Seriation and the Existential Principle
The View from Oregon – 344: Friday 06 June 2025
In 2022 I gave a couple of presentations—both in Scotland, as one invitation followed another—in which I pointed out that the seriation technique that archaeologists use is formally the same technique as astrophysics has employed to reconstruct stellar evolution. Since stellar evolution takes place at a scale of time that makes the observation of a single star inadequate to say anything about the evolution of that star, we have to look at different stars at different places in the sky and at different stages in their evolutionary development, distinguish discrete stages in this evolutionary development, and then arrange these stages into a generic sequence that describes the evolution of all stars of that type (with the type corresponding the size of the star, which stands in direct relation to its mass, temperature, and lifespan). These two very different examples of scientific observation—pot sherds and stars—can obscure the fact that the technique of seriation is formally
continuous across these epistemic domains. In my talks I said that, as exoplanet atmospheric spectroscopy improves (I say “improves” because we already have the earliest proof of concept from the JWST), we may be able to apply the technique to biospheres, yielding one or several generic sequences of typical biospheric evolution. I still think that this is a great idea, but I haven’t yet attempted to write it up as a paper.
I have come to see seriation as a distinctive category of historical thinking that can be applied across scientific disciplines for the reconstruction of histories, whether these be histories of pottery at a number of related archaeological sites or histories of stars and galaxies. And I have come to see historical thinking as essential not only to what are sometimes called the historical sciences, but to all sciences. Some years ago I formulated (but haven’t published or discussed anywhere) what I call the existential principle: everything that exists has a natural history. This is an explicitly metaphysical principle—I make no bones about that—but it is a metaphysical principle that systematically relates existence to the actual world. Unicorns don’t have a natural history, so they don’t exist. Stories about unicorns, on the other hand, do have a natural history. We could reconstruct that natural history and produce a history of unicorn stories, but a natural history of unicorns would be just another unicorn story.
The existential principle as I have formulated it is extremely simple, but any metaphysical idea if challenged with enough examples takes us into philosophically disputed territory. I haven’t put a lot of thought into finding (apparent) counter-examples for the existential principle, but one problem that occurs to me is that existents have what we could call a customary mode of existence, and we have to distinguish between existents that are true to this customary mode of existence and existents that are not. Elaborating this distinction would, I think, prove to be philosophically difficult. Again, to take unicorns as an example, we imagine unicorns as flesh-and-blood biological beings—like a horse, maybe split off from the evolutionary branch of horses at some point—so the way that we think unicorns ought to exist is not the way in which they do exist, i.e., in stories and in imaginative representations. Thus a unicorn fails to exist in the customary way that flesh-and-blood biological beings exist, which is the way that we think unicorns ought to be exist in order to be considered “real.” I think this is intuitive, but expressing this theoretically and in the requisite generality would be a real philosophical challenge. Why do we think that certain kinds of existents ought to exist in certain ways? And what are the kinds of existence? One can easily see that attempting to answer these questions would launch one on a difficult metaphysical inquiry.
Given the existential principle, i.e., given that all that exists has a natural history, the natural history of any existent can be reconstructed, and seriation is the way we reconstruct natural histories, and it is the primary way that we reconstruct the natural histories of existents that present challenges to observation, like stars, whose evolutionary development plays out over billions of years and therefore cannot be exhaustively observed by any human scientific research program. Dyson’s eternal intelligences could plausibly observe the entire evolutionary history of a star, but insofar as eternal intelligences would exist after the Stelliferous Era, there would be no stars for them to observe, and the remaining astrophysical processes (during what Adams and Laughlin call the Black Hole Era and the Dark Era) are proportionally extended in cosmological time, so even eternal intelligence would have a problem observing what remains of the universe.
Now that I think about it, the claim that “the natural history of any existent can be reconstructed” should be separated out as a distinct thesis that needs to be demonstrated independently from other claims about the existential principle. It is entirely possible that there are existents for which the natural history, for some reason, cannot be reconstructed. Say, for example, a unique existent that exists only at one place and during one time period, and it either was not observed during its unique existence or it could not observed (for example, like stars it may exist for too long to be continuously observed)—such a unique existent would constitute a counterexample to the possibility of seriation. Call this the reconstruction thesis, with the weak reconstruction thesis being that the natural history of any existent can, in principle, be reconstructed, in contrast to the strong reconstruction thesis, such that the natural history of any existent can be, in fact, reconstructed.
Some existents exist on a time scale that observing their natural history is not difficult. The lives of fruit flies, for example, can be observed in a laboratory. A long-lived human being can observe the entire life of a human being who doesn’t live as long. Beyond a certain temporal threshold with both a lower bound and an upper bound, observing and documenting natural histories becomes difficult, and at some point it becomes impossible (though whether in practice only, or in theory also, remains to be established). Rapid processes have been filmed with high-speed cameras and made it possible for us to observe extremely short-lived events at a human scale of time. Even shorter scales of time, such as those studied in subatomic particle physics, require more sophisticated technologies. Below the duration of Planck time it’s probably not possible for us to observe anything, although formally we can interpolate a scale of time of any arbitrarily shorter duration than Planck time. At the other end of the scale, with extremely long-lived or extremely large (in spatial extent) existents there are also problems with the reconstruction of natural histories.
I started out with the seriation of stellar evolution, and to a certain extent we can also engage in galactic seriation. The observable universe as a whole, however, admits of only a single instance, so, as far as our knowledge extends at present, we can’t engage in cosmological seriation. However, it is not impossible that some technology will allow us to transcend our universe of origin and to see it from the outside, as it were, and alongside other universes. Should such technologies and techniques of observation ever become available to intelligent beings, it might be possible to reconstruct the natural history of the universe entire.
Even though direct human observation is boxed in by upper and lower bounds of physical scale, these bounds are quite a bit larger than we might expect for an observer of the size and longevity of a human being. Because of our ability to observe the heavens, we’ve had access to observations on a scale far larger than our homeworld. In other contexts I’ve discussed our ability to measure the speed of light before we possessed truly precision instruments because observing, say, the moons of Jupiter, at different places in Earth’s orbit and different relative positions of Jupiter allowed us to compare relatively rough measurements on a cosmologically local scale. And in constructing the cosmological distance ladder we have a tool that can measure the cosmos entire, and all from the surface of a particular planet in a particular planetary system in a particular galaxy. I never cease to be amazed by this accomplishment.
Despite the accomplishment, there remains much that still lies beyond the bounds of our observation—even observation supplemented by seriation. Red dwarf stars, for example, are predicted to have an evolutionary development that extended from hundreds of billions of years perhaps to a trillion or more years, but the universe is only a bit older than a dozen billion years (by most estimates), and by this time scale red dwarfs have barely started their lives. Some of the most interesting evolutionary developments in the lives of red dwarf stars have yet to occur anywhere in the universe. Observation with seriation can reconstruct the stellar evolution of stars that burn for as long as a few billion years, since they’ve had time to run their course, and we’ve had an opportunity to observe these stars at all stages in their history (except for the long term fate of stellar remnants). Of course, seriation doesn’t occur in a vacuum; astrophysicists also model the interiors of stars, and the better these models reflect our observations with stars that have run their course of stellar evolution, the better they can be trusted to predict the evolution of M and K stars, the evolution of which we have not observed. Also far beyond our ability to observe is the possibility of the decay of fundamental particles. We don’t know whether or not proton decay occurs, but proton decay is a prediction of several physical theories, with no observational confirmation. If we could wait for the far future of the universe trillions of years from now, we might be able to observe proton decay, but it’s likely to remain hypothetical for some time to come.
The observations of natural phenomenon I’ve used as examples are processes that have been ongoing in the universe in some cases for billions of years, which is why they’ve left evidence of their different stages of development that we can then piece together through seriation. What is distinctive about the suite of emergent complexities that have appeared predicated upon the evolution of human beings and the activities of human beings is that none of these are older than human beings, by definition. We don’t and can’t study the histories of intelligence and civilization at their largest scale, because human-level intelligence has only been around for a few hundred thousand years, and civilization has only been around a few thousand years. If civilization is the kind of thing that can endure for millions of years or billions of years, displaying all the while new forms of development, we can’t know this on the basis of observations of ourselves on Earth.
If our observation included other intelligent species or other civilizations on other worlds, then we could at least begin the process of the seriation of intelligence or the seriation of civilization, but that scientific research program would require both the existence of other intelligent beings and our ability to observe them, and especially our ability to observe them at different stages of development. That, in turn, means the more examples we have of a form of emergent complexity that has fully run its course many times over, the better to converge on a generic evolutionary developmental sequence for that form of emergent complexity. Whether intelligence stays in its embodiment of origin and runs its course on its homeworld until its biosphere is exhausted, or whether intelligence passes through multiple re-embodiments such as I discussed in last week’s newsletter—both possible developmental sequences for intelligence—will be relevant to our future.