Forums → WEPR → Is Determinism Underrated?

So much hangs on getting an acceptable definition of the notion, that I'm not sure there is a standard approach. In other words, giving a right account of the notion may be precisely what's at issue here.

At any rate, you can read about some problems with the notion of determinism here. The trick is going to be giving a definition that is non-trivial but that also doesn't try to define an already-tricky concept in terms that are even more opaque (like defining determinism in terms of causation).

The definition given at the end of the section (under the header 2.5 Fixed) is a step in the right direction, although I think the 'or' statement means to refer to (a), rather than (b):

Determinism requires a world that (a) has a well-defined state or description, at any given time, and (b) laws of nature that are true at all places and times. If we have all these, then if (a) and (b) together logically entail the state of the world at all other times (or, at least, all times later than that given in (b)), the world is deterministic.

Fish, without any outside information other than what is in this thread, you do sound incredibly arrogant. Your general tone is 'this should be patently obvious, but I don't understand why everyone can't see this', which roughly translates to 'Why can't everyone be as smart as me'. Anyhow, putting this aside(I'm not here to discuss this), I'll mainly focus on science.

There is some reasonable basis for determinism, at least at the macro level. If I point a gun at an object and pull the trigger, anyone knows what the result is. Obviously, certain actions make other actions more likely to occur. The thinking then goes, if only I could measure better, etc, I could do this - a theory that was popular back in the 1900s. However, there is fundamental (quantum) uncertainty that can be made arbitrarily large for a given system.

Presumably, you accept quantum mechanics - at least the equations, if not classical interpretations. Part of this includes Heisenberg's uncertainty principle(true for every interpretation of quantum mechanics), which roughly states that (delta)(p*x) > (constant). In other words, it is physically impossible to measure a system beyond a certain degree of precision. This has unfortunate consequences for your definition of determinism - the fundamental subatomic scale of physical processes is inherently unmeasurable.

Now, classical mechanics works(very well) because quantum effects tend to cancel each other(very quickly) out to present the continuous, more predictable world that we observe. So you might think that at least at the macro level, this uncertainty can be essentially discounted. But it's not particularly difficult for me to create a situation where randomness governs macro actions. For example, if I make decisions based on a quantum "coin-flip" (e.g. if a particle's momentum is greater than some number, call that heads, and tails otherwise), I can govern large scale actions based on fundamental uncertainty. If you were to examine a system where I had two buttons(perhaps one of which would cause a very significant action to occur, e.g. nuclear Armageddon), and I informed you that the button I pressed would be determined by the quantum "coin-flip", you would not be able to predict anything of significance. I don't really see how your definition of determinism can really survive this, at least in any meaningful manner.

P.S.
When you respond to things, it's good etiquette to avoid one-line refutations, especially when the person you are talking to has grouped arguments in a meaningful manner. Also, while it's fun to throw one-liner every statement made, that just really amounts to a 'gish gallop', something that's generally not very helpful for a good argument. If you do choose to respond to this(something you are more than welcome to do), please keep arguments grouped by paragraph - it's written this way for a reason.

P.P.S.
This really isn't the place for this, but consider rethinking what is obvious. Infinity is generally considered to be extremely non-intuitive, leading to classic examples such as Hilbert's hotel and battles over sum: (1+2+3+...)=?=-1/12. While infinity is unbounded, unbounded systems fail to adequately account(at least in an intuitive, simple manner) for certain things(infinite infinities, one might intuit, would be larger than a single infinity; this is more or less untrue). The point is, there's often a lot more than what is immediately obvious, and many, many approaches have often been tried(hence why the concept is unintuitive). However, you really do think that you have a quality novel solution(not easy, but not impossible), run a google search for it, and then ask some technical experts in that field to make sure that it gives the right conclusions.

Fish, without any outside information other than what is in this thread, you do sound incredibly arrogant. Your general tone is 'this should be patently obvious, but I don't understand why everyone can't see this', which roughly translates to 'Why can't everyone be as smart as me'.

If you were to examine a system where I had two buttons(perhaps one of which would cause a very significant action to occur, e.g. nuclear Armageddon), and I informed you that the button I pressed would be determined by the quantum "coin-flip", you would not be able to predict anything of significance. I don't really see how your definition of determinism can really survive this, at least in any meaningful manner.

Also, while it's fun to throw one-liner every statement made, that just really amounts to a 'gish gallop', something that's generally not very helpful for a good argument.

Infinity is generally considered to be extremely non-intuitive, leading to classic examples such as Hilbert's hotel and battles over sum: (1+2+3+...)=?=-1/12. While infinity is unbounded, unbounded systems fail to adequately account(at least in an intuitive, simple manner) for certain things(infinite infinities, one might intuit, would be larger than a single infinity; this is more or less untrue).

The following will be focused mainly on determinism. The other remarks will be addressed on my profile.

To be completely honest, I'm not really sure how you are defining determinism. If you don't disagree with the possibility of an immeasurable quantum coin-flip(Laplace's demon is unable to measure this as long as he conforms to the laws of physics), then (pure) determinism really isn't very defensible, no matter what definition you use, since it's always possible to tie ever-larger processes to immeasurable coin-flips. From a retroactive point of view, it is nearly trivial to show how things occur - that is not very difficult.

Roughly speaking, this is what I interpret your determinism as:
1. Suppose a system(biological, physical, or otherwise) is given to us, and we are allowed to physically measure this system to the best of our abilities. We may use any measurements that are physically possible - e.g. we can examine neural states, atomic compositions, etc, but cannot make measurements that would not be possible in real life. No upper limit exists on the number of measurements made.
2. Upon making such measurements, we can then predict(given enough time and computational power) what the system will evolve to. We can assume that we have all physics models that adequately predict such phenomena.
3. Our prediction must be very precise, and must give only one answer. Nothing can be left to chance.

If the above is what you hold determinism to be, then heisenberg uncertainty destroys this reasoning. Uncertainty creates an immeasurable quantity, and from there, it's possible to expand the effects of the immeasurable until the system is essentially random(and presumably not deterministic).

To be completely honest, I'm not really sure how you are defining determinism. If you don't disagree with the possibility of an immeasurable quantum coin-flip(Laplace's demon is unable to measure this as long as he conforms to the laws of physics), then (pure) determinism really isn't very defensible, no matter what definition you use, since it's always possible to tie ever-larger processes to immeasurable coin-flips. From a retroactive point of view, it is nearly trivial to show how things occur - that is not very difficult.

Okay, now I really don't understand what you're talking about.

Quantum indeterminism is mostly independent of Heinsenberg uncertainty. There may in fact be a determinate outcome of a particle's wavefunction, but we can't measure it, regardless. Nobody disagrees with the uncertainty principle, as I've said many times - it's not open to reinterpretation.

You really need to define a set of criterion for your determinism:

As a determinist, I regard everything to be the necessary result of measureable and exact physical processes.

The determinism that I hold to is not dependent upon our ability to measure or predict anything

The two statements do not make any sense together. One says that determinism requires that measurable processes are part of determinism, the other says measurement is not required.

The two statements do not make any sense together. One says that determinism requires that measurable processes are part of determinism, the other says measurement is not required.

Unfortunately, this goes beyond "if I had better instrumentation/computation, I would be able to determine events". Instead, this is "no matter how good my instrumentation is, I will never be able to fully determine events". That's the point of the uncertainty principle. I've said this in the 3rd post, during points 1 and 2 - I've allowed all measurements that are physically possible. If a law(the uncertainty principle) explicitly prohibits any instrument from measuring certain quantities beyond certain precision, then a valid view must not rely on such precise measurements .

If you say that determinism is independent of what is physically possible, then I claim that purple unicorns make all events happen(I will mirror your logic), to show how absurd and worthless that line of thought is.

I say that unicorns determine all events, and are understandable to people with fairy dust. People without fairy dust cannot understandable this, and I don't have fairy dust, but you don't either, so there's no evidence that purple unicorns do not run the world, but I do have a strong belief in purple unicorns. It makes no sense to say that, based on the lack of fairy dust,, to say allmighty purple unicorns don't exist.

If that's the approach you take(if I could defy the laws of physics regarding measurements, this would be true), then anything is possible, and your definition of determinism is worth no more than purple unicorns.

If that's the approach you take(if I could defy the laws of physics regarding measurements, this would be true), then anything is possible, and your definition of determinism is worth no more than purple unicorns.

Unfortunately, this goes beyond "if I had better instrumentation/computation, I would be able to determine events". Instead, this is "no matter how good my instrumentation is, I will never be able to fully determine events". That's the point of the uncertainty principle.

In response to both, the main question is: If determinism exists(but can not be tested), why does it matter? It's as if I said: "I believe purple magical unicorns exist, but I can't ever test/prove/show this, and it won't affect the world in any way if they do exist". The statement I made cannot be disproven, so I've automatically won any argument. If that's the case, then yes, determinism is absolutely true. That's why there has to be some criterion by which determinism must be measured - e.g. predictions.

Fish, you misunderstand the uncertainty principle. One necessary assertion that your definition of determinism makes is that "all things are measurable". This is immediately and obviously untrue in light of the uncertainty principle. The uncertainty principle itself is NOT uncertain; it tells us that what we measure must necessarily be inexact. In other words, the uncertainty principle tells us that we can never acquire enough information to perfectly determine these underlying processes, and so the universe, for a lack of a better word, is uncertain, and therefore not deterministic.

Fish, you misunderstand the uncertainty principle. One necessary assertion that your definition of determinism makes is that "all things are measurable". This is immediately and obviously untrue in light of the uncertainty principle. The uncertainty principle itself is NOT uncertain; it tells us that what we measure must necessarily be inexact. In other words, the uncertainty principle tells us that we can never acquire enough information to perfectly determine these underlying processes, and so the universe, for a lack of a better word, is uncertain, and therefore not deterministic.

All events, regardless of any point in time or frame of reference, are set and incapable of being anything other than what they are.

Causation shouldn't really be a difficult concept. It's just the necessary conjunction of two or more factors or conditions.

In response to both, the main question is: If determinism exists(but can not be tested), why does it matter? It's as if I said: "I believe purple magical unicorns exist, but I can't ever test/prove/show this, and it won't affect the world in any way if they do exist". The statement I made cannot be disproven, so I've automatically won any argument. If that's the case, then yes, determinism is absolutely true. That's why there has to be some criterion by which determinism must be measured - e.g. predictions.

I already have considered a "weaker" version of determinism, one I have not completely discarded; I believe Fish called it probabilism or something like that (there was another associated term I don't remember). Basically determinism would be (almost) true except for the most basic processes; individually they are random, but together they form what appears to be determined processes. Except that unlike in true determinism, Laplace's demon would only be able to infer the future with a 99.999% probability, decreasing with projected time. The reason I am arguing for complete determinism here is that, as Fish then said, this implies random processes. And random processes are naturally less intuitive than defined processes.

2 I did not define determinism in any such way. You will not see the phrase "all things are measurable" or its analog in any of my previous posts. You will, however, find the following:

Well, I was going to comment about the "goalpost moving", but Hahiha beat me to it:

As a determinist, I regard everything to be the necessary result of measureable and exact physical processes

There are processes which determine events, and they are measureable.

You will not see the phrase "all things are measurable" or its analog in any of my previous posts.

More or less, I tend to agree with HahiHa - 'probabilism' is much more reasonable than determinism. Probabilism is absolutely correct - after all, flying jet planes don't randomly drop out of the sky. Furthermore, I argue that probabilism is very intuitive, while determinism is less so, but the two concepts are easily mixed. Many systems are modeled as a chain of transitional probabilities between states, which is by far a more nuanced model than the 1 state, 1 result that determinism advocates.

The only issue with probabilism is the certainty with which it infers the future - I can easily rest the fate of the world on a quantum coinflip(my original example), and in doing so, can cause the certainty to dramatically decrease. Typically, unless we try to actively defeat probabilism, this won't happen, but we can deliberately make systems essentially impossible to predict.

So essentially, my position is: The physical world is very strongly shaped by certain processes. These processes are often incredibly strong and reliable, and are often essentially inevitable. But, plenty of randomness exists - this generally cancels out, but it is possible, with effort, to make a systematically random scenario. To put another way, we are subject to the laws of physics, but the laws of physics do not dictate who we are and what we do.

Maybe I'm blundering into a wording issue, but you did state at least twice that you think the processes leading to events are measurable.

A: the processes leading to events are measurable
B: all things are measurable
A ≠ B

I already have considered a "weaker" version of determinism, one I have not completely discarded; I believe Fish called it probabilism or something like that (there was another associated term I don't remember). Basically determinism would be (almost) true except for the most basic processes; individually they are random, but together they form what appears to be determined processes.

The only issue with probabilism is the certainty with which it infers the future - I can easily rest the fate of the world on a quantum coinflip(my original example), and in doing so, can cause the certainty to dramatically decrease. Typically, unless we try to actively defeat probabilism, this won't happen, but we can deliberately make systems essentially impossible to predict.