Goldilocks Earth

Not so long ago, many people, including scientists, had what they thought was a very good reason to believe that the Earth is unique, the only planet that can support life. Their reasoning went much further: they ‘knew’ that the Earth was the only planet in the Solar System that could support life, because the Solar System was the only place in the universe that had planets. In support of this view was a clear statement of fact: planets round other stars had never been observed, so their very existence was hypothetical. Hypotheticals have no place in solid science: the only reason to think that other planets might exist was pure speculation, based on our limited knowledge about the formation of the Solar System.
To be sure, this suggested the exact opposite: that there is nothing very special about the Sun, so similar processes have probably occurred elsewhere. That meant planets. It was plausible, but there was no proof, so it wasn’t science.
This particular line of thinking has gone the way of the dodo. As I write, we know of 518 planets circling other stars10: the technical term is ‘exoplanets’. More are being discovered every week. It is becoming obvious that a significant proportion of the stars in the universe have planets; possibly most of them. We may never be able to observe the bulk of these worlds directly, but a random sample usually represents a wider truth. Planets are no longer the issue. The only reason we didn’t observe them earlier was that we lacked the technology to detect them. So the frontier of the debate has retreated to the existence of Earth-like planets. Almost all known exoplanets are huge, bigger than anything in the Solar System, dwarfing even Jupiter. The diehards retreated to a previously prepared position, now insisting that the evidence merely proves the existence of gigantic planets that could scarcely be more different from Earth – which to them means no possibility of life on such worlds. Again, there is actually a good reason why most of the known exoplanets do not resemble Earth: the methods used to detect exoplanets work best when the planet is very large.
Improved observation techniques have pushed that frontier back too: we now know of much smaller exoplanets, and can already detect the main gases in their atmospheres. In 2008 Mark Swain’s team at the Jet Propulsion Laboratory in California made the first detection of an organic molecule, methane, on an exoplanet.11 They found it on HD 189733b, a ‘hot Jupiter’ about 63 light years from Earth. Water vapour has been found on GJ 1214b,12 and the same methods should be able to find oxygen.
This suggests that it is unwise to dismiss reasonable possibilities on the ground that there are no observations to support them. You need independent evidence that those possibilities are intrinsically unlikely. A lack of observations can change at any moment with the invention of a new technique. So the current absence of evidence of alien life forms that differ from those on Earth may simply be due to the absence of evidence of alien life forms. Just as the absence of evidence of Earth-like planets was, until recently, due to the absence of evidence of planets – which was not evidence for the absence of planets, Earth-like or not.

The case that complex life elsewhere in the Galaxy, or the universe, is very uncommon has been made, eloquently and comprehensively, in Rare Earth. Ward and Brownlee list a large number of special features of our planet, all supposedly necessary for life to exist, and then work out how likely such a combination of features is. Their result is: very unlikely indeed. They don’t rule out simple life forms, such as bacteria, but they argue persuasively that anything even as complex as, say, a goldfish, must be a very rare thing in our universe. They don’t claim that Earth is unique in having such creatures, but other Earths, if they exist at all, will be very thinly spread.
I’ll list some examples of these features – they are based on solid science, much of it surprising and recent, and of interest in its own right. I’ll restrict attention to three astronomical ones. The first two are relatively new; the third is much older.
1. Jupiter protects the inner planets, Earth among them, from being bombarded by comets. A dramatic instance of this process was the break-up of comet Shoemaker – Levy 9 in 1994. Earlier, the comet had swung close to Jupiter, and was diverted in its orbit so that it would return after a few years. As it approached the giant planet, it broke into twenty pieces, which slammed into Jupiter releasing the energy of six million megatons of TNT – roughly six hundred times the world’s total store of nuclear weapons. If any one of those fragments had hit the Earth, nothing higher than bacteria would have survived, and probably not even them. Without Jupiter, comets would be hitting the Earth every twenty years or so.
2. The Earth’s Moon keeps our planet’s axis of rotation stable. Mathematical calculations show that a world lacking a moon that is large compared with itself will suffer erratic changes in the direction of its axis over periods of tens of millions of years.13 Such large moons are uncommon; it is thought that ours originated from a massive collision between Earth and a body the size of Mars during the early stages of the formation of the Solar System. Such collisions are rare.
3. The Earth is situated within the Sun’s habitable zone: a hollow shell of space inside which liquid water can exist on a planet’s surface. Get too close to the Sun, and water will turn to steam and may boil away entirely; too distant, and it will freeze. The habitable zone is limited: Mercury and Venus, close to the Sun, are on the inside of it; Mars, Jupiter, Saturn, Uranus and Neptune are outside it. We got lucky.
Rare Earth lists several dozen such features, and they are regularly trotted out in television science programmes as proof that Earth is very close to unique. However, the case for Earth’s rarity, like premature news of Mark Twain’s death, has been greatly exaggerated.
Indeed, the importance of each item in the list has been exaggerated. Worse, the significance of any such list has been exaggerated. I’ll go through the three items above in turn, and then turn to my more general objection.
Jupiter. As Shoemaker – Levy 9 shows, there are occasions when Jupiter does indeed protect the Earth from comets. But that does not imply that it always has a beneficial effect. It can also divert incoming comets, causing one that might have missed the Earth to hit.
World governments and NASA are starting to worry about NEOs: near-Earth objects. These are lumps of cosmic rock whose orbit round the Sun can bring them close to Earth. Many are asteroids, bodies ranging in size from a tennis ball to one-third of the diameter of the Moon, although the largest bodies in near-Earth orbits today are much smaller than that. Asteroids are found by their thousands in the asteroid belt between Mars and Jupiter.
Indeed, most Earth-crossing asteroids probably originated in the asteroid belt. If they had stayed there, they would pose no danger to our world. What made their orbits change, to become Earth-crossing?
Jupiter.
Mathematical calculations show that Jupiter has a big effect on asteroid orbits. In fact, as the most massive planet in the Solar System, Jupiter has a big effect on the orbits of small bodies. One of the things that Jupiter can do is disturb suitable asteroids, causing their orbits to elongate until they cross the orbit of Mars. This makes them Mars-crossing, not Earth-crossing. But now, if they come close to Mars, they can be diverted again, and now their orbits can cross the orbit of our own Blue World. So Jupiter centres the ball, and Mars scores.
Jupiter has two faces. One is that of protector. The other hurls rocks at us.
Ward and Brownlee discuss this, and argue that the occasional asteroid impact may be good for evolution, shaking up the biosphere. And so it might, but I wonder why asteroids are beneficial in this respect, while comets are not. It seems like special pleading. In fact, the presence of Jupiter may do more harm than good.
The Moon. Unlike most astronomers, I’m not convinced that the current theory of the origin of the Moon is correct,14 but whatever the mechanism was, it does seem likely that satellites whose size is comparable to that of their primary planet are quite rare. So I’ll concede that. And I agree that the presence of such a body does stabilise the axial tilt. However, it is not at all clear that if a planet’s axis changes its direction over a period of tens of millions of years, which is what the mathematics says, then this poses an insuperable problem for evolution. Earth’s creatures have coped with ice ages that come and go every ten or twenty thousand years, which is far more rapid than a change in axial tilt. Land creatures can move as the climate shifts – we’re talking a few hundred metres per year, and they’re moving faster than that today in response to climate change – unless they run out of land, which can happen. (Our elderly cat moves faster than that when chasing a mouse, but I’m referring to changes in average geographical location.) Birds can fly across open sea. And ocean creatures wouldn’t notice any difference. Since it is generally agreed that life began in the Earth’s oceans, and got very complex there, then the tilt of the Earth’s axis doesn’t matter a hoot.
Habitable zone. The habitable zone of a star is often referred to as the Goldilocks zone because it’s ‘just right’. The problem with habitable zone ideas is not that the concept is completely silly, but that it is too simplistic. It is, for example, not at all clear whether the Earth is within the Sun’s habitable zone. An airless Earth might have boiling hot surface temperatures, like the Moon when the Sun is overhead; on the other hand, one with too little carbon dioxide or a white reflecting surface might be covered in ice, as some of the planet is now, and most of it was during the period of Snowball Earth about 700 million years ago. Both Mars and Venus might support liquid water in suitable circumstances, but in those that currently prevail, Mars can get down to 15 degrees below zero and Venus is hot enough to melt lead. The highest surface temperature recorded on Mars to date is 27°C, but only on rare summer days.
It’s worth taking a closer look at the mathematics of habitable zones, to see where the difficulties arise. The calculations start from a central idea in the physics of heat, called a black body. A green object looks green, to the human visual system, because the object reflects sunlight in a range of wavelengths that our brains interpret as ‘green’. A black object does not reflect any wavelengths in the visible range; black is the brain’s default for such objects. A physicist’s black body is an idealised and extreme version of this: it reflects no electromagnetic radiation whatsoever.
However, reflection is not the only way for an object to emit radiation. A black body at a temperature of zero degrees kelvin – ‘absolute zero’, the lowest possible temperature – would emit no radiation of any kind. But at any other temperature, a black body does emit radiation; it just doesn’t do this by reflection. Instead, it glows incandescently, like a red-hot iron bar. The intensity of radiation emitted depends on the temperature and the wavelength of the radiation. Classical physics predicts that a black body should emit an infinite amount of energy, but that makes no sense. In 1901 Max Planck derived a new formula that agreed with observations, and this was later interpreted as evidence for a quantum world.
Planck’s law can be used to derive a formula for the temperature of a planet orbiting a star, and that lets us calculate where the inner and outer edges of the habitable zone are. There are two versions of the formula. The simplest one models the planet as a black body. However, a real planet will reflect some of the radiation that hits it, and the second model takes this into account by incorporating an extra quantity into the formula. It is called the albedo of the planet, which is the fraction of incoming radiation that is reflected away.
Only the first version can yield a habitable zone that depends only on features of the star. As soon as albedo comes into play, the habitable zone also depends on features of the planet – real or hypothetical – that is under consideration. The second version is more general than the first: if we set the albedo to zero, the value for a black body, we recover the first model. The formula relates the planet’s temperature to the star’s size, its surface temperature, the distance from the star to the planet, and the planet’s albedo.15
First, let’s calculate what the Earth’s temperature would be if its albedo were zero, the value for a black body. The answer is 279 K, or 6°C, placing us just inside the habitable zone. However, if we use the observed albedo, which is 0.3, the temperature becomes 254 K, which is -19°C – well below the freezing point of water. So assuming the correct albedo leads to the paradoxical result that the only known habitable planet in the universe does not lie inside its star’s habitable zone.
To locate the outer edge of the habitable zone, we consider a hypothetical planet whose surface temperature is the freezing point of water, 273 K. Then we solve the equation to derive the distance. For the inner edge, we do the same thing, but using the temperature of boiling water, 373 K. Again, there are two versions. For an albedo of zero, the value for a black body, the Sun’s habitable zone extends from 83 million to 156 million kilometres. If we set the albedo to 0.3, the measured value for the Earth, then the habitable zone stretches from 69 million to 130 million kilometres.
The average distances of the four inner planets from the Sun, in kilometres, are 58 million for Mercury, 108 million for Venus, 150 million for Earth, and 228 million for Mars. So for albedo 0 the Earth just scrapes inside the Sun’s habitable zone ... but so does Venus. For albedo 0.3, only Venus lies inside the habitable zone. The Earth and Mars are too cold, Mercury too hot.
Why, then, is the Earth habitable? Because its atmosphere contains greenhouse gases, mainly carbon dioxide and water vapour, which trap incoming radiation and make it warmer than it would be if no atmosphere were present. But the usual concept of a habitable zone does not take the planet’s atmosphere into account. The idea that a star has a habitable zone, independent of properties of the relevant planet, is an oversimplification. Of course, in a broad qualitative sense it is true that if a planet is too near its star then any water present on its surface will boil, and if it’s too far away the water will freeze. But ‘habitable zone’ lends a misleading air of precision.
Greenhouse warming is just one of a huge variety of effects that between them pretty much demolish ‘habitable zone’ as a useful concept. The surface temperature of a planet depends on many factors, only one of which is how far it is from its star, for a given heat output. For example, clouds and ice can increase the albedo, cooling the planet; so can sulphur dioxide. Carbon dioxide, methane and water vapour can warm it. Feedback loops between different factors further complicate the possibilities: warming seas can create clouds that reflect heat and light back, decreasing ice cover can allow more heat and light in.
Even taking all this into account, it is not true that the only place where liquid water can exist is on the surface of a planet in the habitable zone. For example, it used to be thought that Mercury was locked in a spin – orbit resonance, rotating once during the same time it took to revolve once round the Sun. If so, the same side would always face the Sun, just as the same side of the Moon always faces the Earth (give or take a bit of wobbling, known as libration). In fact Mercury does not do this, but there’s every reason to expect that some worlds somewhere in the Galaxy might be very close to their star – much closer than the habitable zone defined above – and locked in such a resonance. In fact, this is the case for at least one exoplanet.16 If so, one side of the planet would be very hot, the other side very cold ... and in between there would be a belt with more moderate temperatures, suitable for liquid water to exist. ‘Just right’, in fact.
Astronomers are almost certain that liquid water exists on a number of bodies in our Solar System that are well outside the Sun’s habitable zone. Paramount among these is Europa, a satellite of Jupiter. There is convincing evidence that Europa, one-quarter the diameter of the Earth, has an ocean that contains as much water as all of Earth’s oceans put together. Yet Europa’s surface is solid ice. So where is the ocean?
Under the ice.
Measurements of Europa’s magnetic field reveal changes that currently seem to be consistent with only one thing: a worldgirdling ocean upon whose surface the ice floats. The water is kept warm by heat generated in Europa’s core, probably caused by repeated squeezing by Jupiter’s huge and powerful gravitational field. Jupiter’s inner three major satellites, Io, Europa and Ganymede, are trapped in an orbital resonance: while Io goes round four times, Europa goes round twice and Ganymede once. This creates unavoidable tidal forces, and squeezing causes friction, which heats the core. This is not such an outlandish scenario: the Earth’s continents and seabed of solid rock float on a vast underground ocean of magma.17
Europa is not alone in having such an ocean. Ganymede and Callisto may have one too, Io probably has one but it’s sulphur, not water, and Saturn’s moon Titan may have a subsurface ocean of slushy liquid methane.
Finally, there is the obvious point that the Earth’s own extremophiles live in conditions that are outside the habitable zone: water at temperatures above its normal boiling point, and below its normal freezing point. Not far outside, but outside all the same. Could they have evolved in such extreme conditions? That’s less clear, but we’ve already seen that a plausible theory of the origin of life has it evolving first as ... extremophiles.

Protective gas giants, stabilising moons, and Goldilocks orbits ... Rare Earth lists dozens of such factors, and like those three, most of them are open to serious challenge. But there is a more general issue, a mathematical point: logic.
It is all very well to list dozens of special features of the Earth, all of which definitely played a significant role in the evolution of life. But it is wrong to conclude from this (alone) that those features are necessary for life. The correct conclusion is that they were sufficient. ‘Sufficient’ means that with them, life arose. ‘Necessary’ means that without them, it would not have arisen. The two are different, and it is the first that the list of features supports. For you to get wet, it is sufficient to stand outside in the rain without protection. But that’s not necessary. You can fall in a lake or take a bath instead.
Evolution is a universal, and its main feature is that creatures evolve to suit their habitat. If some form of life can exist in some habitat, even if it seems hostile to us, then life can evolve to do so. It doesn’t care about our opinions, because we’re not going to be living there. If we approach questions about alien life with the tacit assumption that the only sensible form of life is us, we will ignore all the other possibilities. The word ‘extremophile’ is humancentred: it starts from where we are, and declares that to be what’s sensible and reasonable. The further away we go from our selfdefined centre, the more ‘extreme’ things become.
I remember a museum exhibit about deep-sea fish, which said something to the effect that ‘their strange shapes reflect the strange conditions under which they live’. It seems to make sense: strange conditions imply strange shapes. Not like normal conditions, which imply normal shapes. Like us. But it’s all back to front. Normal conditions, in this sense, are the ones we are accustomed to. So are normal shapes. But we are as different from the fish as they are from us, in both shape and habitat. To them, we would be strange and they would be normal.
To evolution, we would both be normal – relative to our habitat.

A more imaginative reading of the Goldilocks tale makes the same point, and raises a far more interesting set of questions. Mummy Bear’s wimpy porridge was too cold for Goldilocks, and Daddy Bear’s macho male porridge was too hot, while Baby Bear’s intermediate porridge was just right. And so they were – for Goldilocks.
For Mummy Bear, however, the intermediate porridge was too warm. For Daddy Bear, it was too cold. Goldilocks’ point of view is not privileged. Woolly-minded social relativism though it may be, I think that Mummy and Daddy Bear both had valid opinions too.
Discussions of such things as Jupiter’s alleged importance in protecting the Earth from comet impacts often run along the following lines: ‘Without Jupiter, the Earth would be hit by a comet every twenty years.’ There’s a sense in which such statements are true, but a closer look reveals that they don’t address anything of substance. They’re like sports commentators saying, ‘If only he hadn’t been offside, that goal he just scored would have won the match.’ But if the player had not been offside, he would have been in a different place. To score a goal, he would have to have kicked the ball differently. You can’t just change one factor, offside, and keep the rest exactly as before.
It’s the same with Jupiter. Yes, if you took the Solar System as it is today, and magically spirited Jupiter away, comets would rain in upon an unprotected Earth. But if the Solar System had evolved without Jupiter, it would not be the same – in all other respects – as it is. It would have been quite different. More comets would have hit the Earth in the past, for instance, leaving fewer to hit it now.
The mathematics of many bodies moving under gravity, called celestial mechanics, is revealing an unsuspected aspect of planetary systems. Namely, that they are systems. Over billions of years, they organise themselves in complex ways. The biggest planets, the Jupiter-like gas giants, have the biggest influence. Other, lesser worlds, and even those only slightly less massive, get rearranged until the entire system fits together and acts as a whole. This is celestial Gaia.
Very recently, it has been discovered that Jupiter’s influence on the Solar System has created a kind of celestial subway, a network of gravitational ‘tubes’ that can be perceived mathematically but consist of empty space.18 These tubes are pathways along which matter can move more efficiently. The arrangement has come about as a result of subtle feedback effects, caused by gravitation. The equations for gravity are nonlinear, meaning that effects are not proportional to causes. Nonlinear systems have a tendency to behave in surprisingly complicated ways, and they tend to organise themselves by settling into special forms of behaviour.
Considering a Solar System without Jupiter, and arguing that it wouldn’t be so hospitable for life, makes the same mistake as the sports commentator. It forgets that if you change one thing, you change everything. The evidence to date shows that most solar systems have huge planets like Jupiter. It seems likely that most of them also have smaller planets, though these are very difficult to spot at the moment. If so, then the Jupiters will organise their lesser brethren, and often enough there will be a few small worlds closer to the star, and some big ones further out. So even if it is indeed true that planets like Jupiter provide overall protection against comets and the like, it is no huge coincidence if one of them exists in roughly the right orbit. Nature does not simply build solar systems by plonking down planets at random. They have a selfconsistent structure.
This is not to say that habitable planets are inevitable. There are many ways to fail to be habitable. But there are many stars – at least 4×1022 of them – in the universe, and probably even more planets. There are many ways to be habitable too, and habitable planets will not all be carbon copies of the Earth. Rerun the Solar System again, from different beginnings, and there’s a fair chance that at least one world might still be suitable for life.
At some point in its history. Mars may have been suitable for Earth-like life, a billion or more years ago. In fact, it has been suggested that Earthly life was originally seeded from Mars. The current consensus is that this is probably wrong, but it’s not totally out of the question. It will take a close look at Mars to decide the matter.

In contrast to the Rare Earth story, I will describe some mathematical simulations and models devised by Harvard astrophysicists Dimitar Sasselov, Diana Valencia and Richard J. O’Connell, which suggest that planets capable of supporting Earth-like life may be far more common than has previously been thought.19 Their results also call into question the common view that Earth is the ideal kind of world for the kind of life that we find here.
The starting point for their work is the realisation that the conditions that make our kind of life possible do not necessarily require the planets concerned to be of a similar size to our own. What matters is that they should resemble our own world in one key respect: the occurrence of plate tectonics. There is a growing suspicion that the dynamic movement of continents helps to stabilise the Earth’s climate. In particular, carbon dioxide is recycled from the atmosphere to the ocean floor, where it is taken up by marine microorganisms and turned into carbonate; then the subducted carbonate is turned back into carbon dioxide by volcanoes. A stable climate helps liquid water to exist for geologically long periods of time, and water is required for our kind of life, even if other kinds might exist without it. That, in turn, enables evolution to generate complex water-dependent life forms.
It had been assumed that plate tectonics is rare, and that it requires a world of comparable size to our own. The crust of a much smaller world could not break up into suitable plates, while a much larger world would be a gas giant and not have a surface as such anyway. Sasselov and colleagues have shown that both these assumptions are false. Plate tectonics may actually be very common, and could occur on planets much larger than the Earth. The reason is the possibility of ‘super-Earths’: rocky worlds with a similar geological composition to ours, but having much greater mass. No one had previously investigated the internal geological processes of such a world, probably because no such exoplanets were then known. Indeed, virtually all known exoplanets were so large that they had to be gas giants, and this was still the case when the team began their modelling and published their first paper.
By 2005, however, the picture had already started to change with the discovery of the exoplanet GJ 876d, which orbits the star Gliese 876. This was smaller than the typical gas giant exoplanets then known, though still much larger than the Earth; there were hints that it might be mostly rock, rather than gas. However, there was no good way to measure the planet’s density, which would decide the issue, because the only known method required the planet to cross the face of its parent star when viewed from Earth. In 2009 a new exoplanet was found, CoRoT-7b, which did cross the face of its star. Now a density estimate was feasible, and the result was definitive: CoRoT-7b is made from rock. It has about 4.8 times the mass of the Earth and 1.7 times its radius. By 2010 a second super-Earth that transits its star had been located, known as GJ 1214b, with a density closer to that of water than rock, suggesting that it has a thick gaseous atmosphere. This planet has 6.5 times the mass of the Earth and 2.7 times its radius.
Now there were real planets to supplement the theoretical analysis made by Sasselov and his colleagues, adding to the interest of such calculations. They first showed that there are two main kinds of super-Earth: those with a lot of water, and those with much less. The first kind would have formed quite a long way out from the parent star, where they would pick up large amounts of ice. The second kind would have formed further in, and be relatively dry. Both kinds would acquire a large iron core as the denser parts of their molten material sank towards the centre, and a silicate mantle as the lighter materials rose. The water-rich super-Earths would have very deep oceans above the mantle; the drier ones would have thin oceans, or none.
Because the pressure at the centre of a large super-Earth is higher than it is for our own world, the iron core will solidify faster. This probably implies that such a planet will have little or no magnetic field, and that might be bad for the occurrence of life, because magnetic fields shield the surface from radiation. However, so do deep oceans, and in any case we don’t know how necessary a magnetic field really is. Some Earthly bacteria are radiationresistant, for example.
The interior of a large super-Earth should contain the radioactive elements uranium and thorium, which generate most of the heat that keeps our own planet’s core molten. Because these elements occur in much the same proportions throughout the Galaxy, the large super-Earth would have more of them than our own world does, and its core would be considerably hotter. The extra heat would cause convection in the mantle to be more vigorous, and this in turn would drive the movement of large plates at the boundary of the rock, much as it does on Earth. It turns out that these plates would be thinner than they are on Earth, because they move more rapidly and so have less time to thicken up by cooling. They would be easier to deform, except that the planet’s greater gravity exerts more pressure on fault lines, so the plates don’t slide as easily as they do here. These two effects tend to cancel out, so overall the frictional resistance when the plates slide past one another is much the same, regardless of size.
In short: plate tectonics is likely to be more common on large super-Earths than it is on Earth-like terrestrial planets that are similar in size to our own world. It also happens faster, which means that the cycle of subduction and volcanic activity that tends to keep the carbon dioxide concentration fairly stable would, if anything, work better. So a super-Earth that is considerably larger than our own world would probably have a more stable climate than ours, on geological timescales, making it easier for complex life to evolve.
This analysis completely changes the ‘rare Earth’ picture. Terrestrial planets, roughly the same size as our own, should occur fairly often, but comparatively speaking they are probably fairly rare. But the likely number of super-Earths in the Galaxy is far greater than the number of terrestrial planets, so the prospects for life are much better than they would appear if we were to focus solely on terrestrial planets. It also casts serious doubt on Goldilocks arguments, because it turns out that the Earth, far from being ‘just right’ for plate tectonics to arise, is very close to the lower extreme of the range of sizes for which such effects can happen. If the Earth were slightly smaller, it would not have plate tectonics, and that might have caused it not to evolve complex life.
The ideal Earth-like planet, it seems on this analysis, is considerably larger than the Earth. We just scraped into the acceptable range.
There is a general message in this work, and it is one that needs to be far more widely appreciated. The way to understand how likely alien life might be is not to focus on conditions that are virtually identical to those found on this world, and then argue – typically confusing sufficiency with necessity – that only those conditions are suitable for life. What really matters is just how different a planet can be from ours, and still support its own form of life, adapted to its prevailing conditions by evolution.
How diverse can living creatures, and their worlds, be? You won’t find out if you start by assuming they all have to be just like us.

Perhaps alien life has already been discovered.
In 1997 NASA launched the Cassini – Huygens spacecraft, a mission to Saturn. Seven years later the craft reached its destination. The Huygens probe landed on one of Saturn’s moons, Titan. The Cassini spacecraft went into orbit round the planet. One of the early discoveries – dramatic, though to some extent expected – was that Titan has lakes. Because of the deep cold at that distance from the Sun, the lakes are not of water, but liquid methane and ethane.
Now some scientists are wondering whether Cassini has found signs of an exotic kind of life. This is one possible explanation for the strange behaviour of two gases on Titan: hydrogen and acetylene. There ought to be quite a lot of hydrogen, spread fairly uniformly in the moon’s atmosphere. Darrell Strobel, working at Johns Hopkins University, has discovered that the hydrogen streams downwards through the atmosphere and disappears near the surface. Astronomers had expected acetylene to be fairly common too, produced by simple chemical reactions in Titan’s atmosphere and deposited on the surface.
But there isn’t any.
In 2005 Chris MacKay, a planetary scientist at NASA, realised that hypothetical methane-based microbial life would be very likely to get its energy by reacting hydrogen with acetylene, in the same way that most Earthly life reacts oxygen with molecules that contain carbon. The new observations are consistent with this kind of life inhabiting Titan’s surface, and using up all of the missing hydrogen and acetylene.
Of course this doesn’t come close to a proof that Titan harbours such exotic life forms, and Mark Allen (also at NASA) has suggested that non-living processes are a more likely explanation. Cosmic rays could convert acetylene to more complex substances, for instance, when they collide with its molecules. But it does illustrate the value of not assuming that life everywhere must be very much like life here. By doing so, we could have stood a small but significant chance of missing an alien life form in our own backyard.
Watch this space