The world is so random and it is not.
Mysterious, strange in every spot,
and yet, so beautiful you can see
it can’t be random… or can it be?
The world is so random and it is not.
Mysterious, strange in every spot,
and yet, so beautiful you can see
it can’t be random… or can it be?
Alea Deum 
This was interesting, I have to make some posts on Randomness.
Hi Fran, attached is a copy on the conecpt of randomness from the blog (which is down at this time). Mybe it will interest you, judging from the koan on top 🙂
On the concept of random in Nature
What is random? Does randomness really exist and if so, what does it mean?
Difficult questions that have spawned controversy and narrow-mindness.
In this short essay i’ll try to explain the concept of randomness and study its occurence in natural processes.
It is true that for a long time thinking that anything can be truly random was almost a blasphemy. It took several ages
and many advances in science and technology and philosophy to formulate the concept of random events.
For example, while mathematics were advanced from antiquity, Probability Theory was only formulated about 100 years ago. The advances of Physics (particularly Quantum Theory) gave heavy blows to what i call Naive Determinism. However although determinism has received much criticism and scrutiny it still remains powerful especially in the social arena.
The motivation behind determinism is, in a sense theological, as are most of the determinist’s arguments. Even dialectical materialists have fallen to the trap of naive determinism.
Objections to truly random phenomena
The major objections to truly random phenomena are mainly of 2 kinds.
Every effect has a cause, hence there is no random
Random is meaningless and since everything has a meaning, there is no random
Both objections stem from the same principle which is essentially correct. Thus both arguments although superficially different are only one. The principle is theoretical and states that each cause leads to certain effects which thus obtain meaning. If something could happen without cause then it could not have any meaning attached to it.
A striking example is when somenone behaves unpredictably, we do not appraise her freedom of will to do so, but instead ask her to justify the reason for doing so (the common “Why did you do that?” question).
Now let’s study the notions of randomness used in everyday life and in science and their interpretation.
Notions of Randomness
Unknown Parameters
Known Parameters but Uncomputable
When randomness is met in natural phenomena it is interpreted (by the determinist) in 2 distinct, but essentially the same, ways. Either there are unkown parameters, which when known randomness will disappear. Or all parameters are known but are too complex to be taken into account, but it may be possible someday to compute them.
The hidden variables interpretation of Quantum Mechanics (which is much much controversial) is a typical example of the first kind. The complex interactions of Non-Linear Dynamical Systems and Chaos is a typical example of the second kind.
Now i will try to show that both kinds are just one and the same and that both just bypass the main issue. The known parameters but complex interactions in other words states that certain parts of the computation process (eg not enough digits, lack of resources, etc) are unknown. Thus mutatis mutandis states that something is unkown using different terminology. Thus leads to the 1st kind. Similarly we can go from the first notion to the second if time is taken as a resource for finding out the unknown parameters. Both notions bypass the issue by using an IF-THEN statement (if we knew, if we had more processing time, etc..).
The main issue is that some new, unpredictable information is generated and given to the system. Is it possible that both the determinist arguments are true and still true randomness can exist? The answer is YES.
One extremely general and extremely useful and powerful LAW comes to the rescue. It is the ARROW of TIME and the Second Law of Thermodynamics. Just think that if each effect is fully conditioned by it’s cause (without no variation or mutation whatsoever) then time could go back. Find this hard to understand? Just think of a smashed mirror. If the act of throwing the mirror to the ground could account for the whole effect of the smashed mirror, then it would have the SAME probability that the pieces could recollect (at some other instance) and come back together to the hand. This does not happen in this way. This means that some variation took place which is outside the scope of the cause (or causes) that led to it. In other words, of course there are causes and effects that are generated by them, but not fully conditioned. As such new information is generated which aquires meaning (as it is connected to the rest of the system, a-posteriori). So this synthesizes the naive determinist’s objection together with true randomness. The answer to determinism is to stop being naive and understand that everything has a uniqueness of its own and this is the meaning of true randomness. So to sum it up we have this:
RANDOMNESS = NEW INFORMATION = UNIQUENESS
Welcome Nikos!
Well, that was a hell of interesting comment you made. You seem to redefine Randomness as uniqueness as a solution which is a quite interesting philosophical one… Well, something to be expected coming from a Greek… right? 😉
Nonetheless, one might argue that the Universe is cyclic and nothing is unique since soon or later you will repeat your life. This concept goes from the most terrifying one given by Nietzsche where you repeat your life eternally with all its glories and miseries, to the more mathematical and less scary ones like Poincaré‘s.
Randomness, just like Infinity, are among these concepts we humans will always struggle with… I myself I don’t think I will see a globally satisfying end to this quest, but on the flip side that guarantees unlimited amounts of fun discussions! 🙂
Thanks,
it seems there is a problem entering the email in the comments (so you see a dummy email), but the site url is correct (the site has my email there).
The argument follows from the entropy concept with a philosophical tone, but i would not say an abstract tone.
Poincare’s theorem of return, does not say that the particle returns in exactly the same point (but very close in some sense). So this is compatible as is.
Nietzsche’s theory of eternal return is a continuation of the Myth of Eternal Return common to many people in previous times (since antiquity).
There is a reading of Nietzsche which redefines somewhat this thesis, in the sense, that IF a return is possible the person should ask for just the same life (a part of Nietzsche’s perspectivism), like affirming this life (and making it such that the same would be re-lived). it is very interesting.
In any case, randomness and time and concepts of probability are somewhat disconnected as is (although in practice this is not the case). So trying to produce a unifying view (leaving aside a dead mechanistic framework) can be helpful.
it is possible we will have to face such questions soon (see New Physics etc..)
Another interesting consequence of such a formulation (uniqueness, randomness) is that it implies a kind of “memory” (needed by uniqueness), and this can begin to tackle some (at least) epistemological points..
“If the act of throwing the mirror to the ground could account for the whole effect of the smashed mirror, then it would have the SAME probability that the pieces could recollect (at some other instance) and come back together to the hand. ”
I think I understood everything except this part. Could you further elaborate why this would be necessarily so?
Sorry for late reply, just saw your comment, by revisiting the (quite nice) site of Fran here.
Well one can use a number of ways to justify the part you are puzzled over.
For example:
1. By symmetry arguments, left and right hand sides are exactly the same, nothing actualy changed, so they should have the exact same probability either way.
2. Use (variations of) the theorems refered as “fluctuation theorems”, statistical in essense
3. By time reversal arguments (related but distinct from symmetry arguments), in that the time-reversed process is a one-to-one transformation so it should have the exact same measure
Hope these help,
Nikos
I think false claims in the published literature is much higher than 0.50. See Feinstein Science 1988, Begley and Ellis Nature 2011, etc. Researchers that publish or have to publish are actively sculpting their data sets and methods to get the publishing coin of the realm a p-value < 0.05. So, yes they are cheating, but they are reacting to incentives, of editors and funding agencies. They are skilled at their craft and smart enough not to make their data sets public.
On the other hand, industrial scientists mostly do not have to publish and mostly do not. But we can see the results of their experimental work all around us. Cheap food, marvelous products like smart phones and cars. Etc.