There are various notions of chaos, in myth, science and philosophy:

Nietzsche in the Gay Science wrote:

the total character of the world is chaos for all eternity - in the sense of not of a lack of necessity but of a lack of order, arrangement, form, beauty, wisdom.

This is a repudation of Gods hand in the world (order), but an affirmation of science (necessity) in the world.

But I find it provocative to consider whether a perfect chaos is possible; one that, that is one in which is the exact anti-thesis of order? That is no information can be deduced? Or is it a self-contradictory notion?

For example, take a perfect gas; one can deduce pressure, temperature and entropy; its volume, desnity and mass. These are bulk properties; but one can also deduce information about its micro-motions aka Brownian Motion - that is a random walk as a limit.

For a perfect chaos, one should be unable todeduce any information in principle; if not in fact. One could say then it indescribable.

Is this possible? The apeiron of Anaximander - the boundless - seems a possibility; but as Nietzsche notes, it must have an inner law of necessity that forms the world from it.

Another possibility is Kants noumenon; it is explicitly indescribable.

  • 2
    Can I push on the definition of "perfect chaos?" The word has many meanings, and which meaning you choose will shape the answers. Consider the mathematical definition of chaos actually has some order to it (in the form of stable orbits). Mathematically pure-randomness has no order, but it subject to the Central Limit theorem (which leads to what you see as bulk properties). I bring these up because Nietzche's definition of chaos is less exacting than the mathematical version, and if you are talking about "perfect chaos," exacting definitions become helpful.
    – Cort Ammon
    Jan 20, 2015 at 16:43
  • @CortAmmon: Sure; is there a mathematical chaos that has neither order in its 'micro-structure' or in its 'bulk-structure'? Nietzsches imprecise, because he's talking in very generally terms; and not about mathematics but about matter; so his necessity is determinism broadly construed. Jan 21, 2015 at 13:27
  • The closest I can think of to that definition is Basyean inference with a Jeffries Prior, which I believe can be phrased as claiming not that the world is unordered, but rather that any ordering that may exist is unknown. Is that starting to approach a direction which is reasonable for starting an answer?
    – Cort Ammon
    Jan 21, 2015 at 15:47
  • @CortAmmon: Its closer; but I'd put it as 'epistemological perfect chaos' rather than an 'ontological perfect chaos'; I think its the second which I'm after; its worth pointing out that N doesn't think its possible given his assertion above. Jan 22, 2015 at 10:34
  • he doesn't mean chaos per se, else there would be no repetition surely ?
    – user6917
    Jan 23, 2015 at 6:02

1 Answer 1


Non-expert ideas, but since this question has been up for a couple of days, here are some thoughts..

Although this sounds like a straightforward question, really I think this question unravels into what's been called the 'Laplacean spirit'. Either that, or the terms wind up recursively redefining themselves.

If it were possible for human understanding to raise itself to the ideal of the Laplacean spirit, the universe in every single detail past and future would be completely transparent. "For such a spirit the hairs on our head would be numbered and no sparrow would fall to the ground without his knowledge. He would be a prophet facing forward and backward for whom the universe would be a single fact, one great truth." And yet this one truth would present only a limited and partial aspect of the totality of being, of genuine "reality." For reality contains vast and important domains which must remain forever and in principle inaccessible to the kind of scientific knowledge thus described. No enhancement or intensification of this knowledge can bring us a step nearer to the inner mysteries of being. -Cassirer, 'The Laplacean Spirit'

Also, I think perfect ontological chaos entails perfect epistemological chaos.

Firstly, I would think perfect ontological chaos would entail an infinite domain.

Assume an unchanging binary sequence 1011. Assume we consider it as a closed system, then we might reconfigure our own interpretative system to give it order: 'First half is opposites; right side is all 1s'. Or we might say, 'Alternating sequence, except for last item'. We can collect up all of these interpretations to give the required information resources of the observer system, and the information associated with the observed system. In any case, the information content will obviously be finite, and therefore we have not reached perfect chaos.

Likewise, if we consider this a peek into a larger system, such as ..1011.. , then we still will develop strategies for defining the extension of our knowledge. Strategies that would extrapolate this in unusual ways would be weighted low; obvious strategies would be considered likely. For instance, if we flip a coin a hundred times in a row, and it comes up heads every time, we would expect it to be a trick coin, and therefore would guess heads on a fair toss; our strategy would be 'heads'. In any case, let's just assume our observer system is not 'perfect chaos'. I think a strong argument could be made that perfect chaos cannot 'observe' or 'make meaningful representations'.

I also propose that the 'unchanging binary sequence' example can be generalized to any finite system under observation while maintaining these conclusions.

So now, in order to approach infinite information, we must approach infinite resources for storing our interpretive program and the data under consideration; therefore perfect chaos entails infinite domain.

Now, if we are dealing with an infinite domain but consider ourselves finite 'strategies', then we can only observe this infinite domain partially at any given time. We must make assumptions as we go, but we cannot make sense of it, therefore perfect ontological chaos entails perfect epistemological chaos.

To show this in better detail, let's say that, as we go, step by step observing, we constantly restructure our strategies. Our strategies might make sense to us for some steps, but over all steps, the information content must be infinite. In this case, it doesn't matter that things made sense 'in the moment', and that we were able to handle them with our finite resources, we can be assured that we cannot in the full scope fall back on old information. Whether we actually can experience infinite 'steps' is a question for Zeno, Planck, or spiritualists. In this case, we may be 'saved' from the possibility of perfect chaos by our own finitude.

Either that, or we might redefine perfect chaos to be the maximization of disorder related to one's own perceptual system. But then it no longer seems a philosophically pregnant term.

Our perceptions have a great ablility to 'scope' information, such as when a complex rhythm played quickly enough becomes a 'timbre', and also to link perceptions together, such as when moving images kick in proprioceptive feelings.

  • There's an important and problematic ambiguity in the meaning of "perfect ontological chaos." Presumably, what is meant is the complete absence of any order. But the ambiguity is in what order means. It seems order can either mean "putting all my ducks in a row" or things like 2/9 will have an ever repeating series of 2s, which is an "order."
    – virmaior
    Jan 23, 2015 at 2:23
  • This ambiguity then destroys the entailment because it creates four possibilities: a perfect ontological chaos that entails epistemological chaos, a p. ont that fails to entail (we pattern what has no intentional pattern -- hinged on the ambiguity of "order"). Alternately, the existence of non-intentional pattern meaning order means "perfect ontological chaos" is impossible, but that does not entail that we can understand it (or that we cannot) = possibilities three and four.
    – virmaior
    Jan 23, 2015 at 2:24
  • It sounds like you are arguing about the constructual validity of any definition of information, which is a deep problem I do not fully understand. I tried to give my best understanding. A disorder requires a frame of reference, 'the order', in which to be considered disordered. I tried to encapsulate this in the 'observer system' and the unproved statement 'the example can be generalized to any finite system'.
    – dwn
    Jan 23, 2015 at 3:02

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .