Idealization in the Study of History and Civilization
The View from Oregon — 349: Friday 11 July 2025
I was reading through a notebook from a few months ago and found a paragraph in which I argued that it’s possible to imagine an ideal civilization, but an ideal history doesn’t make much sense. There are a lot of elusive intuitions behind these two interrelated claims, and these claims touch on topics I’ve been thinking about for many years. Taking these two positions of the comprehensibility of an ideal civilization and the incomprehensibility of an ideal history, together they fly in the face of another elusive intuition that intimately connects history and civilization. This is something that I’ve discussed on many occasions, and I hope to write a paper about the relationship of history to civilization. However, my thoughts aren’t yet fully formed on this because, as I’ve already said a couple of times, the background intuitions involved in these judgments are elusive. And we’re struck with elusive intuitions until we have both a science of history and a science of civilization that would make these intuitions explicit and formulate them within a conceptual framework capable of criticism and revision in the light of empirical tests. Until our ideas of history and civilization are formalized, we must make do with a philosophical approach to the problem.
The intuition at work in the idea of an ideal or perfect civilization is that we typically invest the idea of civilization with many ideals, and this has in fact been the basis of much criticism of the concept of civilization over the past few decades, and why one writer (John Armstrong) called civilization a “tarnished ideal.” When Ideals are given as the explanation for institutions and actions that fall far short of the professed ideal, the charge of hypocrisy is not far behind. A gap opens up — sometimes a gap that becomes a yawning chasm — between professed ideals and actual practices that, if allowed to grow, becomes are argument against the ideal, which in this case is the ideal of civilization in general, or the ideal of some particular civilization.
Probably in a thorough discussion of this it would become necessary to have some kind of informal definition of the ideal and the perfect so as the distinguish them if they are to be distinguished, or to show that they are identical (if they are). In ordinary language, “ideal” can carry different connotations than “perfection,” and when we’re dealing with elusive intuitions, subtle connotations may carry meanings that we can’t afford to ignore. If I say that a given civilization possesses such-and-such ideals and has such-and-such perfections, these are, intuitively, distinct claims. For example, “ideals” carries a connotation of possibly being realized in the future, and serving as guides in the present even when not yet realized, while “perfections” carry the connotation of a state that obtains in the present. If I say that a civilization has a given perfection, but that perfection fails to obtain, but I mean that it might obtain in the future, that wouldn’t make much sense. However, if I were to invoke an “ideal civilization” or a “perfect civilization” the connotation changes yet again. We might aspire to having a perfect civilization, in which case a perfect civilization is our ideal. But if we were to invoke an ideal civilization, this might be ideal in any number of ways and not necessarily ideal in the sense of perfection, or some specific perfection (which a given civilization might conceivably possess or not possess).
Ideals also bring in the problems of ideals as they are used in science. I’ve been thinking about this problem a lot lately. It’s interesting to me that in continential philosophy of science there is extensive discussion of idealization in science, while in analytical philosophy there is a grudging recognition of the role that ideals play in science, but no real philosophical engagement with the problems that this poses. This is quite similar to the case with abstraction, which is grudgingly recognized but it is rare to find a chapter on abstractions in an analytical textbook on philosophy of science — though you may find a chapter on theoretical terms that will indirectly take on some of the problems of abstract concepts and their role in scientific knowledge.
One of the most familiar instances of an idealization in science is that of the ideal pendulum. Actual pendulums can be made precise to an arbitrary degree, but they can never converge on the properties of an ideal pendulum. I started to make a list of the properties attributed to an ideal pendulum, and found many accounts of the idea with a number of overlapping concepts found in multiple expositions, and some concepts rather less widely recognized. My list, to date, is as follows:
- Frictionless fulcrum
- Weightless string
- Point mass weight bob
- Absence of air drag
- Short amplitude
- Inelasticity of pendulum length
- Rigidity of the pendulum support
- Uniform field of gravity under operation of the pendulum
A little more research and a little more thought might reveal more properties of an ideal pendulum, but this is a good list to start. Before I did any reading about this I wouldn’t have mentioned “short amplitude” as a property of an ideal pendulum, but I found this in several sources. This is called the small-angle approximation and it makes the mathematics of an ideal pendulum easier. One source says, “If the amplitude of angular displacement is small enough that the small angle approximation (sinθ≈θ) holds true, then the equation of motion reduces to the equation of simple harmonic motion…” This might seem a bit abstruse, but it is of the essence of the problem. There is a difference between an ideal pendulum and an actual pendulum because an ideal pendulum is mathematically modeled with relative ease, whereas an actual pendulum would be extraordinarily difficult to model in every wobble and tremor that it makes.
In any case, we can incorporate a short amplitude without difficulty, but the other properties of an ideal pendulum are difficult or impossible to build, though, as noted above, they can be approximated to any degree of precision. We can’t eliminate air drag, but we can construct a pendulum inside a box from which air has been pumped out, and by technological means we can probably produce a vacuum better than the vacuum of intergalactic space — but it’s still not a perfect vacuum. This might sound like an insignificant detail, but details like this can play an important role in scientific theory.
When at the 1930 Solvay conference Einstein confronted Bohr with the photon box thought experiment, Bohr’s response was that a single electron leaving the photon box would make the box lighter, thus rising in the gravitational field and changing the gravitational influence on the clock that is also part of the thought experiment (because the clock is now higher, further from the gravitational influence, and therefore running faster). If a principle in physics (or an entire branch of physics, like quantum theory) can stand or fall on the influence of the the weight of a single photon on a clock, then the forces acting upon an actual pendulum (in contradistinction to an ideal pendulum) seem relatively coarse in comparison. It may or may not be the case that an ideal pendulum is the same thing as a perfect pendulum — again, connotation matters — but if an ideal civilization is analogous to an ideal pendulum, then we’re talking about something entirely different from the ideals usually associated with civilization and from whatever we might deem a perfect civilization.
The intuitions that seem to be involved in a perfect or ideal history are different from the intuitions involved in perfect or ideal civilization — probably, again, the result of the connotations that come into play in ordinary language. My initial thought was that, whereas we can find a way to make sense of the concept of an ideal or perfect civilization, these predicates don’t seem to bear at all on what we call history. Even if we can’t spell out all the properties of an existing perfect civilization, we can imagine ourselves engaged in a project like this, as, for example, in some grand design of social engineering in which we would formulate the constitution of a perfect civilization from scratch. Howver, I can’t imagine the same for history, i.e., even under ideal circumstances, I can’t imagine a social engineering project in which we would design the perfect history, even if it isn’t our history, and even if it can’t be realized in fact. History must start from what is, and not from what ought to be, so that even if we can project a perfect history into the future, the sequence of events that converge on that perfect history are themselves part of history, and they are moreover part of imperfect history. Perhaps if we started the universe from scratch, and began with a perfectly blank slate, then a perfect history might be written out on that blank slate, but this wouldn’t be our universe, but some other counter-factual universe.
In several of my episodes of Today in Philosophy of History I’ve talked about various criteria for identifying civilization or identifying history. In many discussions there is a de facto background assumption that history only applies to civilizations. This seems to be the case, for example, with Hugh Trevor-Roper’s “purposive movement” criterion for history, which is missing in the events of pre- or non-historical peoples. Events still occur in the lives of individuals who inhabit non-historical societies, but the events are indifferent because, according to Trevor-Roper, they exhibit no purposive movement. Others have made similar arguments without appealing to Trevor-Roper’s criterion, instead implicitly or explicitly invoking some other criterion by which civilized societies have actual histories and pre- or non-civilized societies have no real history. I am not advocating this point of view, but it is a common one and needs to be made explicit.
Obviously, there’s a relationship here between civilization and history (and this is the focus of my unfinished paper), and if something like this relationship obtains, then, if civilization can be called ideal or perfect, then history could be called ideal or perfect. I could even employ another term and further confuse matters by appealing to the possibility of an exemplary civilization or an exemplary history. To be exemplary is, again, slightly different from being ideal or being perfect. A civilization might be exemplary without having realized some ideal it is striving toward, or without being perfect. I think that what Clive Bell called paragons of civilization in his Civliization: An Essay could also be called exemplary civilizations. We wouldn’t call Periclean Athens or early modern France perfect civilizations, and we might not call them ideal (however, I can imagine someone making the argument that these paragons are ideal), but they certainly are exemplary.
This is one way to argue with my initial intuitive claim that there might be perfect civilizations but there is no perfect history. As I have been musing on this problem I’ve come to think that there might be other ways to argue with my initial intuition, and there might be senses in which history is ideal, or perfect, or exemplary. An ideal history analogous to an ideal pendulum wouldn’t invoke any of the glittering generalities of our more common use of “ideal,” as when we might invoke, for example, the ideals of liberty, equality, and fraternity. And ideal history in this scientific sense would simply be a history that a science of history would appeal to as a model, amenable to formalization and mathematical modeling, but only an approximation of civilizations in fact. Actual civilizations also approximate ideal civilizations, but in a different sense. As is to be expected, there are a lot of subtleties involved in the attempt to capture an elusive idea.
