Why Do You Remember The Past But Not The Future?

When discussing eternalism and the block universe, the concept of “now” always ends up getting relegated to an aspect of our consciousness, not something “out there”. “Now” seems to be the boundary between what we can remember and what we can only anticipate. But if, aside from entropy, the laws of physics are reversible and we live in a block universe, why are our memories only of the past? Matt O’Dowd gives an interesting answer.

PBS Spacetime” Why do you remember the past but not the future?

O’Dowd’s answer (in case you skipped the video) is that correlations tend to increase over time, including the correlations in a physical object, such as a brain or rock, and its environment. Of course, in the case of brains, we call some of these correlations “memories”. Another name for many of these correlations might be “information”.

I’ve said before that I think information is essentially causation. And the vast majority of correlations come about by one correlated entity causing another, or by some common cause or set of causes in their shared history. So that fits.

O’Dowd does note the possibility of correlations existing without the causal processes, but they are highly improbable, akin to the probability of entropy decreasing on its own. And it’s worth noting, except in the simplest cases, correlations are only correlations due to connecting causal processes, both in their formation, and in whatever downstream effects the correlation may have, such as us recognizing a scar as the result of some kind of physical strike.

Ever since I found out that the measure of information content in Shannon information theory basically has the same mathematics as entropy, I’ve pondered what it meant. On the one hand, entropy itself is the result of causal processes. So both entropy and correlations seem to be the result of causal processes.

Of course, we do have correlations in our brain of future events, but they are either events we might cause, or see as probable due to some recognized repeating pattern. But these correlations have an uncertainty because of other correlations that don’t yet exist for unexpected events.

As I said above, it’s an interesting explanation. But I’m not yet entirely sure it isn’t tautological. We have memories (correlations) because they tend to increase over time. Still, not all tautologies are obvious, and so this still feels like an insight.

What do you think? Does the explanation work? Or is there a better one?

44 thoughts on “Why Do You Remember The Past But Not The Future?

    1. It okay to pick nits. Hope you’re okay with me picking back. 🙂

      I think correlation does broadly imply causation, just not direct causation between the correlated entities. The causal factors that link them may well be common causes in their remote past, causes that wouldn’t imply anything useful, such as what causes schizophrenia.

      Liked by 1 person

      1. I remember my statistics course at university back in 1967. We ran our calculations on heavy Frieden calculators. We were forbidden to divide by zero because it would have caused the roller to fly all the way to the right, probably killing the student next to me. Anyway, I don’t accept your nitpick. There were specific functions for calculating coefficients of correlation and functions (such as reliability and validity coefficients) for calculating causality. A and B might be correlated to some degree, but whether A determines B (or B determines A) is determined by calculating the coefficients or reliability and/or validity of A against B (or B against A). In general, correlation has always been a weaker relationship than causality (reliable or valid determination). 🙂

        Liked by 1 person

        1. Mike, I’m not sure I got my point across. I’m not contesting the general principle that correlation does not imply causation. But that principle applies to causation between the correlated variables, not to any causation at all.

          Something, or perhaps several somethings, cause A, and something(s) cause B. If they’re reliably correlated, that by itself doesn’t imply that A causes B or B causes A, but there’s a very high probability they have commonalities in their causal history. It might be that both are caused by C. Although C might be a collection of causal factors, or so deep in that history that it’s useless for diagnosing the more immediate causes of A or B.

          Liked by 1 person

  1. That the the laws of physics are time invariant is perhaps overblown. Those laws that are statistical in nature do show evidence of “time’s arrow” and I suspect that those that are not immersed in statistics, the invariance is irrelevant.

    We tend to think the past still exists, thus the large number of time travel fictions being written, but this is illogical. What “time travel mechanism” has the power to get the earth’s continents to drift back to where they were in 345 BCE? Our actual experience indicates that only now(+/-) exists. I say now(+/-) because it seems very hard to define now, so my wonky term is referring to our sense of the now.

    The universe is expanding (faster and faster it seems) and so going back to a less expanded universe does not seem in the cards and by reflection, so does going to a more expanded universe.

    We are all time-travelers, but it seems we are limited to travel at one second per second.

    Liked by 3 people

    1. What you’re describing is a variant of presentism, the idea that only the now exists. For many practical purposes, it’s a useful view. But Einstein’s relativity theories complicate the picture. There are frames of reference where you’re going backward in time right. Granted, short of some form of FTL tech, they’re not frames we can ever access, but they exist.

      More relevantly, there are frames where you’re “now” can be very different from someone else’s “now”, depending on how fast each of you is moving or your location in a gravity well. Simultaneity is relative. That’s what makes eternalism seems like a logical consequence of relativity.

      Of course, if quantum mechanics truly has a random component, then that logic could be undermined.

      Liked by 2 people

  2. O’Dowd’s explanation is brilliant. Philosophers of science (and maybe physicists too) usually talk more about “records” but moving immediately to “correlations” makes it clearer, to me at least.

    On the information:entropy connection, see chapter 9 of E T Jaynes. Also, I’ve heard that Gibbs’s definition of entropy is simply the flipside of Shannon’s information (decreasing Shannon-information = increasing Gibbs-entropy).

    The non-tautological part of the explanation is that correlations tend to increase over time because entropy was so low in the past, back to the Big Bang. And relatedly, the fact that entropy and correlations grow together. Be careful about calling conceptually linked facts tautologies: the fact that our concepts actually apply to the real world (when they do) is a non-tautological element, and it’s a doozy.

    O’Dowd too quickly moves from “why does the brain see the arrow of time so clearly?” to “why do we remember…”. There’s another huge aspect of the arrow-of-time that we experience: the fact that we can affect the future but not (at least, not that we can detect) the past.

    But this has the exact same answer! Namely, those correlations, those marks on future rocks and trees and brains, are the effects we are having on the future. Those are the records of our actions. So, it’s unfortunate that O’Dowd didn’t tackle this issue as well.

    Liked by 3 people

    1. Hopefully I conveyed my uncertainty about it being tautological. And the tautology I had in mind wasn’t between entropy and correlations, although I definitely agree it’s there, but between us forming memories and correlations increasing.

      I see what you’re saying about the arrow of time. It’s a nuance he did pass over. Probably time constraints. Arguably, they’re linked, in that the evolutionary reason we remember the past is so we can use that information to affect the future. It’s biology taking advantage of the fact that patterns regularly repeat.

      Liked by 1 person

      1. Sorry, yeah, I was adding a new pseudo-tautology with the entropy-and-correlations. I get a little fanatical because this is a very big stumbling block for philosophy. Philosophy of mind especially, I think. This is a very subtle way that error can sneak into an attempt at careful philosophical reasoning.

        Are you familiar with the Buckingham Pi Theorem? Working through some examples in mechanical engineering class led me to wonder, how the hell can this possibly work? And then I realized: there are hidden claims embodied in our deployment of certain concepts.

        Liked by 1 person

        1. Oh, no worries at all. And like I said before, even tautologies are not obvious. Sometimes pointing them out is a serious insight.

          Can’t say I’m familiar with that theorem. Looks like one of those “unreasonable effectiveness” of mathematics things. But as you note, it’s often because we have unconscious assumptions binding things together.


  3. Causality does not communicate a unitive field. That is the basic problem in this video and the scientific approaches in general. Not that they are incorrect in the estimations and theories, but that they all assume that causality indicates or identified a ubiquitous ontological maxim.

    Philosophy offers us a more thorough possibility why we don’t remember the future: We are, but we are ideologically informed to exclude certain types of information, or rather, for a better way of putting it, humans generally see what they know.

    Further, the category “human being” is assumed in the ubiquitous general attitude which accompanied “causality”.

    In short, in general modern humanity measures the efficient cause as the basic and foundational causality.

    There is a way to reason which considers the efficient cause as secondary or not basic and not assumed.

    What is correlational thereby indicates the efficiently causal manner of universal form, where as the formal cause thereby is able to understand that the efficient cause is but one form of knowledge in the universe (not of the universe).

    But of course the basic formal cause is not as exciting as the efficient cause, Becuase it doesn’t leave so many openings for interesting imaginary ideas.

    Liked by 2 people

    1. If I’m understanding what you’re saying, the standard causal framework is incomplete. The formal cause sounds like the Aristotelian causal framework. Is that right? And would the formal cause essentially involve platonic forms? Or am I completely off base?

      I have to admit my knowledge of Aristotle and Plato could be better. I read some Plato in college, but wasn’t really into it at the time, and have just never swung back to rectify.

      Liked by 1 person

      1. Aristotle yes have us the possibility. I don’t necessarily prescribe to his proposals exactly, however.

        It seems to me that what contradiction indicates is failure. Not just some strange hole in reasoning. For sure, I was assuming a lot by adding my comment here, a lot that my work is about. But mainly philosophy as a source of true knowledge is now placed second or third or fourth, in front of any other number of disciplines which propose to be able to get out any truth, including science, which seems to be the most popular.

        The eternal differences of opinion, along with what science actually is, which I see is a method, a way of doing things, a way of allowing us to get creative with how we are in the world upon other elements of the world. It seems to me that all these various ideas really indicate that what we’re understanding is causality is really kind of a religious faith in a cosmology, and that’s why things make sense to us the way they do. I associate this kind of “faith” in efficient causality, in creation, in a creator (an agent), that which affects some thing else is said to be arranged in a causal manner, and then we trace that lineage of efficient cause.

        I’m not saying it’s necessarily wrong, but I’m saying that sometimes the answers we are seeking we will not be able to find in that lineage simply because that lineage is based in a particular way of viewing things. It only allows us to view things that are allowed by that view, through that lens, so to speak. This is why the efficient cars is so attractive; this is where the idea of creativity and imagination comes from, because existence is so much larger than that particular framework by which were trying to capture everything: we have to get creative.

        Take for your example of the guys video that your post is really discussing. I really like the video by the way and I do like all those kind of philosophical considerations. But the main thing that I like is that he tells us that time is really just a variable. There is nothing in existence which corresponds with time. It is utterly just a placeholder in a series of formulas that we use to do things. There really is no such thing as time in a physical sense. And so the question is why do we perceive the past as memory and we can’t remember things from the future. And then he goes to talk about the brain and everything like that right.

        Well, All of these relationships that he’s drawing into this magical synthesis is really based on the idea that causality is a singular form. And not ironically. Rather, it is assumed Essential and absolutely singular, a singular way of tracing real things, if I can say it like that.

        Yet if we do look back to Aristotle who gives us four different kinds of causality. One of which which is the efficient cars. We see that he also talks about a formal cars, how causality is the form that things are in. So we could say that a cause of a tree is the form that it is.

        Now, I’m not sure how time would fit into that. And so for sure we could say that from the formal causal standpoint we are seeing the future right now.

        Then we might be able to compare the two causes. Because Aristotle talks about how these four different types of causalities do not reduced to one another. They reduced to their own “metaphysical truth”, so to speak. The idea of “metaphysics“ is something that someone came up with like 1000 years after Aristotle to try and explain what he was talking about or certain features of his philosophy, but Aristotle did not have metaphysics. That is something totally invented by someone else in my opinion.

        But so if we allow for different types of causality, then we can compare what might be going on.

        Because their knowledge becomes something more or something that is rooted in various types of causality, instead of just one. Our imagination only gets bigger, solutions become more encompassing.

        Anyways. It seems to me that presently we are caught in reducing or trying to reduce everything to the efficient cause as if it is the only way that we are able to come upon a truth of reality.

        Sorry the long rambling.


        1. I hadn’t heard that about Aristotle before. I had heard that his corpus was really lecture notes for use in his school, and probably reflect not just his work, but those of others. And that the chapter divisions and titles were later additions. So even if he did write “metaphysics”, a word which just means “beyond physics”, he wasn’t the one who made the distinction between physics and metaphysics.

          From what I’ve read, science’s focus on efficient and material causes started during the scientific revolution. It represented the influence of renaissance engineers, who only had to care pragmatically whether their designs worked. People like Galileo, who was both a natural philosopher and engineer, began the evolution from natural philosophy to science by importing engineering methods.

          While it’s undoubtedly a more pragmatic approach, it has always left a lot of people unsatisfied.

          Liked by 1 person

          1. Indeed. Efficient Causality works; it accomplishes tasks. But I feel that at some point we should be able to begin to be honest with what is happening with knowledge truly. And this is to say to begin to reckon what kind of knowledge is good for what kind of tasks, rather than continue blindly to assert that every task must answer to one type of knowledge.

            Liked by 1 person

  4. Mike, at the beginning of your post, you mention ‘now’, which is generally agreed to be wholly subjective, since there is no ‘now’ in the universe or the laws of physics. That strongly suggests that ‘now’ is a feeling—specifically, the feeling of the immediacy of conscious experience. While any feeling of ‘now’ can be correlated with a clock time by looking at a clock, ‘now’ is not a time.

    After watching some of O’Dowd’s video, my overriding question is: Why is there such an urge towards explanatory complexity? Given multiple explanations, simplicity has always been viewed as a value that elevates one explanation over others—the simpler explanations always tend to be the most elegant as well.

    The answer to the “why do we remember the past but not the future” question is that consciousness is forward-time directional. Memories—a multi-sensory gestalt record of conscious experience—are as forward-time directional as the experience they represent. The forward-time directionality of memory explains how we can tell that a movie of a line of people backing off a stopped train is being shown in reverse. Both learned and built-in reflexes share this time directionality and cannot conceivably operate in reverse.

    As you wrote, the laws of physics are reversible but it’s obvious that forward-time directional consciousness is not. As a thought experiment, imagine reversing the laws of physics—actually running the universe in reverse starting after you read this blog comment. Since reflexes and memories are not reversible, a massive cognitive dissonance would be the immediate inevitable result, resulting quickly in madness and eventually life threatening malfunction.

    From the block-view of Eternalism there is no such thing as correlation/causation nor is there any Arrow of Time—everything just is. Correlation/causation belong to the dynamic-view of the block universe that’s provided by consciousness. In my view, our feeling that there is an objective flowing time in the world is simply an externalization of our forward-streaming flow of consciousness. That mysterious Arrow of Time is an externalization of the stream of consciousness as well.

    Liked by 1 person

    1. Stephen,
      I agree that consciousness is forward-time directional, but the question for many of us is, why? I think O’Dowd is trying to get at the answer from a physics (well, astrophysics) perspective. For most of us, saying that it just is, is unsatisfying. We need more. And the correlations seem like a decent attempt at it, a good supplement to the overall entropy and thermodynamic explanation.

      On the block-view versus dynamic-view, I think there’s room for both perspectives depending on the context we’re discussing. We might conceptualize the whole thing as a block, but when working with more of a subset of it, a smaller more limited scope, the dynamic view will be more productive. It’s basically just finding different models productive at different levels of abstraction. In other words, discussing things in dynamic terms is not a rejection of a broader block view. It’s more seeing the block from the inside as a forward time-dimensional system with all the limitations involved.

      Liked by 1 person

      1. I agree: “consciousness is forward-directional” is not the end of the story, but only the end of the introduction where the main characters are introduced. The main action of the story is to explain this fact about consciousness.

        Liked by 1 person

    2. Mike, you pose the question: “Why is consciousness forward-time directional?” because “saying that it just is, is unsatisfying.”

      Consciousness presents as a forward-directional stream which is an attribute so closely tied to consciousness that I consider it definitional—it’s a fundamental matter of fact about consciousness. While pathologies of consciousness are known that slow, speed up and even stop the flow, no instances of reversal have been observed or reported.

      Consciousness is not a reversible physical system, it’s an outcome of a biological process. I can’t imagine any biological process as reversible. Unremembering a sensory experience followed by unexperiencing the experience followed by unstimulation of sensory inputs is as nonsensical as reversing digestion—taking in waste products, embellishing them with nutritional content retrieved from throughout the body and reconstituting them physically as foodstuffs is an absurdity.

      So, why is digestion forward-time directional? Is saying that it “just is” unsatisfying? Am I missing some elusive deep meaning behind your question?

      Regarding the block-view and dynamic-view, you may have the impression I somewhat deprecate the latter but I don’t. I agree that both perspectives are valid. I’ve commented a lot about the block-view because I believe it’s important to recognize that it’s fundamental, has explanatory value and must be considered in our quest to understand ourselves in the universe. But the dynamic-view—the animation of the block-view by consciousness—is inescapably the view we live in. All of physics, the same physics that implies the block-view, is derived from the consistencies we find in the dynamic-view.

      Ultimately, consciousness is forward-time directional because of the arrangement of events on the block universe that give rise to consciousness.


      1. I think for me, the issue is that that the laws of physics are time symmetric, except for the second law of thermodynamics, and the wave function collapse (if it happens). Consciousness is a physical system operating according to the laws of physics. So why is it irreversible?

        If you’d asked me that before O’Dowd’s video, I probably would have just said entropy. But that feels a bit hand wavy, since the second law applies to isolated systems, and a functioning mind isn’t that. The correlation explanation focuses in and makes the relationship clearer. If you didn’t need that clarification, then well and good. Those of us who felt the lack now have a little bit more. Since our two explanations don’t contradict, I think we can both walk away happy?


        1. I’ll end up happy in any case … 😉

          True, consciousness is a complex physical system but it’s also a biological system. I think that places grave constraints on the viable reversibility of the laws of physics. Let’s reframe the question in simpler terms: Can the low-level physical processes of a single biological cell be reversed?


  5. Hi Mike,

    One element of the time’s arrow conversation not addressed by Dowd, and which I’m not sure he handled properly, is irreversibility. While the basic laws of physics may be time-symmetric, not all processes are actually reversible. Prigogine’s work, for instance, explored the fact that the emergence of complexity is closely tied to, if not predicated upon, irreversible processes. This comes into the conversation in terms of entropy, but one can have entropy generation via reversible processes, too, I believe. So I think there’s a relevance to including the irreversibility in this question of time and memory. It clarifies some things I think.

    Where I’m not sure Dowd got it exactly right is the cosmic ray leaving a mark on the asteroid, and how he showed that essentially unwinding in reverse. The ray passes through the asteroid in the opposite direction and the blemish or scar or glass that had previously formed in the material y the cosmic ray is undone. But I don’t think it works that way…? Some things would be possible in reverse, albeit highly unlikely, like the dust coming back together to form the asteroid, etc., but I don’t think we can reverse a cosmic ray and get back to the grain structure or crystalline structure of the original material.

    I don’t really know how this works with the block universe, but it seems that life in general, and perhaps the faculties of memory specifically, probably have the sensation of moving in a particular direction because of the irreversible processes on which they are based. Even if most of the rest of the cosmos is proceeding reversibly, and without any obvious directionality, irreversible processes do have directionality. I think those are the correlations Dowd spoke of.


    Liked by 1 person

    1. Hi Michael,
      I’m not familiar with Prigogine’s work, but scanning his Wikipedia page, it looks like it’s mostly about thermodynamic or entropic irreversibility. As I’m sure you know, that’s a statistical irreversibility. Any one process is, in principle, reversible, even if it very unlikely to happen, and the probability of it happening en masse is profoundly low.

      On the cosmic ray, I recall him using the word “scar”, but I know what you mean. If a strike did generate heat, then it wouldn’t be sufficient to just have the ray come through in reverse; every other emission in the process, such as generated heat, has to happen in reverse. It just makes the probability of anything like that happening even more infinitesimal.

      As far as I know, the only thing in physics that would be hard irreversible (as opposed to the extremely low probability of reversibility associated with the second law of thermodynamics) is a quantum wave function collapse. If there actually is an objective wave function collapse, then that would definitely represent a truly irreversible process, and it seems like it would undermine a block universe, although I guess you could still have a “growing block”.

      Or are there irreversible processes I’m overlooking?

      Liked by 1 person

      1. Hi Mike,

        I think Prigogine felt there are some classes of systems that display intrinsic irreversibility, or what he calls time symmetry breaking. Here is a quote I found from a paper about his work (emphasis added):

        “Transformed generators of motion, such as the Liouville–von Neumann operator for distribution unctions in classical mechanics, and for density matrices in quantum mechanics, break time-symmetry. Moreover, a product of transformed quantities is no longer a transformation of a product, which is not the case in unitary transformation for integrable systems. In other words, there is an intrinsic fluctuation in the transformed quantities in nonintegrable systems. Probability emerges not from supplementary approximations made because of a lack of knowledge, but rather as a dynamical consequence of resonance singularities in nonintegrable systems. Irreversibility is now formulated in a theory of transformations that expresses in ‘explicit’ terms what the usual formulation of dynamics ‘hides.’ “

        I think he showed or felt he showed that systems far-from-equilibrium, with persistent interactions and many degrees of freedom for energy sharing and transformation between modes—(hallmarks of living systems)—have dynamic properties that are not time symmetric.

        As to the asteroid strike, you may be right that everything is reversible in the sense that we can play the film backwards and recreate the original conditions, but this one is very counterintuitive for me. Some grain structures are path dependent in the sense that the rate of cooling of a material has an impact on the final structure. The asteroid could have formed as part of a very large mass that cooled very slowly, and thus had one grain structure, and a cosmic ray or a high energy particle could create local conditions quite different that produce a different structure in the material. I don’t think one can necessarily just “undo” the second structure by running the tape backwards because different conditions (such as a different cooling rate) may be required to obtain the original condition. I know what you mean about physical laws being time symmetric so I’m probably wrong on this; it’s just very damned counterintuitive!


        Liked by 1 person

        1. “I think he showed or felt he showed that systems far-from-equilibrium, with persistent interactions and many degrees of freedom for energy sharing and transformation between modes—(hallmarks of living systems)—have dynamic properties that are not time symmetric.”

          I think the very notion of freedom within any given system is an insurmountable bias for “physical law” pundits to overcome. We are conditioned by our immediate culture, a culture that does not recognize that all systems are living systems, with each system possessing a limited degree of self-determination within an otherwise deterministic system; the more complex the system the greater degree of self-determination.

          Physical law pundits squeal with exhilaration and forcefully expound how smurfy evolution is by pointing to the diversity and novelty we observe. They forget that evolution is a process and that this “process” is underwritten by a dynamic that makes novelty and diversity possible in the first place. And that dynamic is living systems that are free to express their own unique structural and qualitative properties with other systems. Bias and bigotry rules in the kingdoms of men.


          Liked by 1 person

        2. Hi Michael,
          Appreciate the quote but I find it somewhat obfuscated. He does mention quantum mechanics, and assuming he holds to a collapse interpretation, that would be true irreversibility. (Density matrices, in and of themselves, like wave mechanics, aren’t irreversible in principle. We need a reduction or wave function collapse for that.)

          Much of the rest still seems entropic to me. To be clear, while systems which produce entropy are reversible in principle, in practice they’re effectively irreversible. We might unscramble an egg if we make a massive project out of it, but we’ll create a lot more entropy in the process than initially created by the scrambling.

          On the asteroid, the thing to remember is that you can’t do it with just the asteroid itself, even in principle. Too much of its condition resulted from the environment. So reversing that condition would take opposite processes from the environment. It’s unscrambling the egg on an astronomical scale.

          Which is to say, your intuition, in practice, isn’t wrong.

          Liked by 1 person

          1. Hi Mike,

            Agreed the quote has a bunch of jargon related to Prigogine’s research program. I just want to note that when you say, “Much of the rest still seems entropic to me,” that it doesn’t sound like the more profound elements of Prigogine’s research have quite sunk in. So I want to take a crack at conveying what I think he has accomplished, or at least suggested can be accomplished, and we can use your conversation above with Stephen Wysong to make this clear I think.

            You asked this question, “Can the low-level physical processes of a single biological cell be reversed?” and then answered it, “In principle, yes.”

            I took this rhetorical exercise to imply that (aside from wave function collapse) nothing that happens in nature cannot be reversed, in principle. While it would be very unlikely to occur spontaneously, it could. But I think if Prigogine is right there are processes that occur in the universe for which reversibility is simply not possible, both practically and in principle. His technical writing is certainly over my head, but summaries of his work suggest this was the culmination of his research program.

            And for me, this is a fairly stunning discovery. I mean… it suggests that part of what makes living systems living, is that the processes of which they are composed break time-symmetry. If this is so, the fundamental reason one would have memories of the past and not the future, is because there is inherent directionality to the processes of life. I think this is a stronger argument than correlations, although they could be said to be related. What correlations don’t explicitly involve are two things, which may ultimately be the same: transformation, and the production of complexity, or the ramification of wholeness.

            How does this happen? I’m not a mathematician so I’m going to try and paint a picture using my intuition. What we know about the conditions in which this breaking of time symmetry occurs (from Prigogine’s writing) is that, as I said earlier, it involves systems with many modes and degrees of freedom that experience persistent interaction; that “resonance singularities” arise out of which the system emerges into new “branches” or modes; and lastly, that the transformations that occur tend to result in some degree of coherence. This is how we build complexity I think. The initial system is “transformed” by passing through this resonance singularity into something new. And this passage, so to speak, must involve the breaking of time symmetry.

            This is more profound than getting nicked by a passing pebble of dust. This is the generation of a new and dynamic continuum of order that is its own reference point, in a sense. The system, having achieved a coherent synthesis, is more than the sum of the parts. It is not equivalent to the sum of the parts because it is bound into a singular whole through mutual information—like James might say. But this mutual information is shared throughout the system and does not exist explicitly in one part or the other. It exists only at the level of the system as a whole. And this is why I think the formation of these structures breaks time symmetry: because the newly formed system is not itself time symmetric. It cannot be run backwards. And it creates new information that may further be built upon.

            If you take a very simple example like a Rayleigh-Benard cell, you cannot run it backwards unless you change the direction of gravity. At least I don’t think so. And unless I’m mistaken, when we say a process is reversible we don’t mean that gravity changes sign, or the charge of the electron reverses, etc. (I don’t think so anyway.) But to the example, when you heat water at the bottom of the pan, the density difference between the hot water at the bottom of the pan and the cool water at the top in the presence of a gravitational field creates the circulation patterns we’re familiar with. But if the pan of water were to run in reverse and absorb heat from the air at the top, and then cool at the bottom, there would be no driving buoyancy difference to foster circulation. The system would never “run.” This simple example shows that spontaneous order formation involves a symmetry breaking. And I think this symmetry breaking, particularly with respect to time, is the best answer we have to date for why our memories and our lives have directionality.

            There is certainly entropy involved, but there are processes that involve entropy formation that don’t have the characteristics of these non-linear, far-from-equilibrium systems. Expanding steam through a steam turbine increases entropy while producing electric power. But it’s a class of system that in principle could be reversed. Unlike the convection cells. There is no mode shift through a resonance singularity that yields a novel self-resonating structure at the expense of the degrees of freedom of the original constituents.


            Liked by 1 person

          2. Hi Michael,
            I appreciate the effort at clarification, but a lot of the terminology is still stumping me. For example, the phrase “resonance singularity”. I know what the individual words means, but the phrase translates as “deep failure of mathematics” or perhaps “deep failure of theory”, which strikes me as more of an epistemic issue, perhaps related to limitations explored in chaos theory? If so, then we’re still talking about systems reversible in principle, but not in practice.

            I will say that it’s not going to be that hard to stump me on how a particular complex system could be reversed, even in principle. But stumping me still leaves the general consensus that the laws of physics, aside from the second law of thermodynamics (and possibly wave function collapse), are reversible, and I would think physicists would have taken account of any work that demonstrated true irreversibility in principle.

            (One of the things reading Strevens’ book has made me more aware of, is that scientists make claims in their popular writing they don’t (can’t) make in their scientific papers. So their strictly scientific work can be accepted, and even rewarded, but we should be cautious in transferring that esteem to their looser musings.)

            I will note that in the case of a process involving gravity, gravity imparts kinetic energy to what it’s affecting. If we started something lower in a gravity well and it had the same kinetic energy as something that had fallen to that point, but in reverse, then we’d get the reverse movement. Gravity is interesting however, because I’m not sure how we can talk about reversibility in the case of black holes, but that gets us into the black hole information paradox stuff.

            Overall, I don’t think we should get too worked up about these “in principle” situations, since most of them, due to complexity and chaos, not to mention things like the speed of light limit and universal expansion, are effectively impossible. This seems to make the distinction largely academic.


  6. Hey Mike, thanks for this post, video. I think it provides an actual useful insight I didn’t have before, and it’s based on the cosmic ray scarred rock. If you’re just given the rock and no indication of the direction of time, there are two possibilities: 1. The scar is correlated with a cosmic ray in the future which will erase the scar, or 2. the scar is correlated with a cosmic ray in the past. Correlation isn’t just yes or no. It comes with a quantity between 0 and 1, and the quantity associated with situation 1 (ray in future) is extremely close to 0, whereas the quantity for situation 2 (ray in the past) is high. So in general, we can associate high correlations with one of the directions of time, and call it the past, and call the correlated items “memories”.

    [and pretty sure correlation = mutual information]

    Liked by 1 person

    1. Good point about correlations being probabilistic. It seems like they become more reliable if they’re part of a repeating patterns, which makes it increasingly likely they’re actual correlates of some past event. And future correlations, by their very nature, can’t achieve anything like that reliability.

      I totally agree that correlations are mutual information.


  7. The more interesting question might be why do we falsely remember the past?

    When we realize that false memories are possible, possibly even frequent, then the idea that a memory is simply a recording falls apart. Even when we examine the apparent cosmic ray scarred rock, we are dealing with interpretation. Could the scarring been caused by something other than a cosmic ray? Could be it a false record?

    Records themselves exist in time and are subject to alteration. They can vanish. The library of Alexandria burned. If the rock breaks apart, the record of the cosmic ray could be lost or interpreted as something else. False records can be created. Patterns on sand may look to be caused by water but could be caused by wind. The Hitler diaries were briefly history before becoming fraud.

    Liked by 1 person

    1. Good point.

      I think we falsely remember because episodic memory is a reconstruction, a simulation of a past event based on individual semantic memories, many of which have gotten tangled with later impressions. It’s why we’re pretty decent at remembering recent events, but the further back we go, the less reliable it becomes, mainly due to new correlations tangling with the old ones.

      Why did evolution do it that way? Because the real adaptive functionality is prediction, and keeping detailed accurate records, at least for most organisms, isn’t worth the energy investment.

      But as you note, even records themselves are subject to destruction and degradation. We’ve lost most of what the ancient Greeks wrote. What we have is mostly what Islamic and Medieval Christians thought was worth preserving, with an occasional exception like a copy of Lucretius being found in a random archive.

      But in principle, if you could find all the correlations and reverse engineer them, and then reverse engineer the new correlations uncovered, and then reverse those, etc, you could know all the events. Of course, in practice this will never be possible.

      Liked by 1 person

  8. Hi Mike

    I usually resist commenting because since I am slow it is too ‘time’ consuming and my biological clock is beginning to ran out of time as well. So, just some more or less random observations.

    I suspect the reason that there are so many different theories, approaches and conclusions is that there is a great degree of incompleteness in all the theories of time.

    Hameroff and Penrose conceive of consciousness arising as a result of processes in which innumerable clocks are operating, even at the quantum level, throughout the body.When all is put together, each of us represents a unique biological clock, ie a coordinated system in which all the parts have to be in the same time zone. This is obvious since each of us is made up of slightly different materials.

    So, when a photon hits an opsin they will combine if the wavelength – a time keeping device – is the right one. The rotating earth is a clock to which our bodies respond with their diurnal rhythms. Timing thus determines how we feel!

    The solar system system is also a clock and it directly causes our seasons. Thus one can work all the way back to the Big Bang which supposedly started the first clock running.

    All systems thus require a time keeping device or devices. Time, evolution and change thus are of the essence. Time is real and it is complex. How else could we ever detect the patterns within our selves?

    Liked by 1 person

    1. Hi Liam,
      Never be afraid to take your time on weighing in. But I understand completely. I avoid commenting on many popular blogs or news sites because the discussion is just too frantic.

      I’m with you on a clock basically just being a consistent pattern that we compare events to. Interestingly, cosmologists often talk about the overall matter density of the universe being the ultimate clock, which is enabled by the ongoing expansion of space. Although the discrepancies in the measurements of that expansion is starting to call into question how reliable of a guide it might be.

      I also agree that time is real. Although that doesn’t necessarily mean it’s fundamental, not emergent from something else.


Your thoughts?

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.