I have sympathy with this view: "What if causation is a multidimentional, convoluted, worse yet, fractal thingie that defies all methodologies trying to trace it with a finger like a crack on the wall?" Yes the web of interactions in the real world is very complex. But it's not fractal: there seems always to be a lower level where in small enough domains the essential causal interactions are linear. Their outcome depends on context, but nevertheless we can understand the elements of causation by looking at such small local systems.

It is when you put them together to get really complex interaction networks that things get really complex: but even then there are identifiable hierarchical structures and network motifs that let us understand much of what is going on. It is in this context that we can reliably identify both bottom up and top down elements of causation.

So I am not as pessimistic as you.

George Ellis

lol good for you, George!

But I seriously think that information is fractal (and information is related to causation). I base it on my observation of the real world, when I was trying to understand some strange phenomena. To me it appeared that an event had many consequences, large and small, near and far, just as I said above. And I saw those consequences (I am a visual type) as paisleys of various sizes making up a flowery pattern on a fabric. And the interesting thing I saw was this: before an event arrives (a big paisley), there are many small paisleys (and of course a few medium size ones) that arrive before it, in a way, announcing the arrival of the main event. They do it many times, at various times. And they run in streaks, like it befits a fractal thing proper. Likewise, after the main event, there are many "aftershocks", large and small, running in streaks. Then a streak changes as if madam info got tired of her tune.

I hope I made sense. This view of events as paisleys on repeated patterns, large and small, made me think that info is indeed fractal. And information, you must admit, is related to causation.

Also, when we think of causation, we tend to oversimplify and consider real only the obvious things, like, 'heating up water causes it to boil'. Usually, what's left out are things like, why exactly did mom put the kettle on the stove. Was it because she wanted some tea? Or because she expected a company? Maybe a habit; she always does it around that time. Even when a person thinks that he or she had a clear, well defined intent, in reality it is virtually impossible to trace the "intended action" to something concrete. As long as an action appears reasonable in a given context, one can always find a reasonable explanation. The trouble is, reasonable explanations are rarely right. Worse yet, they rob us of our illusion of free will.

I find the first part of your response a bit mysterious: it may be to do with the mind rather than physics.

I very much like the second part about the kettle: it raises the real deep issues. The thing that's puzzling is how these deep questions relate to physics: how can we take the physics seriously, not underestimating it in any way, but not also not trivialising the deep issues? -- in particular, not trivialising life, consciousness, and the issue of free will. That's the real challenge that underlies this whole discussion.

You may have missed the post in this thread by J. C. N. Smith on Jul. 20. It's worth reading.

George Ellis

GE:

Yours was an interesting and informative essay.. As a newcomer to the FQXi community, I feel few of the "community" grade, or even look at, my essay which approaches the problem very realistically, based on an internal view.. Might you look at it, comment if so inclined, and grade it?

To Seek Unknown Shores

聽聽 http://fqxi.org/community/forum/topic/1409

Thank you

TE

    Yes, thank you, that was a very good reading.

    About my mysterious response, I was just in a playful mood last night which continued into this morning (hope you don't mind). What I find mysterious is that I made both posts above _before_ I followed your lead and read "Patterns in the Fabric of Nature" by Steven Weinstein, which brings up the complexity of non-locality of causation and the question of free will. Coincidence?

    It is tempting to see my largely unconscious actions (2 posts above) as precursors to the event about to unfold (reading Steve's essay), as if the arrow of time can be reversed. Since I do not believe in the reversal of the arrow of time, I see this sort of coincidences as the proof of the fractal nature of information (and causality), which I tried to convey above. How else can this be explained?

    Dear George Ellis,

    As a relativist who has written about the structure of spacetime, how do you get causality out of block time? You don't mention block time in your essay, but it would seem from your bio that you think Minkowski spacetime is right, and that leads unavoidably to block time.

    I've argued in my essay that if block time is right, and motion through time is an illusion, then the laws of physics, and crucial principles such as cause and effect, would have to exist within the illusion, because they require motion through time.

    I've also argued that the two levels of time we seem to find, block time and motion through time, can't possibly co-exist as simply part of the nature of the time dimension, as many assume. Given the intrinsic unpredicability of quantum events, the two levels disagree over whether the future already exists. This means only one of these levels can be real - I've examined both possibilities.

    Block time is one of these two possibilities. But it implies that you can't assume the spacetime interpretation of SR is right, and also have causality. This is one of many problems with block time that people often ignore. Because the time dimension is taken to be different from the other dimensions in Minkowski spacetime, people sometimes assume that the nature of the dimension somehow solves the problem. I've tried to show that this can't be the case. Do you have a way to get causality out of block time?

    Best wishes,

    Jonathan Kerr

      Dear Jonathan Kerr,

      that is not the topic of this essay or thread. However I have just put an extensive paper on the archive [ink:http://xxx.lanl.gov/abs/1208.2611]

      Space time and the passage of time [/link] looking at the issue in detail. In brief: there is no problem if one has an Emerging Block Universes (EBU).

      George Ellis

      Dear Pentcho Valev

      Your comments would carry much more weight if you chose to omit the insults and sarcasm.

      You claim "It is tautological - you just define as "top-down causation" influences that go from what you see (or mankind has defined) as higher hierarchical level to what you see (or mankind has defined) as lower hierarchical level".

      There is indeed an issue here: how does one define higher and lower levels? That has to be done depending on context: it is quite different in the physical sciences, the life sciences, and artificial sciences. It can be sensibly done, as is shown in various articles referred to in my essay (indeed one cannot understand complex systems without categorising such levels: see for example Tannenbaum's book on digital computers for that particular case).

      Given that context, this is a sound definition. The scientific issue is whether there are instances of existence of such effects. My claim is that there are indeed many examples showing this is the case, across the sciences. There is much evidence to support this claim.

      George Ellis

      Dear Pentcho Valev,

      "Roughly speaking, in complex systems anything affects anything" Yes that certainly is roughly speaking and unenlightening.

      In biology and in computer systems, the key feature leading to real complexity is existence of Modular Hierarchical Structures, where each word is important: see particularly the books by H A Simon and by Grady Booch that I refer to for an enlightening discussion. Indeed you can't understand biology without taking this kind of structure into account (with cells being the key modules on which it all hinges), see Campbell and Reece for example, nor can you understand digital computers without taking it into account, see Tannenbaum. Example: for the hierarchical structure of genetic modules, see the book "Modularity in Evolution and Development" by Schlosser and Wagner. Then there are various interlocking hierarchies in the brain, even though there is no specific "top", see for example Leergard, Hilgetag and Sporns, "Mapping the connectome: multi-level analysis of brain connectivity" (Frontiers in Neuroinformatics, Volume 6 Article 14, May 2012).

      So there certainly are hierarchical systems out there. It is then a key issue as to how causation works in such systems. That is what my essay tries to address. Again I particularly point to Booch's book at characterising clearly core features (such as inheritance and information hiding) that are needed for complexity to emerge .

      You do not find it profound.Fine, point me to something that is indeed profound and clearly illuminates how genuinely complexity emerges on the basis of the underlying physics (and please don't just give me the statistical properties of complex networks: even if they are useful in some ways, they simply fail to get at the essence of what is going on. Uri Alon's book on network motifs gets much closer to the real dynamics).

      George Ellis

      Dear George,

      The issue is unavoidably relevant to your essay. Your essay discusses causality, which you admit in your arXiv paper can only exist if major adjustments are made to the standard view of spacetime. And spacetime is one of your fields. I'm sure you'll see the need to explain how the subject of your essay works in relation to this, as it's entirely dependent on it - and we're meant to be looking at the foundations.

      In your arXiv paper you've argued very well for a flow of time. I call it motion through time, but we agree absolutely that it must exist. You've shown Barbour and others to be wrong in the idea that motion through time is an illusion, as I have in other ways. And you've argued, as I have, that standard block time is wrong. That means we both think the Rietdijk-Putnam argument, which is a rigourous proof that a fixed future comes out of spacetime, is wrong. Perhaps you'd explain why you think it's wrong, I've done that by suggesting the flaw is in the assumption that simultaneity across a distance has meaning (beyond the light cone).

      You've also come up with an excellent way of hooking the quantum randomness up to the large-scale world in a Schrödinger type way, but using it to show that the future is unfixed. That makes a neat distillation the issue of quantum uncertainty versus the block time fixed future, which is a key part of my argument as well.

      So we agree on a lot. We only disagree on whether the spacetime interpretation of SR can be kept, given this need to adjust the block time picture - we agree that SR itself is right. You're one of a number of people who (comparatively recently) seem to have come round to the idea that time must flow, who've then tried to bring that idea into block time, while keeping spacetime much as it is. This risks being 'cake and eat it' - you might have to choose. Ideas such as the 'crystallysing block universe' are similar to your EBU, or emerging block universe. Basically, the idea is that the block sets as time moves through it.

      I can show a major weakness in this approach, and I see it as an attempt to 'fix-up' the spacetime interpretation, when it simply may be the wrong interpretation. The crux of the Rietdijk-Putnam argument is that an event can be in the past to one observer, but in the future to another. If that idea is wrong, block time falls apart completely. I've argued that it's wrong (because it depends on long range simultaneity).

      But in your language, or that of the EBU, this means that to one observer an event has already been frozen into the block, while to another observer it hasn't. The question of which events have gone into the fixed part of the block is observer-dependent. As you probably see, this greatly weakens that whole approach. How then are we to deal with the physics of how an event gets frozen into the block? Some say recently that the collapse of the wave function may be the process of an event going from future to past. But again, if this is entirely observer-dependent, and depends on how one is moving, then it doesn't look like a physical process. And that problem is exactly what led to standard block time in the first place, along with illusions and all.

      I'd appreciate your thoughts on this, and any comments on my essay, thank you.

      Best wishes, Jonathan

      PS. I see the 'crystallising block universe' and the 'emerging (originally evolving) block universe' come from the same place. The question is the same about all - how they get round the problem of the fact that which events have been crystallysed is observer-dependent (frame-dependent), and therefore seems to reside in the observer's perception if spacetime is right, just as with standard block time. JK

        Dear George Ellis,

        Since you are calling yourself a relativist, I doubt that you are open for what the topic of the contest demands: "Questioning the Foundations". Doesn't this include Cantor's naive set theory, Einstein's special theory of relativity, Lorentz covariance, a priori existing Parmenidean spacetime, etc. too?

        While I was initially fascinated by the many reasonable views you uttered, my doubts in correctness of relativity rose each time I read an essay or an other paper of you. I nonetheless accepted that there are many arguments in support of relativity.

        Situation has suddenly changed after I tried and managed to understand an experiment by Norbert Feist. See Fig. 5 of my recent essay. I fear I cannot expect an other factual reply than silence from anybody who has to fear loss of reputation or who is simply lazy.

        Do not get me wrong. I do not exclude that some relativity-related theories are about as useful approximations as is according to Ebbinghaus/Lessing Georg Cantor's naive set theory which was based on an obvious error.

        Please do not feel hurt. You might understand my motivation and uncompromising rudeness when I tell you that some years ago a Hendrik van Hees blamed me for damaging the reputation of Otto-von-Guericke-University Magdeburg after I suggested that the ear performs a cosine rather than Fourier transformation. Fortunately my boss declared the matter undecidable because utterly foundational. MP3 works.

        Respectfully,

        Eckard Blumschein

          Dear Jonathan

          "I'm sure you'll see the need to explain how the subject of your essay works in relation to this, as it's entirely dependent on it - and we're meant to be looking at the foundations." It works by taking into account what I say in my arXiv paper on the passage of time.

          "The crux of the Rietdijk-Putnam argument is that an event can be in the past to one observer, but in the future to another. If that idea is wrong, block time falls apart completely." But I argue in my paper that what observers think is past or future on other world lines does not matter. What matters is (a) what happens on their own worldlines, which must have a proper temporal ordering, and (b) what happens in terms of interactions between events on different worldlines, which are mediated by timelike and null curves. Spacelike surfaces and instantaneity do not enter into it. Finally (c) there are preferred worldlines in spacetime, as I explain, so in fact the Lorentz symmetry of the theory is broken in realistic solutions of the equations.

          In any case for the purpose of the present essay, what I need is that causality works in local situations as described by ordinary physics, with a well-defined flow of time as embodied in the standard equations of physics such as Newton's laws and Maxwell's equations and the Schroedinger equation, see the Feynman Lectures on Physics. All the evidence for ordinary physics shows this is true, indeed physics as we know it would not exist if it were not so. That is all I need for my essay.

          George Ellis

          Dear Eckard Blumschein

          " Since you are calling yourself a relativist, I doubt that you are open for what the topic of the contest demands: "Questioning the Foundations". Doesn't this include Cantor's naive set theory, Einstein's special theory of relativity, Lorentz covariance, a priori existing Parmenidean spacetime, etc. too? "

          I chose to deal with a particular topic that I regard of importance, and you state you doubt that I am up to what the competition is about because I did not deal with a whole set of different topics. What is the point of this gratuitous insult?

          "Situation has suddenly changed after I tried and managed to understand an experiment by Norbert Feist. See Fig. 5 of my recent essay. I fear I cannot expect an other factual reply than silence from anybody who has to fear loss of reputation or who is simply lazy." This kind of accusation does not motivate me to read your essay or respond to it. On the contrary.

          "You might understand my motivation and uncompromising rudeness ..." I have no intention of responding to any further such rudeness. It is incompatible with my attempts to have cordial collegial discussions on the topic of my essay.

          yours sincerely

          George Ellis

          Hello George,

          Thank you for your reply. I didn't mean there was any need to justify your essay, I believe in causality too. I was just trying to show the relevance of the block time issue, and of bringing it into the discussion on this page.

          You say that in the EBU "what observers think is past or future on other world lines does not matter". I'm sure it doesn't in many ways, but the point I've made it that it matters if we're trying to pinpoint the crystallation of an event, and look at what could cause it in that context. It's all very well saying that an event goes into the fixed past because of the collapse of the wave function. But if to some observers it already has, while to other observers moving differently it hasn't yet been crystallised, then we end up with what looks like a perception-based thing, and exactly the same setup that led to the problems of standard block time in the first place.

          I think it's very good that you're argued for a flow of time, or motion through time, as strongly as you have. I'll look at your work further, and will almost certainly refer to it in mine. People are coming round to the idea that block time is wrong, partly because it seems that something has to give if we're to get to quantum gravity. To me, a slight tweak to spacetime is not enough, and the problems with it do not wash out so easily, as I've shown. I think we need to face up to the fact that we simply have the wrong interpretation in front of us. Spacetime looks right, and it has helped us simplify a lot of theories. There's a great reluctance to reverse out of the cul-de-sac. But it's an interpretation, and if one gets conceptual problems with an interpretation, then one probably needs a new one.

          I'd appreciate any thoughts you might have on my essay,

          Best wishes, Jonathan

          You mentioned that the issue was whether algebra can deal with a Modular Hierarchical structure, and the need for graph theory. Just wondering if there are top down aspects to a supernova, just to have a nice physical example to chew on, especially for being so 'event-like', and anthropic questions ? If so, it seems like there is more going on than modular hierarchy and graph theory, like why can't elements build up without kicking leptons out of the neighborhood.