How Many Earths?

How Many Earths?

Let’s start with a couple of thought experiments.[1] Suppose there is a certain planet that has been around for a billion years; call it Janet. Not much has changed in it during this time and it has received no bombardments from abroad. Then one day it is subjected to an intense heat wave caused by getting too close to a sun or by incoming meteors. This influx of heat causes Janet to undergo a thorough geological transformation: the rocks that compose it metamorphose into different kinds of rock. We feel inclined to say that Janet is not the same planet she was; indeed, Janet no longer exists, having been transformed into a different planet. The case is like transforming a person so comprehensively that we cannot speak of the same person. Janet now looks completely different, is half her previous size, and has a different molecular composition. She might even have been split into two. Janet is no more. Now consider another planet, Joan, that receives a different kind of treatment: Joan is subjected to heavy sustained bombardment from outer space, adding tremendously to its mass. It is now ten times bigger than it was and hosts types of matter alien to its original makeup. It isn’t itself internally modified, but it is covered with alien material. We would say that Joan still exists but the planet that has been created is not identical to Joan; it is a new planet. It is as if the new planet has swallowed the old planet whole. Here we would speak of two planets not one—old Joan and new Joan. Obviously, we can construct other scenarios of a similar kind; for example, we could imagine a planet, Julie, that is pulled apart by a strong gravitational field and now consists of spatially separate parts held together by a cosmic thread—it looks like three balls held together by string. Does Julie still exist? Judgments of planetary identity are not always straightforward; some changes can make us question a planet’s identity. We might even discover that a planet we thought was one is really two: under its surface are two previous planets that have tenuously joined forces, giving an appearance of unity that is not borne out by the facts (they have quite different geologies).

My question, then, is whether the earth is as unitary as we tend to suppose. Is it really one planet? What makes a planet into a single object? What are the criteria of identity associated with the sortal “planet”? Consider a planet very different from Earth called Wendy: Wendy is spherical like Earth but is made of only one type of rock, say granite; it is homogeneous all through. There is no molten core, no atmosphere, and no alien bombardment in its past. It is exactly as it was when it came into existence ten billion years ago—no life on it, no weather to speak of, no geological upheavals. Wendy is completely static. Nor does it vary from place to place—nothing like the arctic or the tropics. It is a very dull planet. But it is certainly unitary over time; it has not gone out of existence since its birth, replaced by a numerically distinct planet. Wendy hasn’t changed a bit. It’s the same old dependable Wendy over a lifetime of ten billion years. This is nothing like planet Earth, whose career has been notably dynamic: huge geological changes, much bombardment, considerable accretion, hard on the outside and soft in the center, continuously changing, and home to myriad life forms. Earth is barely recognizable from its early days; not the same planet at all. Wouldn’t it be proper to speak of Old Earth and New Earth? Isn’t that more respectful of the facts on the ground? Doesn’t it give a more accurate picture of what we loosely call “Earth”? Don’t we talk the way we do out of convenience not ontological veracity? It’s like the way we talk about towns: towns change dramatically over time, yet we speak of them as one. Isn’t the London of today a different town from the London of its first incarnation (a bunch of mud huts by a river)? Aren’t there really many Londons, over time and at a time? Don’t we speak of a single London only by convention? Certainly, being in roughly the same place is not sufficient for numerical identity.

I would therefore like to propose that we revise our thinking about the earth’s identity to take account of its actual nature. There are really several Earths. True, they are tightly crowded together, like cattle in a corral, but they are sufficiently distinct to warrant names of their own; the collection of them is crudely designated “Earth”. And I recommend going the whole hog: there are many Earths. There is crust Earth, mantle Earth, outer core Earth, inner core Earth, ocean Earth, atmosphere Earth. Crust earth can itself be divided into different Earths, according to the type of rock forming the rock strata. Then we have the different regions of surface Earth: arctic Earth, temperate Earth, tropical Earth. In addition, and importantly, we have botanical Earth, animal Earth, and psychological Earth.[2] Each of these is part of the totality we call “Earth”, though spatially separate. They each constitute different “worlds”: the world of molten lava, the solid crust world, the oceanic world, the plant world, etc. The main division is between the original world of early Earth consisting largely of hot molten rock and the cooler post-bombardment world of later Earth. The latter then divides into pre-biotic Earth and biotic Earth. We might then go on to distinguish the pre-mental Earth from the mental Earth. It would not be unreasonable to suppose that the original planet does not exist anymore, having been replaced by the newer model, the result of serious work being done on the original (like a thoroughly renovated house). That hot hell-hole no longer exists, being replaced by our very habitable semi-paradise. It looks and feels completely different. Its geological composition is completely different. It is no more the same planet than the original Earth was the same celestial object as the cloud of dust from which it condensed. Too much transformation, too much metamorphosis. From lava to life. What if planet Earth is transformed even more dramatically in the future, say by nanotechnology driven by AI—different geology, different climate, different everything, completely unrecognizable? There can be many Earths over time and many at a time. We live on a multitude of Earths, just as we live in a multitude of towns. Each should have its own name in an ontologically perfect language.

The earth is like the brain. What we call the brain is really composed of many sub-brains, and it is pretty arbitrary where we draw its boundaries (are the retinae part of the brain, or the whole nervous system?). We already have labels for the parts considered as separate modules; it is those that matter not the brain as a totality. The whole brain has no identifiable function; it’s the parts of it that have functions. It is easy to imagine a brain as a loose collection of separate entities linked causally; there is no necessity about cramming all the parts into a confined space. The various sub-brains could each have their own place of residence distributed all over the body (intellect in the toes, say). It’s the components that matter not the way they are packaged. Similarly, the earth is a package of sub-earths, each with its own identity, its boundaries unclear (does it include the clouds and atmosphere?). How many brains do I have? How many earths do I live on? Many, in both cases. Words like “brain” and “planet” are not well-behaved sortals, carving nature at its joints; they are folk terms introduced for convenience. They suggest more homogeneity than actually exists in their designations. We exist in a plethora of worlds; we live on a plurality of planets. If the molten core of Earth were to evaporate, leaving only the crust, we would still have a planet called Earth to call home—crust Earth, that is. If the crust were to disappear, leaving only the molten core, we would still have an Earth, though not one we could live on. I think, in fact, that throughout history the reference of “Earth” in people’s mouths was the part of Earth on which they lived—a small section of the earth’s surface. Most of what we now call Earth was outside their ken and conceptual repertoire; now we take in much more, but remember that many people know nothing of the full reality of the place in the universe they occupy. The folk never intended to take in the whole kit and caboodle, just a local slice of it. They were right in that there are many Earths; there is no single Earth, no natural unit, no primordial individual substance. There is no single thing that persists from Earth’s birth to now. There are many Earths whose fate we are (rightly) concerned about, not just one. When you gaze at one of those pictures of Earth seen from a distance, you are seeing many Earths not a single Earth. If Earth dies, many Earths die. If some Earths die, not all do.[3]

[1] I view this essay as a contribution to the philosophy of astronomy, including Earth science. It might help to consider my discussion of “planet” in conjunction with Quine’s discussion of “rabbit”, though rabbits are better defined.

[2] See my “A Philosophy of Nature” on the continuity of biotic and non-biotic Earth.

[3] We are told that Mars used to be a warm watery planet and is now a cold dry one. That planet is no more, having been replaced by a much less hospitable planet. What we see now is a dead body, a remnant. How many planets orbit the sun? We are accustomed to saying eight, but properly individuated there are many more. We alone have about twelve by my count. If we took all of our life forms, along with a suitable quantity of soil and rock, to another solar system and set up camp, we would still be living on Earth by my reckoning—an Earth anyway. If a cataclysm stripped Earth of all life, leaving only an inorganic hunk of rock, we would be left with an Earth of sorts, but not the one we know and love. We could call this the “Many Earths Hypothesis”, though it is not so much a hypothesis as a fact. The word “Earth”, without any attached modifier, suffers from indeterminacy of reference; context usually helps to narrow it down. It is high time we got more specific about what we are talking about.