NationStates Jolt Archive

What is the nature of probability?

Is it an objective part of the universe, or is it's existence contingent upon our subjective limitations?

Probably neither of the two.

It is simply a measure of a combination of our ignorance and the uncertainty inherent in the universe. How accurately we can measure under uncontrolled circumstances is unclear.

Probably neither of the two.

It is simply a measure of a combination of our ignorance and the uncertainty inherent in the universe. How accurately we can measure under uncontrolled circumstances is unclear.

Shouldn't that be "both"?

I really should have reworded it, because it is obvious that there is a probability that exists due to ignorance, so I was pretty much interested only with the former part of my question.

I have no idea, but I do know that the probability of me posting in this thread just because I love polls like that is very high indeed. :)

I believe that the second choice will take an early lead because it is blatantly ironic, but as it takes a lead, the first choice will come back.

I believe that the second choice will take an early lead because it is blatantly ironic, but as it takes a lead, the first choice will come back.Probably.

i'm probably not going to post in this thread, but i have to say i lied in the poll.

This is probably the greatest post in the history of NSG.

What is the nature of probability?

Is it an objective part of the universe, or is it's existence contingent upon our subjective limitations?

i asked my calculus TA (who specializes in probability). Apparently, its something you cannot make a lot of money off of researching.

The nature of probability, if it is a combination of two things, is neither one thing nor the other - hence the neither rather than both. But both would have worked to get the message through.

Is probability an ontological fact? That is your question then. I would have to say that if quantum mechanics is in any way close to a description of reality then probability is an essential aspect of that reality. The doubt is that this is only a conditional statement and the condition is one that, at present, we have no way of knowing to be true.

The nature of probability, if it is a combination of two things, is neither one thing nor the other - hence the neither rather than both. But both would have worked to get the message through.

I think that the two ideas of probability are totally unrelated.

Is probability an ontological fact? That is your question then. I would have to say that if quantum mechanics is in any way close to a description of reality then probability is an essential aspect of that reality. The doubt is that this is only a conditional statement and the condition is one that, at present, we have no way of knowing to be true.

I will betray my scientific ignorance, but does quantum mechanics deal with uncertainty of observation, or uncertainty of reality? That is pretty much the question I am asking in the first post, and now it seems to be an unanswerable question.

What is the nature of probability?

Is it an objective part of the universe, or is it's existence contingent upon our subjective limitations?

Probability doesn't exist. It is an abstract concept to figure out the potential that something might happen.

probability is not a thing. It's a concept. It's a measure of how likely something is to come about.

Nothing more, nothing less. While certain physical characteristic dictate probability, probability itself is not a hard coded law or effect.

Probability doesn't exist. It is an abstract concept to figure out the potential that something might happen.

This seems intuitively wrong to me for some reason, but I can't put my finger on it just yet.

Probability doesn't exist. It is an abstract concept to figure out the potential that something might happen.

Is it possible to do away with the uncertainty of observation?

Who knows. Who cares.

I think that the two ideas of probability are totally unrelated.
They are certainly unrelated in their origin, but in their effect they are indistinguishable.




I will betray my scientific ignorance, but does quantum mechanics deal with uncertainty of observation, or uncertainty of reality? That is pretty much the question I am asking in the first post, and now it seems to be an unanswerable question.

QM has probability built in, it is not uncertainty, but a distribution of the probability of a collapse of any state vector wave function to a point across all of space (and time - if you want to separate that from space). The probability is not, as most seem to think, about whether Schrodingers cat is alive or dead, but about the distribution of this cat over the event space that its state vector describes. The actual evolution of states, from one arrangement to another, is completely deterministic (Schrodingers equation is deterministic), it is just that if the resulting superposition of states is reduced by an observation, then there are calculable probabilities as to what will be observed, and these probabilities will be the same every time. (What is observed, of course, need not be.)

The question then resolves to a question as to the veracity of QM. As it is, to date, one of the most precisely verified scientific theories - competing only with General Relativity for this title - its veracity would not be in much doubt were it not for the fact that it is incompatible with General Relativity.

Who knows. Who cares.

You probably should have kept that to yourself.

Is it possible to do away with the uncertainty of observation?
I like to consider the following view of mechanics.
Relativity replaced Newtonian physics in the hyperfast and superlarge domains.
Quantum mechanics did the same in the ultrasmall domains.
Chaos did away with all the Newtonian mechanics that was left.

All physical systems contain a degree of unpredictability and are chaotic (cannot be known exactly). The potential that a system exists in any state is 1. The potential for a system to exist in any substate is a fraction of 1. A chaotic system exists in a definite set of substates that are governed by some set of initial conditions, however it is impossible to predict that at any given time which state exactly it will be in. However, because the chaotic system has a definite set of rules, it can be known or predicted with a certain degree of certainty which states it is most likely to be in, or in which state it is most likely to be found in most of the time.

Therefore yes, probability exists as a non-abstract, physical concept describing the behaviour of a given system. I believe also the Heisenberg uncertainty principle is employed as a physical concept in many derivations, rather than a simple instrumental limitation. But that's where my quantum got sketchy.

Is probability an ontological fact? That is your question then.

Probabilities of 0 and 1 are ontological facts, no? Ergo, probability is an ontological fact.

Probability doesn't exist. It is an abstract concept to figure out the potential that something might happen.

Probability is not solely concerned with the future. We can also use it to describe the present or the past, or indeed hypothetical cases which could never come to pass. None of these latter examples are concerned with what might happen, but with what is happening, has happened or hypothetically could happen.

Therefore yes, probability exists as a non-abstract, physical concept describing the behaviour of a given system. I believe also the Heisenberg uncertainty principle is employed as a physical concept in many derivations, rather than a simple instrumental limitation. But that's where my quantum got sketchy.

If there is uncertainty in observation, how can we preclude uncertainty in reality?

Alternatively, how can we say that our percieved uncertainty of reality is not merely uncertainty of observation?

Who knows. Who cares.I probably do.

If there is uncertainty in observation, how can we preclude uncertainty in reality?

Alternatively, how can we say that our percieved uncertainty of reality is not merely uncertainty of observation?

There is, according to theory, no possibility of measuring something to any greater degree of accuracy than is allowed by Heisenberg's uncertainty principle. As such we cannot now, or ever, know the exact value of any particular quality that we are measuring. As this limitation is inherent in the nature of the universe (as currently understood), and not in our measuring techniques, it seems reasonable to state that the uncertainty is one that is real, rather than one that is a simple failing on our part. However, this uncertainty is not connected in any way with probability, it is just a measure of error. The probability that is inherent in the system comes in through the process of superposition of potential states, which can only occur when the states are not being measured.

(Chaotic systems are not probabalistic - they are purely deterministic, but uncertain due to the limits of our ability to measure the initial state.)

I will betray my scientific ignorance, but does quantum mechanics deal with uncertainty of observation, or uncertainty of reality? That is pretty much the question I am asking in the first post, and now it seems to be an unanswerable question.

The scientific method itself relies upon the assumption that the universe is deterministic - that true randomness does not exist. The idea is that, if we knew all of the underlying mechanisms, as well as all of the initial and boundary conditions for a process, we could predict it with absolute accuracy.

We use probability to describe those things for which we don't have (and may never have) enough information to predict with accuracy. When we start looking at the quantum level, it seems that we may never be able to collect all of the information we would need to precisely define the mechanisms or the conditions (as any attempt to collect such information changes the information we are collecting), so we model it with probability instead.

But, at its core, probability is a matter of our ignorance, not a property of the process or object being observed.

There is, according to theory, no possibility of measuring something to any greater degree of accuracy than is allowed by Heisenberg's uncertainty principle. As such we cannot now, or ever, know the exact value of any particular quality that we are measuring. As this limitation is inherent in the nature of the universe (as currently understood), and not in our measuring techniques, it seems reasonable to state that the uncertainty is one that is real, rather than one that is a simple failing on our part. However, this uncertainty is not connected in any way with probability, it is just a measure of error. The probability that is inherent in the system comes in through the process of superposition of potential states, which can only occur when the states are not being measured.

(Chaotic systems are not probabalistic - they are purely deterministic, but uncertain due to the limits of our ability to measure the initial state.)

As you said, the two types of uncertainty are indistinguishable, so how is it known that the uncertainty caused by measurement does not mask an underlying uncertainty?

As you said, the two types of uncertainty are indistinguishable, so how is it known that the uncertainty caused by measurement does not mask an underlying uncertainty?

It isn't. It is assumed - an axiomatic assumption on which the method of investigation generally employed (the scientific method) is based.

As you said, the two types of uncertainty are indistinguishable, so how is it known that the uncertainty caused by measurement does not mask an underlying uncertainty?

It isn't known. The value may well be undefined below a certain degree of accuracy in reality. The point is though, that this does not introduce probabilities into our description of the universe. It simply introduces a limit to the resolution with which we can examine reality.

Probably neither of the two.

It is simply a measure of a combination of our ignorance and the uncertainty inherent in the universe. How accurately we can measure under uncontrolled circumstances is unclear.

Close.

The universe is not uncertain. There are rules and everything goes acording to plan.

What is uncertain is our knowledge about it. About the rules. Most of all, about human behavior.

I have no doubt that if we could enter all the knowledge about all the variables of the present, knowing all the rules of the universe (natural and human), we would be able to predict the entire future.

Probability doesn't exist. It is an abstract concept to figure out the potential that something might happen.

Agreed. Probability is a atempt to understand how a process work and the potential for it to hapen like we want it to. Of course, since we dont fully understand the process nor all the variables that act on it, you can only judge theoreticaly which and how these variables will behave and try to create a patern based on the few information you have.

You can try creating a probabilyt by observing a lot of experiences on the process and try to pinpoint how the controled variables act on it and on the outcome.

It isn't. It is assumed - an axiomatic assumption on which the method of investigation generally employed (the scientific method) is based.

Sorry Dem, I have to disagree with you here. It is not assumed, nor would such an assumption be axiomatic to the scientific method in general. The scientific method does not demand absolute precision in measurement, it does demand, however, deterministic behaviour.

The same set up, the same operations -> the same result.

Quantum mechanics has challenged this, but it is not the first system to do so. Statistical thermodynamics, used to predict the behaviour of gases etc. is also a probabilistic system, in which the averaged behaviour over a large number of individual items is predicted. We can predict what a gaseous region will do as a whole, we cannot predict what any one element (atom - molecule) of that gas will do.

The universe is not uncertain. There are rules and everything goes acording to plan.

We have no conclusive evidence to show us that this is, in fact, the case.

Close.

The universe is not uncertain. There are rules and everything goes acording to plan.

What is uncertain is our knowledge about it. About the rules. Most of all, about human behavior.

I have no doubt that if we could enter all the knowledge about all the variables of the present, knowing all the rules of the universe (natural and human), we would be able to predict the entire future.

Go study the Bell inequalities then. The position you are taking here was experimental disproven in 1981 by Alain Aspect and colleagues in Paris.

There are rules, yes, but these rules a probabilistic, and the limits to measurement and knowledge are imposed by the nature of the universe, not by our failings, as I have indicated above.

Additionally, you are claiming that every action you make is determined by the state of the universe immediately prior to that action. You may be happy with that position of being a predetermined puppet, but I am not. It may just be wishful thinking, but I prefer to believe in some degree of self determination.

Sorry Dem, I have to disagree with you here. It is not assumed, nor would such an assumption be axiomatic to the scientific method in general. The scientific method does not demand absolute precision in measurement, it does demand, however, deterministic behaviour.

The same set up, the same operations -> the same result.

That's exactly what I said. I didn't say anything about absolute precision in measurement. Determinism is an axiomatic assumption of the scientific method - which means there is no underlying true randomness.

Quantum mechanics has challenged this, but it is not the first system to do so. Statistical thermodynamics, used to predict the behaviour of gases etc. is also a probabilistic system, in which the averaged behaviour over a large number of individual items is predicted. We can predict what a gaseous region will do as a whole, we cannot predict what any one element (atom - molecule) of that gas will do.

But, if we knew the initial and boundary conditions and underlying mechanisms, we could predict what a single element would do - that is the nature of deterministic behavior. It is a matter of our ignorance - the fact that we cannot collect all of this information - that limits our predictions to probabilities and bulk behavior.

It isn't known. The value may well be undefined below a certain degree of accuracy in reality. The point is though, that this does not introduce probabilities into our description of the universe. It simply introduces a limit to the resolution with which we can examine reality.

Then quantum mechanics are pretty much irrelevant, other than making my original question seem to be an insolvable metaphysical connundrum.

There are rules, yes, but these rules a probabilistic, and the limits to measurement and knowledge are imposed by the nature of the universe, not by our failings, as I have indicated above.


How was this shown?

We have no conclusive evidence to show us that this is, in fact, the case.

Agreed.

But we try to prove it.

The scientic method has the assumption that to prove how a process work, you have to limit the number of variables, or at least know how they influence the outcome of the experience.

That way, by changing the variables, making tests and new experiences, you can cut down to the "laws" or "variables" that truly product the outcome of the process.

An example is when you get a medicinal plant in nature, do a research on all the chemicals present, separate one by one and try to find out, by using each one of them in a diferent test subject, which one produces the benefits found initialy in the medicinal plant, that has a lot of chemicals (variants) that have no directly influence on the outcome you wish to have (medical benefit).

Now, the uncertain law or the law of the impredictable consequences say that you will never be able to fully understand the universe, and that means it will always be uncertain. I agree to that. But that doenst mean that there arent rules that we will never know about that directly influence all reality, including our behavior, mind and life.

In my opinion, they exist. Its easy to understand how gravity works, and that if you drop something it will go towards the center of the world. But you cant truly understand how all the planets on the universe act on this object, and how their atrection changes the absolute event. Yet just because we cant understand it doesnt mean it doesnt exist, and that they doensnt act on the object.

I went into psychology just because of it. Because I want to understand the laws humans are controled by and at the same time understand how each one of us is so diferent. If you take an iten, go to a clean room and release an object 100 times, probabily 100 of 100 times it will drop and hit the ground. That is because you control the variables. Some of them, at least. There can be an earthquake, the floor will break and the object will fall to the floor below or into the head of someone. That would mean out of 100 times, 99 times it hited the ground.

That is "easy" to predict. What is hard to predict is if you get a kid into the room. He will try to take the object, jump to get it, steal it from you, bother you and will just disrupt the entire outcome of the probability of the desired consequence.

That is the reason why the scientic method tries to keep the value of human perception to the minimun. Humans are such a huge variable that makes all probabilities hard to predict.

That is why I went into psychology. And also the main reason why, in my opinin, the universe will always be impredictable and random/chaotic (at least in our perception, because I believe that we ultimately dont have "control" over our destiny, meaning everything is predicted).

But, nonetheless, if we could gather all the information about the currently present, how everything influences everything, how all the laws and paterns of the universe work, and managed to create a model with all these informations, you would be able to predict the universe ad infinitum.

That is, of course, a theory that will never happen. If that were to happen, we would become the closest to god that we can. We would become ominiscent and be able to fully understand and know the future and the past.

Just a theory. But I believe in it. In my opinion, if god do exist, he is just like that.

Go study the Bell inequalities then. The position you are taking here was experimental disproven in 1981 by Alain Aspect and colleagues in Paris.

There are rules, yes, but these rules a probabilistic, and the limits to measurement and knowledge are imposed by the nature of the universe, not by our failings, as I have indicated above.

Additionally, you are claiming that every action you make is determined by the state of the universe immediately prior to that action. You may be happy with that position of being a predetermined puppet, but I am not. It may just be wishful thinking, but I prefer to believe in some degree of self determination.

Ok, Ill study them. Probabily not today through, maybe tomorrow.

The rules are a probability only because we dont understand all the variables. If we were to understand all the variables (impossible, because for even the most simple of actions there are infinite variables) and how the influence the outcome, you would be able to have a perfect and complete understanding of the outcome.

Ok Im out of time, Ill only be able to answer tomorrow.

Who cares about probability? What really matters is improbability, and, more specifically, infinite improbability.

What is the nature of probability?

Is it an objective part of the universe, or is it's existence contingent upon our subjective limitations?

It's the absolute square of the quantum wavefunction.

That's exactly what I said. I didn't say anything about absolute precision in measurement. Determinism is an axiomatic assumption of the scientific method - which means there is no underlying true randomness.

Please don't confuse probabilistic states with randomness. QM is deterministic, we agree on that, but it has probability built in as part of the system. From any given, specific and detailed set of initial conditions the system evolves in a completely determined manner to a future set of conditions, just that these conditions are described in terms of probabilities. An electron at point A will, in the future be 50% likely to be at point B and 50% likely to be at point C (A gross simplification - but correct in principle). This will be true of any electron at point A given that the conditions are the same. The theory is thus, deterministic, but ti does not tell us where the electron is, it simply tells us the probabilities of where the electron will be when we detect that electron.


But, if we knew the initial and boundary conditions and underlying mechanisms, we could predict what a single element would do - that is the nature of deterministic behavior. It is a matter of our ignorance - the fact that we cannot collect all of this information - that limits our predictions to probabilities and bulk behavior.
No. QM does not permit us to say exactly what a given element will do. It tells us instead the exact probabilities of what any single element will do. There is a subtle but important difference. Determinism and probablistic behaviour are combined in one system. Hidden variables that would allow us to dispense with the probabilities have been shown to be impossible. There is a good wiki article on this (http://en.wikipedia.org/wiki/Bell's_theorem) (too long to post here).

probability and the diversity of it, is the only real nature of the universe we live in. it does not mean that anything never happens or always does, but that somethings do happen more often, sometimes overwhelmingly more often, then others, and allows that somethings happen a whole lot more often when other things happen first. that doesn't mean one or more self awaire intrests could not exist either, but if they have ever taken a direct personal hand in its detailed shape and form, compelling evidence of their having done so is conspicuously absent. nor is probability such an arbitrary willy nilly thing, that we lack some considerable degree of influence over at least certain aspects of it. though it is of course, widely, blindly, often adamantly, misunderstood.

take the expression "one in a million" for example. sounds pretty unlikely that you'll ever live to see such an event, doesn't it?

yet when you are talking about an anomaly that occurs only once in say a million cycles of something, and those cycles occur, not millions but billions of times every second, a one in a million chance under those conditions, becomes something quite exceedingly common.

that is why people can pretend to be objective while using that pretention to deliberate distort perceptions. real science makes every effort to avoid this, unfortunately commercial interests and the biases of organized so called belief, are too often heard with much louder voices.

people in my generation feared science because hitler had claimed it as his excuse. but hitler's claim to it, was an outright lie and fabrication, based on only marinaly probable possibilities, themselves, for the most part, already discredited by objective evidence even in his own time.

if any thing, he is a parable of the dangers of emotional suseptability to persuasion and the coerciveness of human society and their leading people to ignore objectively observable probability.

=^^=
.../\...

I know this: understanding probability is an important aspect of being a gambler. Life itself is full of gambles, therefor understanding probability is probably important.

Please don't confuse probabilistic states with randomness. QM is deterministic, we agree on that, but it has probability built in as part of the system. From any given, specific and detailed set of initial conditions the system evolves in a completely determined manner to a future set of conditions, just that these conditions are described in terms of probabilities. An electron at point A will, in the future be 50% likely to be at point B and 50% likely to be at point C (A gross simplification - but correct in principle). This will be true of any electron at point A given that the conditions are the same. The theory is thus, deterministic, but ti does not tell us where the electron is, it simply tells us the probabilities of where the electron will be when we detect that electron.

The theory doesn't tell us where the electron is, because it cannot. But something - some set of rules and some initial and boundary conditions - determines whether that electron will be at point B or point C. The fact that we do not know what these things are is the reason that our theory must describe it in terms of probability.

The point here is that I am not confusing probability with randomness - but most people do. True randomness couldn't actually be described with a probability field. But what people generally term as "random" is a matter of probability. It is a matter of probability because, as human beings, we don't have enough information to describe it more accurately.

No. QM does not permit us to say exactly what a given element will do.

Indeed. But QM is our model used to describe the mechanism, not the mechanism itself. Each individual element is moving based upon the underlying mechanisms which control it - as well as the applicable initial and boundary conditions. The fact that we cannot model with that accuracy does not change the fact that each individual element will be doing something - and what it does will be controlled by these things.

It tells us instead the exact probabilities of what any single element will do. There is a subtle but important difference. Determinism and probablistic behaviour are combined in one system. Hidden variables that would allow us to dispense with the probabilities have been shown to be impossible. There is a good wiki article on this (http://en.wikipedia.org/wiki/Bell's_theorem) (too long to post here).

Knowing the variables is impossible. However, if those variables do not exist - if there is not an underlying mechanism that determines exactly what the single element will do - then the system is not actually deterministic. Our inability to describe the actual mechanisms and have the actual variables limits us to probabilistic models.

Indeed. But QM is our model used to describe the mechanism, not the mechanism itself. Each individual element is moving based upon the underlying mechanisms which control it - as well as the applicable initial and boundary conditions. The fact that we cannot model with that accuracy does not change the fact that each individual element will be doing something - and what it does will be controlled by these things.

Knowing the variables is impossible. However, if those variables do not exist - if there is not an underlying mechanism that determines exactly what the single element will do - then the system is not actually deterministic. Our inability to describe the actual mechanisms and have the actual variables limits us to probabilistic models.

OK, we agree up to this point. Then it gets a little bit tricky. The big problem with QM has always been one of interpretation. One can either assume that it is literally describing reality in some way, in which case it is modeling the mechanism itself rather than describing the behaviour of the mechanism, or we can say that as a predictive theory it is very good, but it has severe problems as a descriptive theory. It seems to me that you are of this second school, you accept the predictive power of QM, but would prefer something a little less counter intuitive as a description of reality. I, on the other hand, fall more on the side of accepting QM as being the best description of reality to date, despite its mind twisting implications. As the violation of the Bell inequalities demonstrates that there can be no local hidden variables, we are left with either a violation of General Relativity or the acceptance that the universe is fundamentally probabilistic. As General Relativity has held up extremely well to experimental verification, I side with the probabilistic field.

Now this opens a question about the deterministic nature of the physical universe. If science is to function at all it requires that experiments be repeatable. This repeatability can, however, be statistical in nature. As has happened with gas laws and the like. QM allows for statistical determination. We can predict, reliably, what the distribution of outcomes will be. What we cannot do is predict what the outcome of any one individual event will be. Now if you view determinism in a very strict manner, then QM is not deterministic due to this inability, however, if that is the case then determinism went out the window with radiation and thermodynamics. A slightly more tolerant version of scientific determinism allows that all that is needed is for the probabilities of any particular outcome to an event be known and that over a large number of events that these be confirmed. This allows for repeatability and, as such, rescues the scientific endevour from the spectre of non deterministic behaviour.

It is normal to view a probabilistic model as being part of a deterministic world. I am aware that this view is often justified by claiming that the probabilistic aspect is due to our inability to describe the exact mechanisms or initial state (or both) and that in time we will see that the outcome of any one event is completely determined. QM, however, does not allow this escape. It forces us to face the probabilities as being inherent in the subject, and not as being failures of technique or knowledge. However the probabilities are, themselves, completely determined by the theory. It then becomes a question as to how we interpret what these probabilities mean. I hold them to be literally true descriptions of the state of the universe. The electron is not at B, nor at C, nor at neither, nor at both it exists only as a field of probabilities until some measurement causes this field to collapse into an electron. It is this collapse, as of yet a completely inexplicable event, that breaks the determinism, not QM.