeable enough to allow new molecules in. From there, chemistry proceeds to something that is self-catalyzing and rather like a protocell – similar to the bubble model of abiogenesis.
Given this admittedly out-there assumption, I can invoke evolutionary processes to expand this strange biosphere. Cells with vacuum-tight walls, high radiation tolerance, and specialized molecule-sized valves survive better; on the surfaces of the ponds as well as beneath them and inside ejected meteoroids. Photosynthesis takes over from feeding on primordial and cosmic-ray-seeded high-energy compounds. Heterotrophs feed on the plants and on each other; larger and better-armored multicellular plants enjoy a survival advantage and larger animals evolve to feed on them. Eventually, three-dimensional forests grow out from small cores – collecting a hundred times as much light as compared to if they only coated the rock. I was inspired to this design by some speculation by Freeman Dyson and possibly by some of the pictures in Le Petit Prince. Conveniently, these forests will be incredibly hard to observe over interstellar distances: they are dark in the visible, and infrared telescopes don’t have the same resolution as optical or radio arrays. Even if they are seen, the forests will be confused for boring dead carbonaceous asteroids until and unless their orbits are known well enough to measure their masses from mutual perturbations – showing how low their densities are.
The first rock was destroyed in a collision within a few hundred million years of the first appearance of life, scattering microbes across the system in the debris. Objects within the entire habitable zone have things growing on them; going too far out from the star causes un-insulated cells to freeze. On a timescale of a hundred million years, each individual forest is destroyed. On a much much shorter timescale, the plants break off rocket-propelled seed pods and solar-sail leaves that search for other objects using chemical-imaging tracers (the sail leaf idea was used by Larry Niven in the form of the ‘sail seed’). The animals that don’t have sufficient hibernation and delta-v abilities of their own hitch rides.
And, because I said there would be one, an intelligent species of animal evolved in one of the larger forests and then spread. I called them the neari – the etymology is a bit strained, but I hope it sounds cool enough.
Anatomy and Psychology
There is gravity, even on the smallest asteroids, although it can be effectively canceled for fast-spinning objects. The neari can handle continuous loads of ~0.03 m/s^2 without having too many problems, and can do brief jerks up to several gees. But they don’t have a physiologically preferred up and down or left-and-right, since the gravitational pull is a very gradual change as they navigate through the forests. So I designed a body plan for them inspired by sea urchins, crabs, and T4-bacteriophage: an icosahedral central body with limbs (equally validly called ‘arms’ or ‘legs’) at the vertices. The facets have specialized functions, but the neari are equally capable from all directions:
Four facets contain mouths, with airlock throats leading into a central stomach (the mouths are also used to excrete waste). Four are optical eyes, based around fish-eye lenses. Four have lines of infrared-sensitive pits. Four are blank on the surface, and underneath them are the neari’s central brain lobes. And four have organs that work only in near-vacuum: nose-eyes. This is an idea from Hal Clement: in vacuum, molecules travel long distances without collisions. So a pinhole camera can make an image in molecules as easily as in light, as long as whatever it is looking at is putting out any detectable number of particles. Neari exoskeletons and plant leaves smell only faintly. A comet outgassing is detectable tens of thousands of kilometers away – far further along its tail. An simpler version of this organ is used by the rocket pods and sail leaves to home in on objects.
The neari grow by molting and shedding their exoskeleton, a process which also allows the regeneration of limbs. Molting grows less efficient with time, and eventually accumulated damage leads to death – although prompt injuries will kill a neari far more quickly. The other way neari die is by giving birth. To solve the problem of exchanging genetic material and maintaining a pressurized environment for the embryonic neari to grow in, two neari (of any two of the three different sexes) join along the blank faces and slowly fuse together. The embryonic neari grow inside the merged exoskeleton, fed by their parents’ bodies.
These biological differences led to some important psychological differences between neari and humans (or ursians). Neari never know their parents – they are raised by their aunts/uncles and cousins. The individual is generally seen as less important compared to the family than it is for us. And sex is not a recreational activity. I have not fully explored how such different psychology will be reflected in neari culture. There is one other relevant psychological trait, from the hazard of being stuck in space just out of reach of a claw-hold: neari get very anxious if they aren’t holding onto something.
Culture, Technology, and Starships
The neari culture at Druk is in many senses less fragmented than that of the ursians, even though they’re spread out over cubic astronomical units. As of 70,000 BCE, you might have been tempted to call them a bronze-age culture. But there is no such thing as technological levels – real technological developments are spurred by the environment, the available resources and knowledge, and random moments of inspiration. The neari may have not have any metal other than nickel-iron, but they formed it into knives and hooks and parabolic focusing mirrors. Reflecting telescopes and heliographs gave light-speed communication across all of Druk, at least at telegraph-equivalent bandwidth. Writing preserved information. The neari had chemical rocketry, using rocket pods, and solar sails, made by cultivating and then trimming the leaves of sail plants. Travel between the forests, while slow and mass-limited, was inexpensive, and the neari understood Newtonian mechanics, radiation pressure, and numerical integration.
That said, the forests were isolated enough from each other that they retained very different cultures. Wars were not unknown, although they were limited in scope by the very low population density. Enough cooperation between forests did happen to organize large scale projects, including starships.
Druk has two red dwarf stars very close to each other, with an orbital velocity of ~75 km/s. With a little extra help from the super-jovian gas giant (>5 AU from the stars) and Wyvern (~500 AU), a careful gravitational slingshot maneuver past both stars can give an escaping interstellar spacecraft a velocity of ~300 km/s. And the neari already have things that can be turned into starships: comets (I got this idea from Greg Benford and David Brin). Several cubic kilometers of comet ice/organics contain ~3e16 J of usable chemical energy, enough to support 100 neari and their life-support system for tens of thousands of years. Select a comet about to make a close flyby of the stars, trim its orbit with high-speed impactors so that it gets ejected at high speed, jump on board before the flyby, afterwards adjust your trajectory using the gas giant or the white dwarf, and then burrow down under the surface for the long ride.
The problem with these comet boats is that they can only be targeted onto a certain range of trajectories, only a small fraction of comets are suitable (e.g. the comet has to be big enough to survive the stellar flybys), and the accuracy of the navigation can only be so good. The comets jet out gas unpredictably at perihelion, and even if the comet boat’s trajectory was known perfectly that of the destination star isn’t. The neari can measure parallax and proper motion accurately, but radial velocity is only approximate. A 0.1 km/s error at launch would be 1000 AU at arrival, and the comet boat can’t change its velocity by too much more than that en route.
For purposes of the setting:
The neari did start to spread across interstellar space ~70000 years ago – launching comet boats every few thousand years. But only one of them got close enough to its target star for the neari then on board to ditch the comet and stop. The others are either confirmed to be dead, still flying through the void, or unknown in status because they stopped replying to heliograph calls centuries ago. So Druk and Wyvern (the latter has a small debris disc out to a few hundredths of an AU from the star that the neari have settled) are still the main place to find the neari and there they are hard to see from parsecs away.
Does all of this work for a species that can travel over interstellar distances without being readily detectable, while simultaneously not spreading so fast that we should expect them to be here already?
Since we are playing Hal Clement’s game, we’ll start by dissecting the world-building behind one of his stories.
Mission of Gravity, its sequel Star Light, and the associated short stories Under and Lecture Demonstration were arguably Clement’s greatest works, and the world-building behind them was the subject of Whirligig World. The stories are largely set on Mesklin, supposedly a 16 Jupiter-mass planet in orbit around one of the two stars of the 61 Cygni system. While there were several claimed detections of such a planet between the 1940s and 1970s, we now know that there are no objects more massive than Jupiter around either star. The initial claims were based on astrometry, measuring the positions of the 61 Cygni stars relative to each other, and the claimed detections were marginal and turned out to be slight systematic biases in the data. The current limits are from high-precision radial velocity, which provides much better sensitivity except for objects on orbits oriented at almost exactly right angles to the line-of-sight to the stars. Such orbits aren’t stable around 61 Cygni except very close to each star, because they would be out of the plane of the stars orbit around each other. This gives us our first problem with Clement’s world-building:
1. There is no such planet as Mesklin.
But since Clement wrote the original Mission of Gravity in 1952/1953, we should perhaps excuse this one. At the time, the claimed detections of a similar body were in the literature. He wrote Star Light and the shorts many years later, but ret-conning the whole setting to put it somewhere else would not have made that much sense.
In order to have a solid surface for the local intelligent species (and visiting humans) to stand on, Clement gave Mesklin a composition similar to what we now know to be the composition of Titan – dominated by ices, water and ammonia, which are rocks at a surface temperature of ~100 K (although the plot of Lecture Demonstration focuses on how water-ammonia mixtures can have a lower melting point than either of the pure chemicals). Methane was a liquid, and drove a complicated meteorological cycle. Presumably, there would be some amount of silica as well, but there were only traces of hydrogen and helium.
There are several problems with this, all from Mesklin’s shear mass. It’s so heavy that it should have accreted huge quantities of hydrogen and helium while it was forming. There is no method of planet formation that gives such a large object without lots of gas. More directly, there simply wasn’t that much non-hydrogen non-helium material in the 61 Cygni system when it formed. Between them, the two stars in the system contain only 12 Jupiter masses of such stuff, and only a small fraction of that amount would have remained outside the stars and been available for planet-building.
2. A non-gas-giant planet the size of Mesklin cannot normally exist.
This one is much less excusable – Clement knew his astronomy well enough to understand the total amount of material available to build planets, and how little of it is potentially solid material. And something with 16 times the mass of Jupiter would normally be considered a brown dwarf – it will be fusing deuterium into helium in its interior.
If we grant this impossible planet existing, then we have another problem. It will be very very hot, because of gravitational energy released when it accreted. Mesklin would be about the size of Jupiter – the size is determined primarily by electron degeneracy pressure in the interior, rather than by the chemical composition, much like for a white-dwarf star. But it will have two hundred times the gravitational binding energy of Jupiter. That is a lot of energy. When it initially formed, it will be plasma and gas all the way through to the core (although electron degeneracy makes the definition of ‘plasma’ a little strange). It will cool with time, but cooling models of brown dwarf stars say that even after ten billion years Mesklin would still have an effective temperature of at least 500 K after 10 billion years, depending on the opacity of the atmosphere, and 61 Cygni is only 6 billion years old. No liquid methane, no solid ice, no liquid water, no surface to speak of – hence ‘effective’ temperature rather than ‘surface’ temperature.
3. Mesklin cannot have cooled to the point that it would be solid.
This one again is not excusable, given the standards that Clement set himself.
A solid and cold object the size of Mesklin would have very high surface gravity – far too high for any human to endure. Clement realized this, of course, and came up with a way to decrease the gravity over a portion of the surface. He gave Mesklin a spin period of ~18 minutes, which gives a much lower gravity at the equator than at the poles – it is just below the spin rate at which material would fly off into space. Mesklin would then reconfigure into a flattened shape, the exact contours of which depend on the equation of state of the interior – how density changes with pressure. Clement confessed to an approximation here: he modeled Mesklin as an oblate spheroid, which is not correct (see his 2000 Addendum to Whirligig World), and specified a surface gravity of ~3 g at the equator. The gravity at the poles worked out to 700 g in Clement’s approximation; values for different equations of state were all >200 g.
However, such a rapid spin rate cannot be maintained, or originate in the first place. During planetary and stellar accretion, objects spin down to well below breakup speed. Angular momentum is transferred from the material that accretes onto the star or planet to material that escapes, by gravitational tides. For large objects, particularly stars, the object and the innermost part of the accretion disc are both ionized, and something called magnetorotational instability also transfers angular momentum outward, giving a relatively slow-spinning object. Clement can be excused from not understanding this, since the theory for it has only been developed over the last twenty years or so.
But Mesklin would have some elongation. This means that tidal interactions between it, the two stars in the system, and the moons that Clement specified Mesklin to have would act to rapidly de-spin it. The masses of Mesklin’s moons were not specified, but they would have very rapidly evolved outward to very distant orbits or to escape Mesklin entirely.
4. Mesklin cannot retain its rapid spin, or any satellites in close orbits.
While magnetorotational instability is a recent development, the theory of tidal damping isn’t. So Clement missed this one as well.
Edit: Mesklin’s inner moon was supposed to have a semi-major axis of 90,000 miles. But Mesklin’s Roche Limit would be ~300,000 miles. Any satellites within that distance would be shredded to pieces by tides. Clement should have caught that too.
There is one final error, which I noticed even the first time I read Mission of Gravity. The plot of the book is that humans have landed a special research rocket at the north pole, but it broke and was unable to launch again (how to build a rocket able to accelerate at 700 g was never addressed, nor was it adequately explained what science they hoped to learn). Since the book was written before computers became as ubiquitous and capable as they are now, the solution was to land humans at the equator, have them make contact with a group of the local intelligent species, and trade weather reports, satellite recon, and radio communications for retrieval of the rocket’s payload. The trader captain they make contact with, Barlennan, is 18 inches long and 2 high and looks like an armor-plated centipede, but he’s shrewd at negotiating and goes along with the plan. The adventure story of his crew’s travel across the planet, first accompanied by one of the humans and then with only radio links as they get to higher-gravity regions, is the bulk of the book and lets Clement show off his world.
But near the end of the book, they get to near the landing site and find their way blocked by a cliff one hundred meters high. That’s the problem. Under 700 gravities, the pressure on the ice underneath the cliff will be ~7 mega-Pascals for each meter of height. The yield strength of ice at 90 K is about 70 mega-Pascals (it is weaker still at higher temperatures, which is why we don’t have 7 km high ice cliffs on Earth). That means that Mesklin can’t have cliffs higher than ~10 m at its pole. The material would promptly collapse (and be turned into steam from the energy release). So those few chapters of the book don’t work; nor does much of the plot of Under , where Barlennan and several of his crew are washed into methane-eroded caves under the cliff-face. Neither the cliff nor the caves could exist.
5. There cannot be tall cliffs or caves at Mesklin’s pole.
In the 1950s, nobody had measured the yield strength of cryogenic ice, but scaling from values at warmer temperatures would say that the cliffs and caves were impossible. Clement should probably have figured this one out as well.
There is an additional complication: at a temperature of ~90 K and a depth of 20-30 m under 700 gees, ice transitions from ice Ih or Ic to ice IX or ice II. That increase in density and change in mechanical properties would do something odd to the geology. This hasn’t been investigated in detail, because on real objects with much lower gravity, the phase transitions happen at much greater depths with much greater temperatures, giving a sub-surface liquid water layer and then a transition to Ice V, VI, or VII.
So, unfortunately, there are at least five ways that Mission of Gravity is impossible, and Clement should have seen four of them. Perhaps he did, and ignored them for the sake of his story. And least you think that I’m beating up Clement in particular: when we dissect other works, you’ll see that Clement’s errors are relatively subtle in comparison to most. And, as the name implies, hard sci-fi is difficult to do correctly. Clement played the game very very well, although he didn’t play it perfectly.
Are there any problems with Mission of Gravity that I’ve missed? I haven’t touched on the problems of biology based on liquid methane as a solvent – biology able to produce plant fibers can can be woven into few-hundred-meter-long cords with twice the tensile strength of Kevlar.