Thanks for the read Bill! I'm not sure why those two things would be contradictory in your mind. But I suspect it might have something to do with how I'm using the word "emergence" and what you associate with it.

It's worth noting that emergence can be used semantically in two ways (above you can see George Ellis's comment about this as well). You can use it to say "the patterns emerged from the simple chemical interaction." In this manner it usually means getting something complex from something very simple: it's fundamentally historical. This isn't the usage herein.

There is another way to talk about emergence. For instance, if I had a bunch of NOR logic gates, and I hooked them up to make a complicated circuit that enacts many different kinds of logic functions (like ORs and ANDs and NOTs), you would say that the circuit and other logic functions "emerged" from the underlying NOR gates. This is the way I use emergence in the paper.

Since you asked, here's a brief summary of causal emergence (takes in deep breath): the causal structure of systems can be treated mathematically as a communication channel over which states are sent over time (much like sending messages), and it turns out that describing/observing/intervening upon the system in terms of a higher scale can actually make the channel transmit more information because these higher-scale descriptions are a form of channel coding. Whoof!

Thanks very much Wilhelmus. I didn't explore anything down at the quantum level for my own essay. To me it seems that relying on quantum effects to explain agents is like trying to solve a single mystery by combining two mysteries, which generally just makes everything even more mysterious, but I'm eager to read your essay and find out. I will check it out there.

All the best,

Erik P Hoel

Thanks George. I definitely agree that emergence can be taken in a historical sense (evolution, development, complexity from simplicity) and a level sense (function, scale, causation). It's only in the latter sense that it involves issues of supervenience.

I had a question about your comment: what exactly do you mean by "it is the higher levels that decide what will be done and the lower levels that carry out the work?" Can you explain that a bit more? It may illuminate some of our differences in approach. Because if the lower levels are carrying out the causal work in the system, didn't the lower levels really make that previous higher level decision?

What we are saying is in a sense the opposite of this: causal emergence is only when a macroscale outstrips the microscale in terms of information and causal work, so proving that causal emergence occurs involves directly assessing the causal structure at both the microscale and the macroscale and comparing the two. I think it's really helpful to use simple but well-defined systems like Markov chains for this exact reason: you can derive all the possible supervening scales along with the full causal structure.

All the best,

Erik P Hoel

Dear Erik,

Thank you for your well written and detailed essay. I thoroughly enjoyed reading it. As others have commented, I particularly agree with your analysis on emergence across multiple levels of scale and particularly liked the way you tied them all together in your conclusion, "purposeless microscale descriptions are like a low dimensional slice of a high dimensional object". I voted on your essay a few days ago, but just thought I'd give you a more detailed reply on how much I enjoyed it.

Regards,

Robert

    Thanks so much Robert! Very nice to hear it. Just finished your essay - I greatly enjoyed your breakdown of Maxwell's Demon, and it got me thinking about the long history of cases of trying to get something from nothing. In Maxwell's case, it's a violation of the 2nd law. I think for a long time people have thought of emergence in that way - it's almost like getting something from nothing, because how could you possibly gain any information or causal work going up to a macroscale? It seems like squeezing something from nothing, and I think it is this that's the intuitive force behind the "exclusion argument." But there's a few cases where some people have figured out how to squeeze something from nothing (metaphorically, obviously). One of those is Claude Shannon's noisy-channel theorem. At first it really seems a really noisy channel can only transmit very low amounts of information. Then Shannon showed that through channel coding the information can be radically increased - without altering the channel! By saying that causal emergence comes from treating a system's causal structure as a channel, and that macroscales are encodings for the channel, I'm piggy-backing on Shannon's "something from nothing" proof. So causal emergence is kind of like getting something something from nothing (without altering the system).

    Anyways, just wanted to let you know your essay inspired me to think about it with a new analogy.

    Thanks so much!

    Erik P Hoel

    Dear Erik Hoel,

    just rated your essay and gave it a high score. Your concept of causal emergence is intriguing and you should further investigate it. It also poses interesting teleological questions.

    Best wishes,

    Stefan Weckbach

      Thanks so much Stefan, I'm so glad you found it intriguing. I definitely plan on investigating it further (when I find the time!). In terms of teleology, I think you're right. However, I'm always wary of those kinds of words, so my own personal stance is to try to explain what looks like teleology (apparent teleology) without coming to overt metaphysical conclusions.

      Thanks so much for reading and rating!

      Erik P Hoel

      Hi Eric P. Hoel,

      I offer a complementary suggestion. In addition to pushing along trying to elaborate on the usual assumptions, you might also pause and see what challenges those underlying assumptions. I have an essay that introduces some of the challenges facing the scientific vision of life,

      http://fqxi.org/community/forum/topic/2783

      If nothing else it might introduce some additional puzzles to mull over.

      I hope things are going well for you.

      Ted Christopher

        Thanks so much George! Actually, Larry was on my PhD thesis committee at UW-Madison. He does excellent work.

        There's a handful of analytic philosophers who have thought about these issues, starting with Yablo. There's also List and Menzies, as well as Shapiro and Sober. All these people do incredible work and have all touched on issues related to causal emergence at some point or another, although most are focused more on problems of mental causation. None have, as far as I know, argued explicitly for the theory laid out here and elsewhere.

        One constant problem that I have with this research is the consequence of framing it in terms of the exclusion problem. It's a good way to frame it because it hammers the problem home, but it's a bad way because the exclusion argument is a well-known philosophical issue and people then immediately assume this is a philosophical solution to a philosophical problem. But as I indicate in the essay, I'm using the exclusion argument as a stand-in for a more general issue concerning causal structure, information, and model choice.

        Ultimately, I think this requires a scientific (or mathematical) theory, composed of: A) formalizing supervenience as changes in scale or as highlighting only subsets of the system's state-space; B) some sensitive measure of causation and/or information (I've used information theory and Pearl's causal calculus) that can handle things like noise, is proven to be related to various important causal properties, doesn't give nonsensical answers for simple scenarios, etc; C) actually checking and proving that B can be higher across various scales made with A; D) explaining why it's theoretically even possible that the macro can beat the micro; E) hopefully some applications.

        Originally we argued in 2013 for the D that macroscales reduce the noise in the system (over both the past and the future), and that's how causal emergence occurs. I think there's another interesting way of framing it, which is that macroscales can be thought of as codes (as I argue here and elsewhere), and the macro can beat the micro because of Shannon's noisy-channel coding theorem. Hopefully both these explanations help with E: actual applications.

        Dear Dr. Erik P Hoel,

        Please excuse me for I have no intention of disparaging in any way any part of your essay.

        I merely wish to point out that "Everything should be made as simple as possible, but not simpler." Albert Einstein (1879 - 1955) Physicist & Nobel Laureate.

        Only nature could produce a reality so simple, a single cell amoeba could deal with it.

        The real Universe must consist only of one unified visible infinite physical surface occurring in one infinite dimension, that am always illuminated by infinite non-surface light.

        A more detailed explanation of natural reality can be found in my essay, SCORE ONE FOR SIMPLICITY. I do hope that you will read my essay and perhaps comment on its merit.

        Joe Fisher, Realist

        Hi Ted - thanks so much for stopping by. I strongly agree that we all make assumptions.

        I checked out your essay and was very glad to see you mention Rafael Yuste - he's my principal investigator here at Columbia University. I did want to say that, while I disagree with some of your examples seriously challenging contemporary neuroscience, I absolutely agree with you that little attention has been paid in neuroscience to the consequences of hydrocephalus. IF it's true that people are operating normally but have drastically reduced gray matter (such as 10 to 20%) we're going to need to drastically rethink some things. However, if I remember correctly recent research has questioned these numbers.

        http://blogs.discovermagazine.com/neuroskeptic/2015/07/26/is-your-brain-really-necessary-revisited/#.WMVsVRLyuRs

        Thanks so much for reading!

        Erik P Hoel

        Hi Erik

        Your essay is awesome, you basically crushed it. The point about the kinetics of the system and the signal propagation time setting scales for "identity" was the first point that I found truly insightful, it reminded me of the idea that if we wanted to imagine something crazy like the universe being one big mind then it would never be able to actually finish a thought because it is expanding faster than it is possible to send signals back and forth across its entirety, and probably can't cross the percolation threshold for correlations as a result. The other point I loved was about needing to include the environment in Romeo's causal structure. In the phenomenological tradition, it was clear since Husserl that "consciousness" can only ever be "consciousness of phenomena", it's only really since philosophy of mind took this very ahistorical turn against "reductionism", I feel, that we lost sight of this fact. All of your other claims are correct, lucid, and I think should be uncontroversial for anyone familiar with modern neuroscience, information theory or stochastic dynamics. But it's exceptionally well-argued and clear. If you have a moment to look at my own entry at any point, I think we make many of the same arguments though I wrote in a somewhat different style. In any case I'd appreciate any feedback you could offer.

        Again, total slayer of an essay.

        Joe

        Thanks so much Joe - highly appreciate it. Although I wouldn't say any of this is "uncontroversial"! Try getting funding for it hahaha.

        I just finished your essay, which I enjoyed, particularly your writing. There's lots of stuff going on in there but just wanted to mention the relationship to my own essay here, which is your segment on higher-level explanations. You say: "In this way, we compress our explanations of phenomena, with the useful result that they can be communicated and shared with fewer bits, thus requiring less work to understand."

        I completely agree - this is totally necessary for human communication, or something like science where we communicate facts or data to one another. What I argue in my essay is that there's another possibility for an information theory metaphor beyond just compression for these types of cases: coding. So higher-level explanations aren't always *merely* compressions, sometimes they are also codes; in addition to being compressed, they also error-correct, meaning they can have in theory more information than whatever underlies them.

        Thanks for commenting, and I enjoyed your essay,

        Erik P Hoel

          Dear Erik,

          I also find your essay excellent and very stimulating. A few questions.

          The emergence of (apparently) purposeful behavior, intentions, goals, agency, is clearly a complexity booster, and it would be great if we could see it at work even in discrete, finite, *deterministic* formal systems, below the level of biology.

          In my mind, one of the best examples of emergence is represented by the 'digital particles' of cellular automata (Conway's GoL gliders, Wolfram's ECA 110 trajectories), and you do mention them in your essay. However, those are completely deterministic systems, while the causal emergence that you formalise in terms of input state and output state distributions, their correlations, mutual information etc., seems to fundamentally require a stochastic basis. Is there a way out to this difficulty? Can the theory be extended to formal, algorithmic, deterministic models? That would be very useful since in that context, as far as I know, emergent phenomena cannot yield more two-levels -- a limitation pointed out also by G. Ellis (I can't remember where, but earlier than in his current essay).

          The idea of substrate independence, or multiple realisability, is exciting, and provides a new (at least to me) perspective for looking at the whole business of emergence. However, while I find the example of waves quite appropriate -- with laws that are independent from the nature of the molecules in the medium -- I am not convinced that cellular automata are a good example. In the Game of Life, localised structures emerge only from a very specific underlying rule: should you change a single bit of that rule, those structures would vanish. What is true is that the digital particle, say a glider, is 'implemented' by an ever changing subset of underlying cells, but I don't think this counts as 'substrate independence..., or does it?

          To me, the idea that the fully-fledged causal structure of an open system may require the participation of the environment has appeared intriguing at a first reading but ... obscure at a second one :-] How should one interpret the red transition arrows in Figure 3? What is the formal meaning of 'outsourcing causal structure' to the environment, in terms of state transition systems? Should I imagine that some of the states of Romeo's brain are out in the environment. In which sense?

          Finally, I'd be curious to check the possible applicability of causal emergence to the very fundamental model of causal sets (Sorkin, Bombelli, Rideout, Dawker, etc.), which is essentially based just on causal relations between unstructured events. Any idea?

          Thank you and all the best!

          Tommaso

          http://fqxi.org/community/forum/topic/2824

            Thanks so much for the close read Tommaso! You ask some really great questions - so forgive the lack of brevity in my replies!

            "one of the best examples of emergence is represented by the 'digital particles' of cellular automata (Conway's GoL gliders, Wolfram's ECA 110 trajectories), and you do mention them in your essay. However, those are completely deterministic systems, while the causal emergence that you formalise in terms of input state and output state distributions, their correlations, mutual information etc., seems to fundamentally require a stochastic basis. Is there a way out to this difficulty? Can the theory be extended to formal, algorithmic, deterministic models?"

            Actually, causal emergence can work in completely deterministic systems (see Hoel et al. 2013 "Quantifying causal emergence" in PNAS for some examples). I mention this in the essay but only very briefly in the technical endnotes (section D). What is necessary is that these deterministic systems are degenerate - that is, multiple states can deterministically lead to the same state. As far as I know, a lot of CAs (in fact, most) fit this bill, so this could be comfortably applied to CAs. However, for totally time-reversible systems, if you take that system as a whole (the entire CA board, as it were) causal emergence cannot work. I also address this objection in the endnotes (section D).

            "That would be very useful since in that context, as far as I know, emergent phenomena cannot yield more two-levels -- a limitation pointed out also by G. Ellis (I can't remember where, but earlier than in his current essay)."

            Interesting. I address this a bit in the conclusion and also in the endnotes (E) where I ask what this all means metaphysically. The two options I find attractive are supersedence (like the total causal work of a system is x and the winning level gets all of x) and layering (x is distributed in some way based). In the former case there is only ever one level, and in the latter case, there could be far more than two levels.

            "To me, the idea that the fully-fledged causal structure of an open system may require the participation of the environment has appeared intriguing at a first reading but ... obscure at a second one :-] How should one interpret the red transition arrows in Figure 3? What is the formal meaning of 'outsourcing causal structure' to the environment, in terms of state transition systems? Should I imagine that some of the states of Romeo's brain are out in the environment. In which sense?"

            Great questions. Wish I had had the space to elaborate more and give a more formalized definition of teleological (apparent or otherwise, I'm neutral) causal relationships. I tried to get it across in a thought experiment of Romeo's brain: if you take a system in isolation, you can still derive a causal structure from it. There's a certain set of causal paths within that system, from state to state. Let's say that for this system that causal structure is pretty sparse (not very complex). Now you expose the system to the environment. You can imagine several different responses. Maybe the causal structure stays sparse. Maybe it becomes extremely noisy and the causal structure loses a lot of its complexity (like every state leads to every other state over time, or something). Or maybe you expose all these "hidden" structured causal paths (one example is shown by the red arrow of a single relationship that is exposed by interacting with the environment). The internal causal structure is no longer sparse - it has gained in complexity.

            Just to make it even more concrete and simple, imagine a simple generally feedforward neural network with input nodes and output nodes, but considered in isolation. Sparse internal causal structure. Then imagine hooking up the output nodes to the input nodes in some complex or interesting way. You could change the internal causal structure pretty significantly, just by "crossing the streams" (sorry) and allowing higher-up events to influence lower-down events in the network now. The formal definition that would go along with the thought experiment would be the degree of that change. The thing that makes this "teleological" in the case of Romeo's brain is that those output-to-input connections that complexify the internal causal structure exist solely through Romeo's interaction with his environment, not through explicitly added physical connections (but mathematically it might be the same). So it's not that the states themselves change, it's the connections between the states may be enriched. In analytic philosophy language, the formal way of phrasing this (mentioned only briefly in the essay) is that the full causal structure of Romeo's brain is not locally supervenient on the physical structure itself.

            "Finally, I'd be curious to check the possible applicability of causal emergence to the very fundamental model of causal sets (Sorkin, Bombelli, Rideout, Dawker, etc.), which is essentially based just on causal relations between unstructured events. Any idea?"

            It's a great possible connection! But I don't want my speculative reach to exceed its grasp by any more than it already has!

            Thanks so much for your wonderful questions and detailed reading,

            Erik P Hoel

            Dear Dr. Hoel,

            I am quite interested in your essay, with its emphasis on the emergence of agency from lower-level structures.

            I also address the issue of agency in my own essay, "No Ghost in the Machine". However, rather than a vague spontaneous emergence at some level of complexity, I argue that recognition of self, other agents, and a causal narrative are built into specific evolved brain structures that create the sense of consciousness as part of a dynamic model of the environment. The reason that this is such a difficult problem is that we are being misled by the subjective perceptions of our own minds.

            Alan Kadin

              Nice essay Dr Hoel,

              Your ideas and thinking are excellent, 'Kublai Khan remains silent, reflecting. Then he adds: 'Why do you speak to me of the stones? It is only the arch that matters to me.' Polo answers: 'Without stones there is no arch

              Agents are generally somewhere above biological mechanisms but below economics on the ladder. They are a major part of the slim section that corresponds to the scale of our everyday experiences etc...'

              I fully agree with you Here I am proposing still higher level agent, at the universe level ............

              .....................please have a look at my essay, where ...............reproduction of Galaxies in the Universe is described. Dynamic Universe Model is another mathematical model for Universe. Its mathematics show that the movement of masses will be having a purpose or goal, Different Galaxies will be born and die (quench) etc...just have a look at my essay... "Distances, Locations, Ages and Reproduction of Galaxies in our Dynamic Universe" where UGF (Universal Gravitational force) acting on each and every mass, will create a direction and purpose of movement.....

              I think intension is inherited from Universe itself to all Biological systems

              For your information Dynamic Universe model is totally based on experimental results. Here in Dynamic Universe Model Space is Space and time is time in cosmology level or in any level. In the classical general relativity, space and time are convertible in to each other.

              Many papers and books on Dynamic Universe Model were published by the author on unsolved problems of present day Physics, for example 'Absolute Rest frame of reference is not necessary' (1994) , 'Multiple bending of light ray can create many images for one Galaxy: in our dynamic universe', About "SITA" simulations, 'Missing mass in Galaxy is NOT required', "New mathematics tensors without Differential and Integral equations", "Information, Reality and Relics of Cosmic Microwave Background", "Dynamic Universe Model explains the Discrepancies of Very-Long-Baseline Interferometry Observations.", in 2015 'Explaining Formation of Astronomical Jets Using Dynamic Universe Model, 'Explaining Pioneer anomaly', 'Explaining Near luminal velocities in Astronomical jets', 'Observation of super luminal neutrinos', 'Process of quenching in Galaxies due to formation of hole at the center of Galaxy, as its central densemass dries up', "Dynamic Universe Model Predicts the Trajectory of New Horizons Satellite Going to Pluto" etc., are some more papers from the Dynamic Universe model. Four Books also were published. Book1 shows Dynamic Universe Model is singularity free and body to collision free, Book 2, and Book 3 are explanation of equations of Dynamic Universe model. Book 4 deals about prediction and finding of Blue shifted Galaxies in the universe.

              With axioms like... No Isotropy; No Homogeneity; No Space-time continuum; Non-uniform density of matter(Universe is lumpy); No singularities; No collisions between bodies; No Blackholes; No warm holes; No Bigbang; No repulsion between distant Galaxies; Non-empty Universe; No imaginary or negative time axis; No imaginary X, Y, Z axes; No differential and Integral Equations mathematically; No General Relativity and Model does not reduce to General Relativity on any condition; No Creation of matter like Bigbang or steady-state models; No many mini Bigbangs; No Missing Mass; No Dark matter; No Dark energy; No Bigbang generated CMB detected; No Multi-verses etc.

              Many predictions of Dynamic Universe Model came true, like Blue shifted Galaxies and no dark matter. Dynamic Universe Model gave many results otherwise difficult to explain

              Have a look at my essay on Dynamic Universe Model and its blog also where all my books and papers are available for free downloading...

              http://vaksdynamicuniversemodel.blogspot.in/

              Best wishes to your essay.

              For your blessings please................

              =snp. gupta

              Dear Erik,

              congratulations to your beautiful essay. It is both brilliantly composed and very interesting. I'm also always keen to learn about new aspects of IIT.

              I have two comments:

              -First, a minor criticism: Of course "quark clouds" do not constitute atomic physics. Quarks constitute the atomic nucleus while chemistry depends on processes involving electrons in the atomic shells.

              -Second, is "causal emergence" really so surprising? To me "causal emergence" seems to appear whenever an experimentalist is performing a series of measurements and statistical uncertainties average out (essentially the "law of large numbers" in stochastics). Probably even more prominently in quantum mechanics the fundamental level appears to be less deetermined than macroscopic classical physics. Can you elaborate a little more about whether the causal emergence you are talking about is different? Do I misunderstand the concept?

              Also I'm a little confused about the exact relationship of under- and overfitting (overfitting seems clear to me but can somethink like underfitting be possible without causal breaks?)

              When you talk about the emergence of teleology in agents I was also wondering how strongly this phenomenon is related to what you call substrate independence and to the relationship between matter and information. One may argue that pure information-based conecpts such as the "bill of rights" or the "contents of the bible" are totally substrate independent, and that agents exist somehow in between such imaterial objects and material physics. Would you agree with this interpretation? Finally, especially exciting to me are the role of the environment and of the perspective, also since that may provide a link to the foundations of quantum mechanics and to our own essay.

              Best regards, Heinrich

                Thanks so much Alan for reading. I don't think of this as a form of vague or spontaneous emergence - it's quite structured and non-arbitrary. It's nothing like the example you give in your own essay of the idea that past a certain (arbitrary) level of complexity, consciousness spontaneously arises. I have some thoughts on your own essay that I will comment on there - thanks so much!

                Erik P Hoel