• T Clark
    4.2k
    OP Part 1

    There is a group of issues that I’ve been wrestling with lately. They are ones that come up a lot on the Forum. Specific issues include determinism, predictability, probability, reductionism, emergence, free-will, causation, chaos theory. I don’t want to retread all the recent threads, so I’ll focus on a fairly specific issue. How is determinism different from predictability.

    In a recent thread, I wrote the following:

    It feels intuitively to me that in some, many, most? cases unraveling cause is not possible even in theory. It's not just a case of being ignorant. Part of that feeling is a conviction that sufficiently complex systems, even those that are theoretically "caused," could not be unraveled with the fastest supercomputer operating for the life of the universe. There is a point, isn't there, where "completely outside the scope of human possibility" turns into "not possible even in theory." Seems to me there is.T Clark

    If something is completely unpredictable, does it still make sense to say it is caused. Isn't cause inextricably tied up with prediction? It may be possible to model and predict a coin flip or build a machine that can flip a coin with near perfect uniformity, but how about 1,000 flips using 1,000 random coins flipped by 1,000 random people?T Clark

    For now, I don't want to get into an argument about the difference between causation and determinism.

    In response to this, Wittgenstein posted the following:

    With regards to a complicated system, l have found the following article whose link is below quite useful. From what l have understood partially is that, a deterministic system can be unpredictable because the uncertainty and the error in the initial measurement of the system will cause drastic change in the calculated outcome.

    .... closer look reveals that determinism and predictability are very different notions. In particular, in recent decades chaos theory has highlighted that deterministic systems can be unpredictable in various different ways.

    https://www.google.com/url?sa=t&source=web&rct=j&url=http://philsci-archive.pitt.edu/12166/1/DeterminismIndeterminismWordPittsburghArchiveWithF.pdf&ved=2ahUKEwift_CkjonkAhUL-2EKHSknAvwQFjABegQIDxAH&usg=AOvVaw3LDopPZI0btavaExsb5oik
    Wittgenstein

    I had some problems with the article, and it covered more that I want to cover here, but it raised interesting issues I want to think more about. And I recognize that “determinism” does not necessarily mean the same as “causation,” but I don’t especially want to get in to that here.

    I’ll leave the OP there for now. I don’t like posts that are too long. I’ll immediately follow up in my next post with more specifics on how I see the issue.

  • Terrapin Station
    13.5k
    "Determinism" is easily ambiguous, because it can be used in both an ontological and an epistemological sense.

    "Predictability" is only used in an epistemological sense.

    If we're using "determinism" in the epistemological sense, it makes to see it as synonymous with predictability.

    If we're using "determinism" in the ontological sense, it's definitely different than predictability.
  • T Clark
    4.2k
    OP Part 2

    First, some definitions. Here are some definitions of “determinism” from various places:

    • This is from the article Wittgenstein linked to - “A system is deterministic just in case the state of the system at one time fixes the state of the system at all future times. A system is indeterministic just in case it is not deterministic.”
    • Causal determinism is, roughly speaking, the idea that every event is necessitated by antecedent events and conditions together with the laws of nature.
    • Determinism is the understanding that all events, including moral choices, are completely determined by previously existing causes.

    I want to be clear that I am talking about a strong determinism - protons bouncing off of each other, not the kind of determinism that comes from people being affected by their genetics and environment.

    Here is a definition of predictability from the Wikipedia – “Predictability is the degree to which a correct prediction or forecast of a system's state can be made either qualitatively or quantitatively.”

    OK, down to business. Here is what the linked article says:

    It has often been believed that determinism and predictability go together in the sense that deterministic systems are always predictable. Determinism is an ontological thesis. Predictability – that the future states of a system can be predicted – is an epistemological thesis…. However, a closer look reveals that determinism and predictability are very different notions. In particular, in recent decades chaos theory has highlighted that deterministic systems can be unpredictable in various different ways.

    The article does talk about Chaos theory a bit and I think it might be helpful to discuss it. I may bring it up later.

    Looking on the web, turns out I’m not the first to raise the determinism vs. predictability issue. There’s a lot to look through. Here’s a link to an article I found that had something in particular to say. Its called “Determinism and the Paradox of Predictability.”

    https://link.springer.com/article/10.1007/s10670-009-9199-1

    Here’s what the authors say:

    The inference from determinism to predictability, though intuitively plausible, needs to be qualified in an important respect. We need to distinguish between two different kinds of predictability. On the one hand, determinism implies external predictability, that is, the possibility for an external observer, not part of the universe, to predict, in principle, all future states of the universe. Yet, on the other hand, embedded predictability as the possibility for an embedded subsystem in the universe to make such predictions, does not obtain in a deterministic universe.

    I didn't read the whole article. I just liked/hated the first sentence so much I wanted to use it. I say “bologna.” Well, no, that’s a bit strong, but, as you can see from the quotes in my first post, I don’t agree.
  • T Clark
    4.2k
    "Determinism" is easily ambiguous, because it can be used in both an ontological and an epistemological sense.

    "Predictability" is only used in an epistemological sense.

    If we're using "determinism" in the epistemological sense, it makes to see it as synonymous with predictability.

    If we're using "determinism" in the ontological sense, it's definitely different than predictability.
    Terrapin Station

    In the past I've said that epistemology belongs as part of metaphysics along with ontology. Actually, at heart, when I say that there's no difference between determinism and predictability maybe I'm taking the first step in arguing that there's no difference between ontology and epistemology.

    Not ready for that in this discussion.
  • frank
    3.5k
    There is a point, isn't there, where "completely outside the scope of human possibility" turns into "not possible even in theory."T Clark

    Are you familiar with this guy?

    Probability is a way of expressing prediction, but it doesn't apply to unique events. Trying to squash the concept to fit leads to the conclusion that the outcome of any particular event had a 100% chance of happening.
  • T Clark
    4.2k
    Are you familiar with this guy?frank

    To a certain extent. Actually, his name comes up a lot when you type "determinism vs. predictability" on the web. How is he specifically relevant?

    Probability is a way of expressing prediction, but it doesn't apply to unique events. Trying to squash the concept to fit leads to the conclusion that the outcome of any particular event had a 100% chance of happening.frank

    We're talking about predictability, not probability. Again, how is this relevant to the issue as I've laid it out?
  • T Clark
    4.2k


    In the past I've said that epistemology belongs as part of metaphysics along with ontology. Actually, at heart, when I say that there's no difference between determinism and predictability maybe I'm taking the first step in arguing that there's no difference between ontology and epistemology.T Clark

    I get the feeling I've stepped off a cliff with this statement. On the forum, unlike in roadrunner cartoons, I have time to step back before I fall. I'll say it differently - just throwing out the words "ontology" and "epistemology" doesn't really respond to my posts. It doesn't really say anything. How about a bit more in depth response.
  • Wittgenstein
    191

    Are you familiar with this guy?
    To a certain extent. Actually, his name comes up a lot when you type "determinism vs. predictability" on the web. How is he specifically relevant?

    That guy is in mentioned in the article you posted, l think he hasn't checked it out yet. :smile: I will try break down the article in some basic points so we can remove the scholarly jargon and discuss the real matter at hand and l will fail possibly but let's give it a try.

    1. asymptotic unpredictability. (AUP)
    This unpredictability is due to how little changes or inaccuracy in initial measurement of system will drastically undermine the accuracy of our predictions in a deterministic system. The reason is that such small or little conditions or values of the system spread over all the predictable range in a short time.
    However asymptotic unpredictability is not unique to chaotic systems as some systems can be simple and not complicated and still have asymptotic unpredictability.
    ... has argued that approximate probabilistic irrelevance is the kind of unpredictability that is unique to chaos. Unlike asymptotic unpredictability, approximate probabilistic irrelevance is a probabilistic concept of unpredictability. According to this concept, any measurement (i.e. knowledge of the initial states that the system may currently be in) is irrelevant for practical purposes for predicting outcomes sufficiently far in the future
    Here we can see that the author (Werndl) is placing probability as a kind of predictability. Hence instead of saying all choatic system are associated with AUP, he states that are probability of future states cannot be obtained from initial values of the system.

    2.The link between determinism and stochastic systems
    Hence what is meant by the phrase that the deterministic model and the stochastic model give the same predictions is that the possible observed values of the stochastic system and deterministic system are the same, and that the probability distributions over the
    sequences of observations of the deterministic model and the sequences of outcomes of the stochastic model are the same.
    How does having the same prediction indicate a similar system ? There is also a problem with predicting stochastic systems behavior but simple deterministic system can be easily predicted. I hope l am not missing something here. The author doesn't seem to connect the 1st topic with the second one.After that, they discuss underdeterminism, which is refuted by stating that evidence supports deterministic systems over stochastic systems, hence they are favourable. Personally, l dont think this is related to our topic.
  • T Clark
    4.2k


    One of the reasons I didn't like the article you linked to was how poorly I thought it described the issues you are talking about above. I guess the point it was trying to make was that there are, after all, good reasons why a deterministic system might be unpredictable. That brings me back to my previous statements

    It feels intuitively to me that in some, many, most? cases unraveling cause is not possible even in theory. It's not just a case of being ignorant. Part of that feeling is a conviction that sufficiently complex systems, even those that are theoretically "caused," could not be unraveled with the fastest supercomputer operating for the life of the universe. There is a point, isn't there, where "completely outside the scope of human possibility" turns into "not possible even in theory." Seems to me there is.T Clark
    If something is completely unpredictable, does it still make sense to say it is caused. Isn't cause inextricably tied up with prediction?T Clark

    I just don't see how it makes any sense to say something is deterministic if it can't be used to determine, i.e. predict, something. A quick survey of the web on this issue shows that a lot of other philosophers have felt the way I do, although the majority seem to disagree with my position.
  • Janus
    8.5k
    Determinism is an ontological thesis. Predictability – that the future states of a system can be predicted – is an epistemological thesis…. However, a closer look reveals that determinism and predictability are very different notions.T Clark

    I think the reason these two notions are often conflated is, in part, due to the fact that the ideas of determinability and predictability are, at least in relation to outcomes, synonymous.
  • fdrake
    2.7k
    Just want to throw this into the hat.

    You flip a fair coin, the probability of heads is 0.5, the probability of tails is 0.5. The outcome is not predictable, but the probabilistic behaviour can be fully specified. Whether this behaviour arises from true randomness or as a result of an intricate dependence of the dynamics of coin flipping to the forces applied to lend it rotation and project it through the air, the distribution of heads and tails is still part of the system. Probability's a latent structure of even fully deterministic systems.
  • Janus
    8.5k
    Probability's a latent structure of even fully deterministic systems.fdrake

    Would you say probability is a "latent structure" of the deterministic system itself or merely of our attempt to understand it?
  • T Clark
    4.2k

    Just to be clear, I disagree with the quote you attributed to me. That's not my position, it's the position of the paper I referenced.

    I think the reason these two notions are often conflated is, in part at least, due to the fact that determinability and predictability are, in some sense, synonymous.Janus

    It seems you're saying they are conflated because they mean the same thing.
  • T Clark
    4.2k
    You flip a fair coin, the probability of heads is 0.5, the probability of tails is 0.5. The outcome is not predictable, but the probabilistic behaviour can be fully specified. Whether this behaviour arises from true randomness or as a result of an intricate dependence of the dynamics of coin flipping to the forces applied to lend it rotation and project it through the air, the distribution of heads and tails is still part of the system. Probability's a latent structure of even fully deterministic systems.fdrake

    When I look at the definitions of determinism above, it seems to me it's talking about, not the ability to predict that after 1,000 throws there will be about 500 heads, but the exact sequence of results - h,t,h,h,t,t.....t,h, h, h, t, t.

    Seems to me that probability has a place in this discussion, but I'm not exactly sure how. Maybe I have not been clear enough about what I mean by 'predictable."
  • Janus
    8.5k
    It seems you're saying they are conflated because they mean the same thing.T Clark

    Just to be clear I was neither attributing nor not attributing the idea expressed in the text I quoted to you; I was merely responding to the text.

    And yes, I am saying that determinism and predictability are often conflated, probably at least in part due to the fact that 'determinability' and 'predictability' mean the same thing in respect of outcomes; that coupled with the obvious relationship between the ideas of determinability and determinism.
  • Wittgenstein
    191

    You should try to refute AUP, even though it seems wrong intuitively. Maybe it is talking about lack of information causing the unpredictability and we can perhaps predict chaotic system if the initial values are accurately known or maybe it is inherent in the system
  • T Clark
    4.2k
    And yes, I am saying that determinism and predictability are often conflated, probably at least in part due to the fact that 'determinability' and 'predictability' mean the same thing in respect of outcomes; that coupled with the obvious relationship between the ideas of determinability and determinism.Janus

    I don't think that is the reason predictability and determinism are assumed to be equivalent. I think they genuinely are, in all meaningful ways, equivalent at the point, as I said,

    where "completely outside the scope of human possibility" turns into "not possible even in theory.T Clark

    At that point, it becomes what is often called a distinction without a difference. I think that's the heart of my argument. To me, if prediction of a system's behavior is "not possible even in theory," it doesn't make sense to say it is deterministic.
  • T Clark
    4.2k
    You should try to refute AUP, even though it seems wrong intuitively. Maybe it is talking about lack of information causing the unpredictability and we can perhaps predict chaotic system if the initial values are accurately known or maybe it is inherent in the systemWittgenstein

    I don't think I was clear enough. AUP doesn't seem "wrong intuitively" to me, I just don't think it changes my argument. As I said, it seems to me it's just one of the

    good reasons why a deterministic system might be unpredictable.T Clark

    Or am I missing something?
  • Wittgenstein
    191

    I just don't see how it makes any sense to say something is deterministic if it can't be used to determine, i.e. predict,

    I have two questions for you regarding this statement.
    Choatic system are deterministic but does it say in theory ( not practise ) that it is impossible to predict the future states, as you have mentioned in a discrete manner ?

    Let's suppose it does, but why is predictability a neccessary condition for a deterministic system ?
  • T Clark
    4.2k
    Choatic system are deterministic but does it say in theory ( not practise ) that it is impossible to predict the future states, as you have mentioned in a discrete manner ?Wittgenstein

    That was one of the points I was trying to make. In my opinion if something is so difficult to predict that it is and will never be possible to do so, it doesn't make sense to call it deterministic. To me, that would be the same as saying even if only God can predict it, it's still deterministic. I think that's what people are saying, and I don't agree with it.
  • Janus
    8.5k
    Sure, you can say that. But there is nothing stopping people speculating as to whether nature is deterministic or indeterministic, and finding that such speculation does make sense to them. Laplace, for example, thought nature was deterministic, and that if you could know the position, momentum, mass and direction of all fundamental particles at any time, you could predict all subsequent events.

    Of course, he also recognized that this is not possible for human beings, but only for his hypothetical "Demon". Later, someone (I forget who right now and can't be bothered searching it), showed that, even assuming that nature is utterly deterministic, such prediction would be impossible even in principle due to the so-called Three Body Problem.
  • fdrake
    2.7k
    Would you say probability is a "latent structure" of the deterministic system itself or merely of our attempt to understand it?Janus

    It can be either or both. I'll focus on it as a latent structure, seeing as the epistemic angle is well known.

    The distribution of outcomes H-T-H-... or whatever derives from the dynamics of coin flipping but is a legitimate part of the system as much as the centre of mass/centroid being in the same place on the coin. Another example there is if you throw a football around (which is very spherical) the probability that it comes to rest on any given side should be approximately equal to the proportion that side is of the surface area. I think this notion is summarised by all points on the faces of the coin and the faces of the football being generic, so that when you partition the surface into the faces (heads/tails, each football face) the uniform distribution holds of the partition.

    A more realistic system might be a melting candle in a restaurant near a door. The candle's lit, the wind's blowing through the door. The gusts are of random strength and direction, but they tend to come in through the door. The candle will melt more on the side furthest away from the wind's gusts since they will bend the flame that way. An adequate description of that system would be some model of how the candle melts in the absence of wind plus some directional wind that changes the proximity of the flame to one side of the candle. If you could find the effect of a gust on the rate of change of the candle wax melting in that direction, then make the gusts random (but clustered in appropriate directions and of appropriate strengths), that would be an adequate description of the system.

    AFAIK that can actually be done with bridges, stress testing can be done in mathematical models by looking at them in a random form, adding random forces and torsions. Seeing how the bridge would respond to them.
  • Wayfarer
    8.6k
    A system is deterministic just in case the state of the system at one time fixes the state of the system at all future timesT Clark

    One thought that comes to mind, is that, unless you're a systematic philosopher, the world is not a system. It is rather more like what is required for there to be systems. But I am inclined to believe that the world must transcend any notion of 'system'.

    Are you familiar with this guy?
    — frank

    To a certain extent. Actually, his name comes up a lot when you type "determinism vs. predictability" on the web. How is he specifically relevant?
    T Clark

    'This guy' is Simon LaPlace, who is given the title 'France's Newton'. He was a genius, inventor of many fundamental ideas in modern cosmology and social science, including the concept of demographics (among many others.)

    What I think Frank is referring to is directly relevant, and as it hasn't been spelled out yet, let's do it. 'LaPlace's Daemon' says:

    We may regard the present state of the universe as the effect of its past and the cause of its future. An intellect which at a certain moment would know all forces that set nature in motion, and all positions of all items of which nature is composed, if this intellect were also vast enough to submit these data to analysis, it would embrace in a single formula the movements of the greatest bodies of the universe and those of the tiniest atom; for such an intellect nothing would be uncertain and the future just like the past would be present before its eyes.

    — Pierre Simon Laplace, A Philosophical Essay on Probabilities.

    Now, regardless of the merits of this statement in light of what has happened since 1814, when it was published, I feel as though this statement is hugely relevant to this thread, as I think this is the source of the whole idea of 'determinism' which so many people who turn up on this forum and post seem to take for granted. I often feel like asking them if they've heard of this guy, but I'll hold off for now.
  • Harry Hindu
    2.3k
    That was one of the points I was trying to make. In my opinion if something is so difficult to predict that it is and will never be possible to do so, it doesn't make sense to call it deterministic. To me, that would be the same as saying even if only God can predict it, it's still deterministic. I think that's what people are saying, and I don't agree with it.T Clark

    So, what can be the case is only what you can understand, or know, it to be? Sounds like solipsism to me.

    Before God created humans, the universe was deterministic but when he created humans, the world became indeterministic? God isn't a god if it's omniscience is erased with the creation of humans. It doesn't make sense. It seems to me that it is entirely possible that the world is a certain way that is different than our knowledge of it (indirect realism).

    Our knowledge is fallible. Randomness is the result of a lack of knowledge of some system. Once we acquire the necessary knowledge the system becomes predictable. Predictions and randomness are ideas that exist in one's head as a result of one's knowledge. What may appear random to you is predictable to me because we both have different knowledge of the system.
  • Harry Hindu
    2.3k
    If I were to write a complex computer program for the behavior of a human-like robot, there would most likely be bugs that would need to be worked out after the initial release. The program is so complex, and its interactions with the world so varied, that I can't predict the outcomes of every type of interaction that may occur. This isn't because the world is indeterministic. It is because my knowledge and memories are limited.
  • frank
    3.5k
    You flip a fair coin, the probability of heads is 0.5, the probability of tails is 0.5.fdrake

    This is way of presenting logical possibility. True, it's derived from facts about the system, but it's still at most a way of weighing expectation. It's an expression of uncertainty. We actually have no knowledge about which side will land face-up unless the system is rigged.

    The way this can be misunderstood is to imagine that probability is about a branching future.
  • fdrake
    2.7k
    This is way of presenting logical possibility. True, it's derived from facts about the system, but it's still at most a way of weighing expectation. It's an expression of uncertainty. We actually have no knowledge about which side will land face-up unless the system is rigged.frank

    And yet the real instances of flipping fair coins produce this kind of distribution. Very similar to sex proportions in birth. The claim that 'if we knew all relevant information then the future would be fixed' isn't inconsistent with probability; when what flipping a coin is contains that uncertainty about the future.

    Who is doing all this knowing? Where is it in the system?
  • T Clark
    4.2k
    Sure, you can say that. But there is nothing stopping people speculating as to whether nature is deterministic or indeterministic, and finding that such speculation does make sense to them.Janus

    Yes, I agree completely. "You can say that, but there is nothing stopping people speculating.." Sounds like a definition of "philosophy" to me.

    I say this all the time - This is a metaphysical/epistemological issue. It's not a matter of fact, it's a matter of choice and usefulness. I think making the distinction between determinism and predictability is not useful in most cases and I think it is often misleading. Here's a quote from William James's "Pragmatism."

    Pragmatism, on the other hand, asks its usual question. "Grant an idea or belief to be true," it says, "what concrete difference will its being true make in anyone's actual life? How will the truth be realized? What experiences will be different from those which would obtain if the belief were false? What, in short, is the truth's cash-value in experiential terms?"

    The moment pragmatism asks this question, it sees the answer: True ideas are those that we can assimilate, validate, corroborate and verify. False ideas are those that we cannot. That is the practical difference it makes to us to have true ideas; that, therefore, is the meaning of truth, for it is all that truth is known-as.


    I guess that makes me a pragmatist.
  • T Clark
    4.2k
    One thought that comes to mind, is that, unless you're a systematic philosopher, the world is not a system. It is rather more like what is required for there to be systems. But I am inclined to believe that the world must transcend any notion of 'system'.Wayfarer

    That quote was from the journal paper Wittgenstein linked me to. I wasn't trying to make a point by using that word. If it confuses things, we can just use one of the other definitions. That's why I like to provide more than one when I can.

    What I think Frank is referring to is directly relevant, and as it hasn't been spelled out yet, let's do it. 'LaPlace's Daemon' says:

    We may regard the present state of the universe as the effect of its past and the cause of its future. An intellect which at a certain moment would know all forces that set nature in motion, and all positions of all items of which nature is composed, if this intellect were also vast enough to submit these data to analysis, it would embrace in a single formula the movements of the greatest bodies of the universe and those of the tiniest atom; for such an intellect nothing would be uncertain and the future just like the past would be present before its eyes.

    — Pierre Simon Laplace, A Philosophical Essay on Probabilities.

    Now, regardless of the merits of this statement in light of what has happened since 1814, when it was published, I feel as though this statement is hugely relevant to this thread, as I think this is the source of the whole idea of 'determinism' which so many people who turn up on this forum and post seem to take for granted.
    Wayfarer

    How is that different than the definitions of "determinism" provided in the OP? Obviously, it provides a more detailed description, but otherwise I don't see any inconsistency. I don't know if you read the article, but the authors call out this specific quote for criticism.

    I often feel like asking them if they've heard of [Werner Heisenberg], but I'll hold off for now.Wayfarer

    I specifically left out quantum mechanics from this discussion because I wasn't sure how it fit in. I have no objection to you bringing it up. It is my understanding that QM is considered a deterministic theory.
  • T Clark
    4.2k
    Our knowledge is fallible. Randomness is the result of a lack of knowledge of some system. Once we acquire the necessary knowledge the system becomes predictable. Predictions and randomness are ideas that exist in one's head as a result of one's knowledge. What may appear random to you is predictable to me because we both have different knowledge of the system.Harry Hindu

    If I were to write a complex computer program for the behavior of a human-like robot, there would most likely be bugs that would need to be worked out after the initial release. The program is so complex, and its interactions with the world so varied, that I can't predict the outcomes of every type of interaction that may occur. This isn't because the world is indeterministic. It is because my knowledge and memories are limited.Harry Hindu

    As I said previously:

    There is a point....where "completely outside the scope of human possibility" turns into "not possible even in theory."T Clark

    At that point, in my, and others, opinions, it stops being deterministic.
  • T Clark
    4.2k
    'if we knew all relevant information then the future would be fixed'fdrake

    As I've said, there are times when knowing all relevant information isn't possible, even in theory. Even if it were possible, that information would also have to be processed in order to make a prediction.
bold
italic
underline
strike
code
quote
ulist
image
url
mention
reveal
youtube
tweet
Add a Comment

Welcome to The Philosophy Forum!

Get involved in philosophical discussions about knowledge, truth, language, consciousness, science, politics, religion, logic and mathematics, art, history, and lots more. No ads, no clutter, and very little agreement — just fascinating conversations.