Science Forums

# Why Determinism could never be disproven or shown to be unlikely

## Recommended Posts

You're still missing the point, in both your latest posts.

1) Claiming that "nature must decide" is assuming a cause and that implies determinism.

I think determinism is usually taken to mean a lack of free will, not a lack of causality.
I really don't think he had meant what you imply. To the contrary determinism is a stretch of causality. Anyway, lack of free will is not the core meaning of determinism, it has been proposed as its corollary but IMHO it's a semantic issue.

Again, I think you're using terms--morals and ethics--in a way that is not the way in which they typically are used. If behaviors are not freely chosen, if they are completely determined, then they do not possess moral or ethical dimensions.
Drawing absurd conclusions about no guilt, no laws and no prisons is a pseudo-argument against determinism, simply because these things, and morals and ethics, are a part of the behaviour of what is being discussed. Remove these things and you are removing part of the causes, which themselves have their own causes.... Laws, ethics etc. don't cause crimes to be committed and neither do they totally prevent them totally, but they have their part in determining people's actions. Therefore they certainly can be part of a determinist viewpoint.

• Replies 85
• Created

#### Popular Days

The Algorithmic Definition of Probability states that a random number sequence is defined as an ordered set where the shortest possible algorithm that can derive that number is longer than the the set itself. That is, its not compressible: you need to have the sequence in order to define it.

...

Decaying alpha particle,

Buffy

Practical determinism? - a straw man

I understand exactly what you are saying, but that is not non deterministic randomness. It might outline a situation where we should consider something random and nothing more for practical reasons. Your definition fits the rolling of a dice perfectly. I roll a dice 20 times in a row and create a sequence of numbers 1 - 6. If I wanted to create that same sequence of numbers using an algorithm, I would have to have some function that took into account a persons inclination to put the dice at a certain point in his hand to start with and an inclination to shake the dice a certain way, and then all the physical factors etc etc. That is much more complex than a sequence of 20 numbers. But the result of rolling the dice is still deterministic.

It doesn't matter if you could ever know or not, although it is easy to say that it is impossible to know something when certain changes in your ability to interact with an object or advances in data storage and processing might allow you to know things you thought you couldn't before.

Common Determinsm

It is more like a question of whether or not there is something to know. Or maybe could god know it or have knowledge of some underlying system that would tell him about it - Where god might be defined as the most knowledgable and most physically capable investigator ever, or maybe just someone that knows everything without having to look.

Saying that random equates to the fact that small changes in input variables cause large changes in the output is not "like" anything but exactly what it is. What it means is that you can see for example if you roll it a different way or weigh one of the sides down etc you can see how it effects the outcome and that the outcome is not cause by some "magic randomness" Your argument seems to be nothing more than saying that coming up with a model describing random behavior would be difficult and complex. This is not evidence against determinism, at least not the kind of determinism people defend.

Your "Practical Determinism" disproven in even classical physics

How is such a practical difficulty any different than any non random occurence? If I play baseball I don't calculate the mass of the baseball and its initial velocity etc etc to figure out exactly how to throw it. The classical physics algorithms are more complicated (to us at least) than the result of the ball flying through the air. This has never been a disproof of determinism.

Determinism never was determinism because it was knowable in the most practical sense.

And on a side note, stating that something will never be knowable is also a violation of the limits of induction. In otherwords, you can't know that it will never be knowable - you think so now, but there could always be additional information that you could find that would both show why you thought that and why it was wrong.

Common determinism is obvious and necessary

Buffy, even without being able to come up with an algorithm to predict the roll of a dice we can see the outcome happens for reasons. Sometimes we can even change the reasons and change the outcome with a certain probability. We can see that things affect stock market prices without being able to use them to predict stock market prices. It is not scientific because it is blatantly obvious, and science knows better than to comment on things that average people are most likely to review themselves.

And yes, the concept of determinism, the real determinism that is immune to your objections here, is very important to philosophy. Perhaps as a non deistic religion if you wish to call it that - it is needed to found understanding about things like morality and ethics and life improvement etc etc.

##### Share on other sites

I understand exactly what you are saying, but that is not non deterministic randomness. Your definition fits the rolling of a dice perfectly.
It doesn't sound like you did understand what I was saying, because my argument was quite the opposite from "rolling...dice perfectly." Indeed it betrays a lack of understanding of the distiction between probability and randomness. We can develop probabilistic curves based on data sets that can *on average* easily model real world *random* processes. But my point about randomness does indeed incorporate all those possible factors that might alter a *specific* throw of the dice, those factors being based on other prior factors which are based on other factors prior to those ad nauseum ad infinitum. Thus its not difficult to show that any attemt to model all those factors would require incorporating the entire history of the universe in order to predict it! There's your information problem!
It doesn't matter if you could ever know or not, although it is easy to say that it is impossible to know something when certain changes in your ability to interact with an object or advances in data storage and processing might allow you to know things you thought you couldn't before.
Consider, in order to have a computer that was capable of storing the information state of everything that has ever changed in the history of the universe, it would, by definition need to be bigger than the universe itself by dozens of orders of magnitude! Now, would you like to argue that you can build a computer that's larger than the universe? Would you like to argue that this is still really in the realm of physics?
It is more like a question of whether or not there is something to know. Or maybe could god know it or have knowledge of some underlying system that would tell him about it - Where god might be defined as the most knowledgable and most physically capable investigator ever, or maybe just someone that knows everything without having to look.
...and that, dear, is what we call metaphysics. Thinking there ought to be "something to know" is a belief, not a theory backed up by data.
Saying that random equates to the fact that small changes in input variables cause large changes in the output is not "like" anything but exactly what it is.
Again, you're completely glossing over a very important distinction: divergent series is a realm of mathematics that is by no means intrinsically linked to randomness. It can indeed use random inputs, and that's interesting: it shows how a sequence of random events becomes dominant--resulting in dramatically chaotic behavior--or is drowned out by the functions that govern the system--hiding the underlying random nature of the constituent events. But to call it the definition of randomness simply betrays a lack of understanding of the subject.

To equate this as being the definition of randomness is useful to your argument of course, because it covers up the major flaw in your argument that depends upon "nothing being random." You're simply denying randomness. I understand why, I just think its important to point it out! :warped:

Your argument seems to be nothing more than saying that coming up with a model describing random behavior would be difficult and complex.
See above: you're missing the argument completely. Its not that I'm whining that "its too hard!" Its that it can be shown that the required computational and data storage requirements to represent it would exceed the capacity of the universe. And that makes it "impossible", not just "really hard."
Determinism never was determinism because it was knowable in the most practical sense.
Which kinda proves my point! It makes showing determinism a completely metaphysical endevour! :shrug:
And on a side note, stating that something will never be knowable is also a violation of the limits of induction.
Tell that to Kurt Goedel!
Buffy, even without being able to come up with an algorithm to predict the roll of a dice we can see the outcome happens for reasons.
No, the outcome happens. A series of outcomes with a function space (those dice only allow a limited number of outcomes that must repeat), will create probabilistic models that might give one the illusion that there's a simple set of rules, but randomness does exist.

The History Channel was running a show last week about a bunch of kids from MIT who set out to "beat Vegas" by getting really good at card counting (I actually know one of the people who worked in this ring). You can count the cards perfectly, but you still never know what the next card is going to be, and these folks had to learn to withstand *huge* losing streaks, no matter how much they "knew" about the future. Yes you can argue that the biggest aspect was the magnification of the effect of the dealer drinking a chocolate milkshake three years ago, but that isn't *all* of why the next card is a deuce: these events cascade all the way back to the Big Bang, no matter how you slice it. And this is before we even start talking about the fact that its unclear that even *if* you had all that data, that quantum randomness really is linked to it at all, which I won't bother with since its not germane to the purpose of this thread.

And yes, the concept of determinism, the real determinism that is immune to your objections here, is very important to philosophy. Perhaps as a non deistic religion if you wish to call it that - it is needed to found understanding about things like morality and ethics and life improvement etc etc.
That's an opinion, and a religious one at that. morality and ethics can be shown to be a *necessary* evolutionary result in social groups: Mr. Dawkins spends a good deal of time on the topic in his latest book that withstands formal argument even if you hate his personal opinions about God.

I don't mind you having this opinion, but realize that if its the last best one you present in your argument, that there's not a whole lot supporting it.

Necessary and sufficient,

Buffy

##### Share on other sites

The ever-growing less than infinite model

Perhaps the universe is infinite. Noone ever cared to build a model that considered all factors of the universe. Rather we build models to consider factors that effect the outcome of single events we want to predict, within certain ranges of precision. In doing this we consider some factors negligable when they are considering our desired precision. Sometimes when information and processing power is limited we aggragate the effects of various factors into a random variable that is a result of those various factors and then use the random variable as a factor in something else. As our abilities grow, we can break down the same random variables into any combination of more negligable factors, clear input factors, and perhaps other random variables depending on what effects it and how complicated those input variables are. And in doing this, our model becomes better, perhaps allowing us to predict with less variance than before.

I can build a model for investing in stock market options that correlates the volatility of stock prices during various kinds of world events so that I can develop a breakout strategy to be used only when serious stuff goes down. Thats because I am a determinist and can clearly see that "random" events are a result of other events IE they are deterministic. Yes I know you have agreed this type of reasoning is possible, but I wonder if you realize that it would not be possible if the events were not deterministic - at least when you use our definition.

No determinism, no correlations

If the prices are non deterministic and just jumping around because they feel like it (oh wait, there is no because since non determism = no because... so you can't say things like that) there is by definition no rhyme or reason to their actions. For that matter correlations indicate having the same cause or that something is caused by something else. So there should never be any signifigant corrleation between any aspect of a non-determistic random variable and anything else. Of course when dealing with anything like dice or the stock market this is disproven to the average person every day since they can do things like smack a flipping coin to change its outcome, or watch how the stock market goes crazy when a terrorist attack occurs.

I hope this shows how nonsensical non determinism is when you use the common definition.

Leaving room for growth

It is not metaphysics to talk about "whether there is something to know" or "whether or not the most knowledgable being could know it". For any given random factor, with a growth in processing ability in the future we could break it down into other factors. So THAT factor at least is more knowable by us than our current knowledge of it. Maybe we could never know ALL of the factors effecting something, but we could break down all of the factors that produce more variance in the outcome than what we can just ignore. Therefore ability to know is relevant because it shows that we can always break down any random factor into other factors and doing so might allow us to "complete" our model in the practical sense.

It's relevance also has to do with the limits of induction or skepticism. There is always room for more information to come along and destroy your belief that it is impossible to know something or impossible for our processing power to grow beyond a certain point.

It is also relevant because it explains how we can still correlate factors to parameters of random variables, or do things like put a dice in a mechanical hand that always rolls it exactly the same way and watch how it always comes out the same. You may not have to be a determinist to be able to understand these things, but you should be.

Who is talking about divergent series?

Saying that random equates to the fact that small changes in input variables cause large changes in the output does not need any further abstraction or metaphors relating it to other disciplines. It is a very simple, very true statement. I should clarify however that "large change" means effect on the output, ie in the case of rolling dice it does not cause you to roll a 56 on a 6 sided die but rather it might cause the dice to land on a different side which causes it to fumble around on the ground differently which causes it to land on perhaps the complete opposite side or perhaps not. If you were to somehow map all potential outcomes of the dice including how that outcome occured (got a 6 by boucing on this corner and hitting the chair leg) THEN you could see that any small change at the beginning would result in a drastically different outcome.

Classical motion models exist in the face of your objection, so does every other model

As far as anyone can see the universe is infinite. Furthermore, there might be more than one of them. Furthermore, noone said that you need to calculate everything in the universe just to come up with a deterministic model for one thing, random or not. Determinism has never been like this. Do I need to know about the behavior of the atoms in my body to calculate where a baseball i throw will land? No, cause it doesn't matter under normal circumstances. So some things have no or negligable effect and can be ignored. Someone might peg me in the face with a baseball before I throw it, and that would certainly have an effect on the outcome. But that typically doesn't happen so I don't base my model on that either. A gust of wind could cause the ball to land slightly removed from my calculation in any direction. But a baseball player who is roughly in the calculated spot is close enough even with this variance to just move his hand to wherever it goes.

Determinism always existed in the face of all the things you mention.

In conclusion your objections are at best a refusal to use the same definition for determinism that the people who defend it are using. Your opinions on any related subject are therefore meaningless...

##### Share on other sites

Kriminal99: I suggest you read up about NP-completeness, the theoretical limit on satisfiability is 558 (Peter Suber, "Three Levels of Truth"), this places a severe constraint on potential knowledge. Also, it appears that there are no methodologies for solving NP-complete problems, this implies that there is no deterministic process underlying systems that are NP-complete.

##### Share on other sites

Kriminal99: I suggest you read up about NP-completeness, the theoretical limit on satisfiability is 558 (Peter Suber, "Three Levels of Truth"), this places a severe constraint on potential knowledge. Also, it appears that there are no methodologies for solving NP-complete problems, this implies that there is no deterministic process underlying systems that are NP-complete.

Speaking of infinitely large problems, how much time could I spend reading if every time someone who disagreed with me in an argument said "if you read X you would know you were wrong"... a whole lot, and a whole lot of those books I read would at best allow me to better explain to that person why they were wrong. Or another thing to ask about that is if it gave you an understanding of why I was wrong, then why do I have to go read it instead of you just converying the reasoning. I am not going to read that rediculous spaghetti reasoning anymore than I am going to conver to each different religion and live as a monk for 5 years in solitude just to have a more accurate viewpoint of the usefulness of religion.

Sometimes when people do this I actually take a quick peek at whatever it is. This guy has an argument in his paper that 1+2=3 isn't really that true because 3+2=1 is false and 13+432=9 is also false. If you want to eek a useful argument out of that muck feel free, I am not going to do it for you.

And btw, to your earlier question the universe was always supposed to be an infinite chain. The big bang is not an ultimate beginning of the universe unless it can explain itself such that noone could ask why did the big bang occur? Why did whatever caused it occur? Why did whatever caused that occur?

##### Share on other sites

And btw, to your earlier question the universe is always supposed to be an infinite chain. The big bang is not an ultimate beginning of the universe unless it can explain itself such that noone could ask why did the big bang occur? Why did whatever caused it occur? Why did whatever caused that occur?

"Always supposed"...by whom? Anyway, this seems a bit off topic IMHO.

##### Share on other sites

"Always supposed"...by whom? Anyway, this seems a bit off topic IMHO.

It was a question asked by philosophers since ancient times - an infinite regress was the result. The best answer anyone could ever come to is that no single cause of the universe like big bang could be discovered in the future, because one could simply ask why did this cause happen, and then why to the next answer and so on. Other people then responded that Chuck Norris is the cause of the universe, and that one cannot ask why Chuck Norris beecause Chuck Norris' existence explains itself. Er... maybe it was god. Of course if you don't believe in god then god does not explain itself nor do you think he is the cause of the universe. Rather there is just the infinite regress.

It is not really that off topic come to think of it, since it again shows that determinism never had anything to do with an infinitely large model of an infinitely large universe, but rather with the fact that there were causes for things and that one could use these relationships to build models to predict outcomes within certain ranges of precision.

##### Share on other sites

...it again shows that determinism never had anything to do with an infinitely large model of an infinitely large universe, but rather with the fact that there were causes for things and that one could use these relationships to build models to predict outcomes within certain ranges of precision.
Whoa! Out of the blue "certain ranges of precision" pops up! You really need to study random numbers and probability some more. Its clear that what I've said so far goes completely over your head.

The whole point I've been trying to make is that convergent and divergent functions *combine* with randomly generated events to *magnify*--not *create*--what appears to be random. You are defining random solely as "apparent" randomness in the aggregate, which is aided and abetted by divergent functions, but that is wholly missing the fundamental concept of randomness.

By bringing up this line about "certain ranges of precision" you've basically blown your entire argument about determinism out of the water: I *absolutely agree* that you can predict *probable* outcomes to within a "certain range of precision", but that's hardly "deterministic!"

I know you've got an agenda here and won't let go, but as many have said already in this thread, the exercise here is getting quite pointless.

Probably certain,

Buffy

##### Share on other sites

A common feature of threads promoting determinism is that the promoters do not present a consistent concept of determinism, this present thread is typically unsatisfactory.

In post 8 Kriminal99 defines determinism: "by true determinism, I mean what it is that people think of when they say that determinism cannot be disproven"

Who are these people? How do I know what they think? If you (K99) know what these people think, say what it is, make a useful definition. If you are attempting to define determinism as "a member of the set of undisprovable propositions", the thread is pointless.

In this thread you have not:

1) defined determinism

2) presented any argument in support of determinism

3) responded to criticism other than by assuming and asserting determinism

4) maintained a consistent position on the character of determinism

You have not dealt with:

1) Erasmus00's point about quantum events

2) Buffy's point about random sequences

3) the implications of NP-completeness

4) any of my points in post 42, including a simple proof of the incoherency of determinism.

Here is a further proof that determinism is false: A Proof of Free Will

This particular piece of "muck" is by a professional philosopher, as was the previous link I offered you. If you think these arguments are unsupportable you will need to offer a reasonable objection, your response in post 57 was pathetic.

##### Share on other sites

Describing somebody's response as pathetic isn't much of an argument either.

This thread is getting nowhere except toward trouble and is edging its way to being closed...

##### Share on other sites

Qfwfq: Certainly it's not much of an argument, it isn't intended as such. Specifically in post 57 I find "I am not going to read that rediculous spaghetti reasoning". This is willfully maintaining ignorance of the matter I brought to the poster's attention. How would you describe a response that shows no knowledge of the matter brought up and no intention to remedy the situation?

##### Share on other sites

Qfwfq: Certainly it's not much of an argument, it isn't intended as such. Specifically in post 57 I find "I am not going to read that rediculous spaghetti reasoning". This is willfully maintaining ignorance of the matter I brought to the poster's attention. How would you describe a response that shows no knowledge of the matter brought up and no intention to remedy the situation?

Uh because if I write a book called the "BIBLE OF DETERMINISM" that is 400 pages long and designed to make anyone who read it throw up and never argue with a determinist again, it is rediculous for me to tell you to go read it and it is not "willfully maintaining ignorance" for you to refuse. The paper you wanted me to read says things like "1+2=3 is not really that true because 3+2=1" which is just plain stupid.

Very simple, if your opinion that I should read the paper is justified, then it means that you know there is a useful argument in the paper. If you know there is a useful argument in the paper, you could just present the argument yourself. If you do not know there is a useful argument in the paper, then your claim that I should read it is meaningless. I choose to read things that I decide are worth reading, which I would be likely to do if people actually who read the potential material won a debate using something other than argument fallacies and obfuscation tactics.

The whole point I've been trying to make is that convergent and divergent functions *combine* with randomly generated events to *magnify*--not *create*--what appears to be random. You are defining random solely as "apparent" randomness in the aggregate, which is aided and abetted by divergent functions, but that is wholly missing the fundamental concept of randomness.

By bringing up this line about "certain ranges of precision" you've basically blown your entire argument about determinism out of the water: I *absolutely agree* that you can predict *probable* outcomes to within a "certain range of precision", but that's hardly "deterministic!"

I know you've got an agenda here and won't let go, but as many have said already in this thread, the exercise here is getting quite pointless.

Probably certain,

Buffy

Ranges of precision has been the argument all along. I guess you finally decided to read your opponents posts. Unfortunately something isn't over my head because you become insecure when you realize you lost the debate.

What "certain ranges of precision" does is blow YOUR argument out of the water, and now that you have finally understood you are simply trying to cover up this fact.

It is very simple - determinism is about creating models that use the rules in the world to make predictions. I launch a baseball and using only factors that are going to effect the outcome with enough variance to make a difference to the catcher I calculate where the baseball will land. Despite your imagined infinite randomness inherent in the universe, I am still able to do this quite effectively. That is because we do not need to calculate the behaviors of all quantum particles in the universe just to figure out where the baseball is going to land, and even if in a given model there is something that creates enough variance that we do need to include it, with advances in technology we can just break that factor down, and any problematic factors that make it up down etc etc past any arbitrary limit but not to infinity, but rather only until we have reduced the variance in our predictions enough for our purposes.

##### Share on other sites

Kriminal99: The link was a reference for the figure 558. The relevant points about NP-completeness were made in post 56. Are you going to address these points?

##### Share on other sites

Kriminal99: The link was a reference for the figure 558. The relevant points about NP-completeness were made in post 56. Are you going to address these points?

You are not speaking english, you are using dogma associated with a poorly constructed belief set that I am not interested in becoming familiar with.

However I will say that what little time I spent looking at that muck I did notice that he created arbitrary requirements for knowledge - that is why recognition that it's statement "1+2=3 is not that true because 3+2=1" is stupid is a general response to your so called "points". The guy starts off in his goal to create this imaginary dillema by demanding that all statements deal with sets and that in order for the statement to be true it must be satisfied by all possible orderings of the set, which of course requires [math] 2^n [/math] to be verified. Too bad statements don't have to take sets as input in order to be true, and that they can still be true without all possible orderings of their set being true.... It seems his imagined support for these requirements has to do with some unfounded allegiance to the scientific method and with empyrically testing things in place of just using deductive reasoning.

And btw the argument in the above post about you presenting the argument from a material if you were able to find a useful argument, and your opinion that I should read it being meaningless if you cannot do so does not depend on the notoriety of the author of the material. - That is called appeal to authority fallacy. You are also required to present the argument in common language in order for it to be considered something other than an obfuscation tactic. These are just fair rules of debate without respect to the social status of any group of people or source, because social status is not a relevant indicator of validity nor do people from other groups respect the social status of your own group.

##### Share on other sites

When a person instigates a discussion such as this and makes the claim that it's impossible to disprove determinism, I, for one, expect them to know what determinism is and to have grounding in the arguments for and against. I have now read the edit to your previous post, "determinism is about creating models that use the rules in the world to make predictions", clearly indicates that you do not know what deteminism is. If that were the meaning of determinism, how would it be subject to either proof or disproof?

##### Share on other sites

When a person instigates a discussion such as this and makes the claim that it's impossible to disprove determinism, I, for one, expect them to know what determinism is and to have grounding in the arguments for and against. I have now read the edit to your previous post, "determinism is about creating models that use the rules in the world to make predictions", clearly indicates that you do not know what deteminism is. If that were the meaning of determinism, how would it be subject to either proof or disproof?

IT ISN'T THAT IS THE WHOLE POINT. Wow good job. You finally figured out what definition determinists use for Determinism. You see the fallacy of the claim "Determinism is false" from day 1 was that people claiming it used a straw man definition for determinism and then tried to force determinists who were NOT using that definition to accept that it was false. The people who believe in determinism never dealt with or had any interest in that version of determinism.

And also the common version of determinism is very relevant to everyone whereas the "straw man" version is pretty irrelevant to a lot of people...

Do I care that calculating the entire universe would require more space than the universe has? No... I never planned to do that anyways. Of course these reckless uses of all qualifiers leave us the ability to poke holes in your arguments which we might do from time to time.

And it is also fallacious to claim that "Person X disproved your point Y", because obviously everyone always thinks that about people they disagree with during the debate whereas the person they are arguing with does not. I deal with as many attempted arguments as I can, if you think I have missed one then just repost it in hopefully more direct terms. God knows I have done that 30000000 times.