The World's Largest Coalition of Nontheists and Nontheist Communities!

Randomness - sand in the Vaseline?

Every single event we cannot accurately predict represents one of two conditions (or a mixture of the two):

1. a lack of information or ability to process that information on our part

2. true randomness

The only way to empirically prove that an outcome is predictable is to accurately and repeatedly predict it. Therefore, the only empirical proof of absolute determinism would be to demonstrate a flawless method of universal prediction - a precise theory of everything - that approaches perfect reliability across an extremely wide range of ongoing systems of causation.

Strong induction and the progressive track record of math and science would appear, according to Occam's Razor, to make determinism a highly likely scenario. However, for this to remain true on an absolute level (not even as proof but as strong likelihood) we must first assume what may very well turn out to be a false dichotomy. Many strict determinists present an either/or picture of the universe that says either determinism is always true or it is not true at all. Since we can see it in action in so many ways, it must, therefore, always be true. While this represents strong induction, it is not empiricism.

For one thing - we can never empirically witness an effect without a cause because we will never be able to recreate the conditions that caused it - since it was not caused. Therefore, any effect with an unknown cause, could represent an effect without a cause. The minute we can precisely reproduce an effect - then we know it can be caused. Until then, we can only assume it must have a cause. Therefore, a causeless effect is not falsifiable. The burden of proof is on the affirmative claim - which, in this case, would be the claim that a particular event had a cause. The singularity that 'preceded' the Big Bang is a prime example of an event or condition that must be proved to have a cause, or we must assume that it did not have a cause, since the burden of proof is on the claim that it did have a cause.

String theory suggests that eleven dimensions may be necessary to contain the 'causal gestalt' that would describe a version of an absolutely predictable universe. Dark matter and energy might help explain why the universe appears to be expanding the way that it is - yet we can't put our finger on that yet either. We still don't know why matter has mass. Why is gravity an asymmetrical force? Why does time appear to be a one way dimension? How did so much matter survive and so little antimatter? What about the horizon problem? There remains a great deal of mystery in the universe. Some of it may be unsolvable.

So, at this stage of the game, absolute determinism is not empirically proved or provable. And, if there is the possibility of completely unpredictable events - then there is the likelihood that those unpredictable events entered the causal gestalt - because they are likely to have deterministic effects which, in turn, become causes.

Since the causal gestalt is intertwined throughout everything (the iron in my hemoglobin that carries the oxygen to my brain so I can think originated in the furnace of a supernovae many light years away, for example), any degree of randomness in the causal gestalt represents some degree of randomness in the collective components of that gestalt that is me. This includes any random aspect of environmental stimuli that my consciousness encounters and is shaped by as a result of that encounter.

Thus, we can talk about relative determinism, or virtual determinism, or practical determinism, or even deterministic probability and be reasonably certain we can rely on it. Perhaps we are compelled to. But, as long as there is any possibility that some things happen for unpredictable reasons, human beings are likely to behave, to some degree, unpredictably as well.

If this unpredictability is actual, then it may very well be that our emerging consciousness becomes, through evolution, better and better at adapting to some degree of utter unpredictability even as we become better and and better at predicting outcomes. This adaptation could be seen as true creativity - an actual ability to extemporaneously adapt to completely unforeseeable events.

So, we are left with little reason to abandon the possibility that our words and actions might be just a little bit more than the inevitable toppling of dominoes into each other. So far, the argument against adopting an existential perception of actual agency is not airtight.

0 members like this

Replies to This Discussion

Permalink Reply by Atheist Exile on February 21, 2010 at 11:06pm

Hi Howard,

I've been checking replies via email and didn't see this new topic of yours until just now. Sorry about that!

"Thus, we can talk about relative determinism, or virtual determinism, or practical determinism, or even deterministic probability and be reasonably certain we can rely on it. Perhaps we are compelled to. But, as long as there is any possibility that some things happen for unpredictable reasons, human beings are likely to behave, to some degree, unpredictably as well."

I thought that paragraph summed up your position pretty well. You didn't mention the quantum realm, which, it seems to me, is a fundamental part of the universe: it even occupies "empty" space (quantum fluctuations). As a bizarre, unpredictable, random realm, it pretty much spits in the eye of causality and determinism. So, causality is NOT absolute . . . but for all practical intents and purposes, it might as well be.

I suppose the universe could host random, uncaused, events but I don't know that science has documented such occurrences outside the quantum realm. It seems to me that chaos could hide randomness within its complexity and we wouldn't know it if we saw it. If I understand you correctly, randomness (uncaused events) might happen, but so rarely that it escapes our detection. If that's the case, then such randomness probably makes very little difference to our experience of reality. I'm reminded of the "black cat glitch" in the Matrix movie. If it's so rare, we might not even notice it and probably can't study it -- except anecdotally.

Truly absolute determinism -- the kind that leads people to abandon the empirical evidence of their experience -- really does seem line an article of faith. I don't know . . . perhaps, as a law of nature, causality really should not admit exceptions.

Until very recently, I emphasized the "animate mode of response" to causality. Many interpreted this as "different rules" for causality that weren't supported by the law of causality itself. But my latest (and last?) defense of free will doesn't need the animate mode of response argument. Neither does it need an exception to absolute causality/determinism. Every effect becomes a cause when it affects something else. So the brain becomes the cause of effect known as imagination. Prescient imagination is what we use to extrapolate cause and effect into the future and thereby inform our decisions in the present. This mental feedback mechanism gives us the temporal advantage over causality we need carve our own paths into the future.

Permalink Reply by Howard S. Dunn on February 21, 2010 at 11:53pm

It seems to me that chaos could hide randomness within its complexity and we wouldn't know it if we saw it.

I would suggest this is more than likely. Let's check out the three types of 'dominoes' that don't fit determinism. But first, let's define the one type of domino that does:

In the causal gestalt (as I am fond of calling the uber-complex, non-linear field of cause and effect) let us simplify in order to comprehend and imagine it as multiple, interacting chains of 'dominoes' falling into each other. When a domino falls into another it causes that domino to fall, in turn. So, at that moment, the falling domino is a cause and the domino it falls into is the effect. However, the effect becomes a cause as soon as it strikes the next domino and so on. Therefore, in a purely deterministic universe, all dominoes are both causes and effects. That leaves us with the possibility of the three other types I mentioned:

1. The domino that doesn't ever fall - it is unaffected and causes nothing to happen. If this domino were to 'exist' we could not detect it. It would look exactly like nothing. And, as we know, that is precisely what the majority of the universe appears to be composed of - nothing.

Now, we are beginning to see that 'nothing' may not exist - it may be filled with dark matter. But, for now, vacuum, darkness, cold, emptiness, ... whatever you want to call it, appears to be everywhere - and pretty important at that. Without the nothingness, the everything would collapse in on itself.

2. The domino that falls without having been struck by another domino - if this domino exists, empiricism will never find it. Empiricism requires the development of an experiment that can repeat the scenario. Since, they would be no cause of this scenario, there would be no experiment that could recreate the conditions that would cause it to occur again. So, the existence of this domino is unfalsifiable, and, empirically speaking, cannot be proved to ever occur.

However, the existence of the singularity before the big bang hints at it. Aspects of quantum physics hint at it. The very complexity of the causal gestalt - chaos - offers plenty of room for it to happen all the time. It might be observed, by chance, occurring all the time - but our inability to cause it to recur would force us to deny its existence. Nevertheless, absence of evidence is not evidence of absence. Determinism is based on the falsifiability of evidence and strong induction based on multitudinous recurrences of observed events both as causes and effects. Induction, therefore, would not account for this domino even it it did exist.

3. The domino whose fall is caused but causes no further dominoes to fall - there is one very likely version of this domino that happens all the time. The thoughts and dreams of any human being, (or animal - for that matter) to some varying degree, affect the behavior of said being and are, therefore, common deterministic dominoes. They are, that is, with the exception of those thoughts and dreams that occur in the mind of said being just before they die - and are not, in anyway, acted upon.

In any case, these dominoes cannot be detected - by definition. Any detection of these dominoes would require an effect.

So, while it seems like nonsense to think that these three described dominoes exist - there is some room to speculate that they may and no empirical way to detect them.

Thus, I have deemed these the three 'nonsense dominoes.' And all three may exist right before our eyes as nuts and bolts, nothing supernatural - simply undetectable - parts of reality. I'm not talking about ghosts or gods. But empiricism has already been proved to be limited. For example, there are edges of the universe we will never see as they are too far away for their light ever to reach us. And we cannot see even the closest star in its current condition. We can only see what it looked like four years ago.

So - the 'nonsense dominoes' may not exist. In that case, I am write of them here through simple compulsion based on the 'dominoes of error' (another discussion). We will never know for sure. The uncertainty principle extends beyond the quantum realm.

Permalink Reply by Atheist Exile on February 22, 2010 at 6:43am

Hey Howard,

Well written. I understand your point better now.

"The domino that doesn't ever fall", according to some, must be before the Big Bang (if at all) because (according to them) EVERYTHING afterward is cascaded from this first cause (Prime Mover). Anything that happened before the Big Bang must be (logically) of the quantum realm. I'm a little murky on this. If space was created by the Big Bang, could there be "empty space" before the Big Bang? If so, was it "filled" with quantum fluctuations, just as the empty space in our universe is? By the way, just about everything is filled with empty space, including you and me . . . and quantum fluctuations occur (allegedly) within those empty spaces. So the empty space we're talking about is at the quantum level.

But I'm not sure I can subscribe to the notion some effects do NOT trace back to the original one (Big Bang). Can causes arise spontaneously, without a prior effect? I can't think of an example of "the domino that falls without having been struck by another domino" . . . but you acknowledge this, so we're not in disagreement. For a while, I was toying with the notion that the brain could be the source of new causal chain reactions but I was later sorry I posted that notion. Luckily, nobody commented on it. :-)

Can an event be called a cause if it has no effect? Your 3rd example: "the domino whose fall is caused but causes no further dominoes to fall" appears to beg this question. If a comatose person can have thoughts or dreams, then perhaps he/she would fall into this category as well as a dying person. Or what if the comatose person can hear or see things in his/her immediate environment but can't produce any coherent thoughts about them?

None of these 3 potentialities seem impossible but they don't seem very likely either: not when you consider that you'd have to break the symmetry of cause and effect. Of the 3, I think #2 ("the domino that falls without having been struck by another domino") is the only one that would not break causal symmetry: we'd simply have multiple or variable sources of causal cascades.

P.S.
On second thought the other 2 seem okay too . . . what law says causal chain reactions can't have a beginning or ending?

Permalink Reply by Howard S. Dunn on February 22, 2010 at 12:02pm

On second thought the other 2 seem okay too . . . what law says causal chain reactions can't have a beginning or ending?

Well, beginnings (2.) and endings (3.) seem to violate determinism - especially beginnings. I would argue that endings could also screw with ultimate predictability - unless you could predict the ending and discount it. However, there is an odd thing about endings: effectively, any previous cause of an ending that didn't cause more than the ending might as well be an ending too - see what I mean? There is a kind of reverse 'decay' that backs out of an ending until you get back to a cause that had other repercussions that did, in fact, continue as causes in their own right.

Take your example of the thoughts and dreams of a comatose person - say one that was in a coma of ten years. There would be a decade of dominoes falling in that person's mind that, if they awoke, would be perfectly valid (deterministically.) Even if there was amnesia, on a subconscious level, those ten years of 'inner dominoes' would probably remain valid. On the other hand, if the person died, all ten years of dominoes would be part of a non-deterministic end-chain.

So, while whether they lived or died might, ultimately, be determined - the deterministic value of those ten years of brain activity would be entirely predicated on whether they ever came out of the coma.

Permalink Reply by Atheist Exile on February 22, 2010 at 8:25am

Two questions for you, Howard,

Who is that guy in your avatar image?

How long have you been discussing free will/determinism/compatibilism?

Permalink Reply by Howard S. Dunn on February 22, 2010 at 11:37am

1. Julien Offray de La Mettrie I like to change it up with historical figures over time to demonstrate that atheism is an old idea and scientists don't hold a monopoly on it.

2. I have no 'formal' education in philosophy and cannot quote a single historic philosopher without Google. (Well. this is not entirely true - especially if you count SF writers like Huxley and Asimov. And, I have read Goethe, Hesse, Voltaire, Camus, Locke, Plato, etc. Just not so I can quote them off the top of my head.)

However, I have been jabbering about this stuff with very intelligent and educated people for over thirty years. I also have a programming background which offers a wonderful analogy on determinism and heuristics - self-programming and simulation systems - and why they aren't consciousness.)

Permalink Reply by Gary Burk on February 22, 2010 at 8:01am

Good stuff Howard!

Doesn't Free Will require randomness? Uncertainty makes free will possible within the boundries of an orderly cosmos. It suggests that although individual events may be random, their aggregate is deterministic.
So, any atom of U235 has a 50% chance of fissioning in 7 million years. We can say with confidence that 7 million years from now one half of the U235 will be gone, but can never predict just which atoms will survive and which come apart. This wonderful "cop-out" allows both randomness and determinancy to co-exist comfortably. (At the very least, it is a plausible explaination of hwo things work). And - it seems to work very well in predicting numerous effects (i.e., it's been well tested)!

Regards,
GaryB

Permalink Reply by Howard S. Dunn on February 22, 2010 at 11:50am

Randomness would fall under 2 and, possibly, 1. And, yes, some version of randomness (true random is what I call nonsense - but in the sense that this type of nonsense may actually be abundant).

In computer programming, while most simulations call for probabilities of events and, therefore, a version of randomness - computers can only so far) generate pseudo-random numbers. These number strings must pass a test that asserts their difficulty in 'reverse engineering' and a couple of other factors. Nevertheless, they aren't actually random as they are generated by mathematical algorithms and, if you know the algorithm and some other factor - usually the time of 'generation' - you will know the entire string and the particular number in that string that was used. Pseudo-randomness is entirely deterministic.

I loved Asimov's approach to the aggregate determinism you speak of in his Foundation series. He extrapolates from real math applications such as insurance actuarial tables to invent a branch of mathematics called psycho-history. This math cannot predict the actions of an individual (a nod to free will) but can accurately predict the future history of very large groups of people. However, it remains a high degree of accuracy and not perfection and, when an individual turns up who has an inordinate amount of power over a large group - the accuracy goes out the window.

In your example, yes, we can predict the half-life of U235 in general - but, of course, there is a different calculation if you use it to create an atomic explosion.

Permalink Reply by Gary Burk on February 22, 2010 at 1:19pm

I think that the normal decay half-life is not a result of stray neutrons disrupting the nucleus, as it is during a chain reaction (an explosion if it happens quickly or a nuclear reactor if its rate is controlled by absorbing some of the neutrons). As of now, there is no generally accepted theory or even promising idea as to why one U235 atom decays while its neighbor remains around for a few million more years. If we look at U238, that half-life extends to about 7.1 Billion years.

A basic premise of quantum physics is that the statistics are predicatable, although the individual events seem to be random (actually, QM assumes that they really are random).
Now that you've mentioned it, I do recall that psycho-history was Harry Seldon's realization that large groups of people would be predictable in the Foundation trilogy.

That may actually be correct but I think will remain unprovable, since the meaning of "large"
in QM is number in the 10^12 and upwards range, far greater than any population we seem likely to reach.

I don't know where an electron in an electric field will go, in fact it may well go in the "wrong" direction. But I do know where an electric current in that field will go, to an arbitrarily great degree of certainty. (Tunnel diodes are common electronics components that count on this phenomenon, atomic force microscopes are instruments that also depend on this so called "tunneling" where charged particles appear on the other side of barriers that they should not have been able to cross).

Regards,
GaryB

Anyhow, the point I was trying to make is that we know that seemingly random events occur at the sub-atomic level, and I surmise that this is adequate to explain in-determinant thoughts. (The notorious "Butterfly principal" is an example of such non-linear processes).

Permalink Reply by Howard S. Dunn on February 22, 2010 at 2:20pm

Anyhow, the point I was trying to make is that we know that seemingly random events occur at the sub-atomic level, and I surmise that this is adequate to explain in-determinant thoughts. (The notorious "Butterfly principal" is an example of such non-linear processes).

I think this is a reasonable stance. Many tend to take on an all or nothing in this discussion - suggesting that, if this randomness disrupts determinism at all - it does it absolutely and, therefore, must not really be valid - since we know it doesn't disrupt it absolutely. I'm not sure why randomness can't be part of the mix.

For example, in a computer simulation, if the random numbers could actually be random, this would not preclude the likelihood that the majority of the code was algorithmic.

Permalink Reply by Gary Burk on February 22, 2010 at 2:49pm

As an aside, there are ways to produce truely random numbers in as simulation. It can be done by capturing the "seed" values from some uncontrolled and unpredictable process. For example, perhaps I need a group of n random numbers. I could go to the New York Times online news, take the first letter of the first word, the second letter of the second word, the third letter of the third word, and so on. When I get to larger numbers, perhaps the 20th random valu it is unlikely that the 20th word would have 0 letters, so I just keep counting through to the 21st word, the 22nd , . . . , untill I count the 20th value. For the 21st random value, I start at that point.
Clearly, this approach can be arbitrarily complicated, for example the 1st letter of the first word of the first article, the 2nd letter of the 3rd word of the 2nd article, the 3 letter of the 5th word of the 3rd . . . I can then use the ASCII code for those random letters with what ever algorithm I wish to generate any range of random values. This type of approach is used to verify results obtained by using easier and more common pseudo-random number generators.

This has nothing to do with Free Will, it's just an interesting approach to a task.

Regards,
Gary

Permalink Reply by Howard S. Dunn on February 22, 2010 at 4:14pm

Cool. Nevertheless - it remains a deterministic process - maybe ... certainly the algorithm of which letters to choose from which paper and the ascii values. The question DOES relate to this discussion in a way. If there is anything actually random about the numbers generated by this method, it would be because the words chosen by the writer/editorial process upon publication were NOT entirely deterministic - which would suggest some element of free will.