Hi Eugene,

First and foremost I appreciate the feedback. I clearly put a lot of myself into the work and it's immensely rewarding to hear that it resonates with people. It would seem that we agree on the 'spirit' of many aspects of the question but I'm uncertain if we agree in the details of the letter. On our point of contention, I am of the opinion that it is possible to confer self-awareness into a dynamic system, and I would point to our existence and the record of our historical evolution as proof of this possibility. I didn't exactly spell this out in the essay, but my opinion is that the computational complexity gain conferred by self-awareness is, in a sense, the 'teleological' explanation for self-awareness. In evolutionary terms, however, it's important to note that this merely gives the rationale for why a feature was kept or 'selected for,' but it tells us nothing about how the feature emerged originally. This is akin to saying that photosynthesis emerged "so that organisms could use sunlight." This statement may be true in a sense but it tells you nothing about how to build a leaf or perfrom repeated inelastic absorption of sunlight. It does, however, help us in identifying bottlenecks which constrain the net energy/information flow through a system until enough potential stress accumulates that new mechanisms emerge which increase the net flux by discovering new ways of channeling energy and information. Photoynthesis is sort of my paradigm example of what happens when a dynamical bottleneck is overcome by discovery of a novel mechanism for energy and information flux, and the discovery of self-awareness by humans may have also been a similar, bottleneck-opening discovery which allowed us to process more information and degrade more free energy because of the commputaional complexity increase it allowed. Unlike photosynthesis, I don't have a mechanistic explanation to complete the teleological explanation, but I presume that the details of the evolution of our own species contain sufficient information to explain this mechanism eventually. Given that we are self-aware, some aspect of the selective constraints on our evolution must lead to self-awareness. The challenge is to identify those constraints and implement them in a machine learning environment. We may be a long way from accomplishing this feat, but our existence alone is enough to convince me it must be possible.

Best,

Joe

Dear Joe,

this is a nice essay. I particularly like your E.coli example illustrating that one process can at the same time be both, a manifestation of mindless mathematical laws and of intentional behavior -- depending on the language you use to describe it (or at which scale you describe it, as I'd put it).

I'm not sure that I agree that there is a "highly-discontinuous change" "when we arrive at 'ourselves'". Couldn't it be that you only perceive this last step as bigger than previous steps, because you can't look at it from an even "higher" perspective? Don't you think that from the perspective of E.coli the step which distinguishes E.coli from everything that was there on a "lower" level would also be perceived as huge?

I think this also touches upon a question you raise: Perceiving the last step leading to 'ourselves' as particularly big, might explain why it is more difficult for us to admit the fragility of what is achieved in this last step.

Cheers, Stefan

    Hi Joseph Murphy Brisendine

    Very interesting essay.

    Meaning based information is of the type that hold at least one possible action between two "existents". If there is no potential action in the relation there is no meaning.

    Yet, as I understood you the inner concrete "I" or "self organization" is kept as imaginary one rather than ratifying its self reality through its life duration. The Cartesian approach to reality is dualistic and causal (any levels of complexities).

    Yes, we do perceive reality objectively, which give the uniqueness in us and in the reality itself. It is an eminent transcendental reality, and through it (with it) comes the glory of phenomena.

    thanks

    yehuda atai

      Thanks Stefan!

      I would use the term scale also, and it's good to get a response from a sober physicalist! There's a sense in which it's important to recognize that there is no discontinuity whatsoever when you just view us biologically, perhaps I didn't emphasize that enough but I was short for space. As animals we're fairly unremarkable, maladapted even, and part of me would respond to this question with "why is ths even a question?" As a scientist, I find part of the answer totally obvious, and I had to recall a lot of what I thought as a philosopher to remember why people even find the question controversial. Having said that, I don't see any reason not to admit that there clearly is *something* at least a little "special" or "unique" abut humans, I mean we can build things that exit the atmosphere, and we can tell a story that encompasses the history of the vast majority of the observable universe. Ok basically we can do physics, and I find that really impressive and cool, and clearly we're better at that than other animals by a considerable margin, and that plus our deep and easily-explained anmal vanity is a huge part of why this question is even a question.

      If I had been dismissive I would contend that even "classical" biology, by which I mean Darwin plus empirical investigation combined with the assumption of the consistency of history, is basically enough to answer this question on its own for reasonable people. Modern biophysics-by which I mean all the evidence and theories of classical biology effectively reduced to principles of physics and chemistry, is like an overwhelming mountain of evidence that should make any form of "vitalism" (the notion that there are special laws of physcis which apply only to life) appear laughable.

      Likewise dualism/pan-psychism/mysterianism and any other number of related terms, which don't all mean the same thing but have in common the feature that matters to me which is that they entertain the analogous idea that there are laws of physics which apply only to mental/conscious/spiritual whatever kinds of phenomena, all of that should appear laughable too, not just wrong but extremely silly. The fact that a great many educated people still take these ideas seriously is evidence to me that what the question is really asking for was a way to explain why we insist on believing we're so hard or impossible to explain from "just math" even though we aren't, and the explanation is actually not that hard. The explanation then, for why this is even a question, is not scientific but philosophical, and I think it has something to do with the explosion of science in modernity as well.

      Note that we have been a species already for 200,000 years roughly, and the part that seems legitimately like a scale-change to me has only happened in perhaps the past 400 years or maybe the lasst couple thousand with increased acceleration in the past few centuries. I think "modern humans" are really where we see a sudden jump in the total amount of information which is assigned meaning. Most animals only find information related to their food, mating, and other immmediate envvironmental concerns meaningful, and they can be better adapted than we are at sensing and responding to that information in their environment, but we find literally every discerible aspect of our environment meaningful. This is a legitimate difference to me, and I think the scale of information we produce, quantifiable things like the number of bits stored on the internet and the likely much larger but not as readily quantified number of bits stored in the physical economy in the form of products and commodities, all of this is demonstrably a larger impact on the structure of the earth than any other individual species on the planet. I cite Scott Aaronson a couple times in there, and when I read his work on the anthropic principle being a "cheap way to gain computational complexity" I immediately thought that this is what really occurred in our brains that caused us to be able to do science, basically the one thing we do that I think is special and wanted to explain, and thought that this was connected to the philosophical rise of dualism and modernity. That's the part of the story that I thought I could contribute uniquely as someone who has existed in both the sceintific and philosophical community.

      Anyway are you in the competition as well? Based on your comments I'm sure I'd agree with your submission. Glad you ejoyed it!

      Joe

      Thanks Yehuda! From your comments I have the impression you are a phenomenologist in the tradition of Husserl. I personally prefer (young) Heidegger and Merleau-Ponty, but I do certainly believe in the glory of phenomenaa! I'll look over you essay now as well I have been meaning to :)

      Joe

      Joe:

      I enjoyed reading your essay but have some questions for you:

      You say, "I would invite you to try and consider how the world would manifest to you if you had no memory..." You ask whether individuals with severe memory impairment possess "selves" and whether such individuals know that they exist? You answer, "I believe the evidence obliges us to answer no."

      You are saying, that individuals with total memory impairment do not know they exist, that is, that they do not have a self. But the two things are not the same. Knowing you exist is not the same as having a self. When I am asleep, I do not know if I exist or not, but does my 'self' really cease to exist until I wake? Does an infant human not have a self until it has a memory? By this reasoncing, if two infants have the same memory, for instance: 'the ice cream I had yesterday was delicious', are you to say both infants have the same self? If they don't have the same self, then self must be more than memory. So, you see, the existence of memory is not, prima facie, the pillar of selfhood as you say it is.

      In many respects you echo Dennett's stance. You say, "As we move from the meaningfulness manifested by bacterial chemotaxis to the meaningfulness manifested by animals in their environments, not much changes except the scale of the processes involved." Similarly, Dennett explains that the difference between simple and complex systems is one of scale--one of degrees of system complexity. I describe this view toward intentionality and representation as a greyscale stance were intentionality is determined by degrees of functional complexity. In my critical analysis (see paper at http://mind-phronesis.co.uk/intentional-stance) I explain why this stance is problematic.

        Hi Mark!

        I did my best to distinguish between selfhood and self-awareness in the essay, or being a self vs knowing that you are a self, but I was covering a lot of ground and I might have not been clear on my position. To answer your questions just from my perspective first, any living thing that embodies a dynamical system "thermostatted" to a fluctuating environment is a self, to borrow the technical term from your work. I definitely do not have the impression that there is a merely incremental scale of "complexity" (however quantified) and there is simply some critical value of this parameter where meaning just pops into existence, I'm not sure if thats exactly what you mean by the incrementalist view but I want to be clear that I'm advancing nothing of the sort. So my assertion was that meaning, which for me is a synonym for intentional self-world relationships, emerges with organisms and the fact that you have, ultimately, a converging series embodied in a physical system repeatedly interacting with an environment. Both the embodiment and repetition are necessary, however, for learning or adaptation to occur (and I consider both of those to be recapitulations of the same dynamics at different scales).

        Self-awareness is a separate concept in addition to selfhood that had to evolve on a stable foundation of many interacting non self-aware selves, and my proposal is that the demonstrable gain in computational complexity generated by self-awareness is the selective mechanism which caused self-aware machines, once they had evolved, to outlearn non self-aware machines. I think the seeds of self-awareness are present in other social mammals if not all animals, but in the past few thousand years we became REALLY aware of our awareness, and aware of our awareness of our awareness ad infinitum and the same with the awareness of other humans. I don't know what the mechanism was that actually caused this change, but I presume it has something to do with social and cultural evolution, given that it's clearly too recent to be anything biologically or genetically special about humans, plus the fact that there appears to be nothing biologically or genetically special about us.

        I hope that clarifies my views, and if nothing else I promise you I'm not defending a naive "just take the standard model and start adding complexity and poof!" view of the evolution of intention.

        I can't comment on Dennet or Searle because I never studied analytic philosophy in any detail.

        Finally, thanks for taking the time to read and comment on my work!

        Joseph,

        Thank you for a deeply thoughtful and nicely written essay. There are aspects of your philosophical view that are very insightful and well presented - for example, in thinking about what it would be like to have no memory, and in your critique of "self" as a basic intuition. And your overview of the progression from thermal physics to bacteria etc. is excellent, very clear-minded.

        I'm sorry to say that I don't agree with your thesis about "anthropic reasoning"... though I don't think you're far off track. I fully agree that from bacteria to our primate ancestors, there's a profound continuity in the nature of meaning and how it evolves. But I would put the discontinuity - the beginning of a new kind of evolutionary process that made us human - much further back, as described in my essay, which is also on the emergence of meaningful information.

        Of course people born into the most primitive cultures still existing on Earth are fully capable of educating themselves in modern science, given the chance. Their native languages all evidence remarkably sophisticated mentalities, beyond any comparison with other animals. Nonetheless, you're right that there are further intellectual discontinuities, as in ancient Greece (and in other cultures where writing began to be a primary medium for cultural evolution), and with the emergence of modern science. And I agree that new dimensions of self-awareness are key elements in these events.

        But I do find baffling the notion that we could tell machine A that it exists - and even more, the thought that this would make all records available in the world meaningful to it. This is surely not in the spirit of the young Heidegger you mentioned in your note above (if by 'young' you mean Being and Time). Self-awareness at any level is not a simple ability or "a cheap trick." It has many "pillars", as you say - not just memory but anticipation, the ability to interpret our perception and create a shared reality, etc.

        Language vastly expanded our mental scope in all these respects, and that happened all over again when written records begin to pervade society... and again with the explicit self-reflexion of modern thought. You ask us to imagine what consciousness would be like without memory... so imagine what it would be like in a culture where no information had ever been conveyed except in momentary, face-to-face contact between people, and no one had ever imagined the possibility of recording anything. Yet amazingly complex societies evolved in these conditions.

        Of course we can't give due weight to everything in such brief essays! - as it is, I'm amazed at how much you've been able to touch on here. And your ending points to the heart of the matter, I think, where you open up the idea of "mutual information". A central point for me is that "I" always evolves in relation to "you". It's true that modern Western thought has evolved as a dialectic between 1st-person and 3rd-person views of the world, and hardly reflects at all on the "You and I" kind of relationship, that Buber emphasized. But I think this is a key failure of the Cartesian/Kantian worldview, that even Heidegger did not fully overcome.

        In any case, I very much appreciate the level at which you're thinking through these issues, and I'm rating your essay at the top of the scale.

        Conrad

          Joseph,

          Thanks for the detailed response to my queries. I can see now there is much more to your ideas than could have been crammed into the essay topic. I did misinterpret some of your ideas and would need to read it again I think now that you have clarified things. It would be nice to cover more of these ideas through discussion.

          In your second paragraph you talk of the emergence of self-awareness and say "I don't know what the mechanism was that actually caused this change" (the change that brought about self-awareness). You may be interested in my hierarchical construct theory essay which also talks about a hierarchy of emergent capabilities that have distinctive attributes. I do give an account of how each level, in a three-tiered hierarchy, emerge and evolve.

          Hi Conrad,

          Thanks for the kind words and I'll be sure to look over your essay as soon as I get the chance, I've been travelling and haven't been able to reply freely. I think that you're correct about the weaker parts of my thesis, and I fully admit that I don't have the education in human anthropology, a subject I've never really studied, to properly address the mechanistic causes of our 'general' intelligence, as opposed to the more narrowly niche-based intelligence of most of our animal cousins. As I alluded to in the comments above, I envisioned the argument around anthropic reasoning as more of the 'teleological' explanation for what sets our species apart, and in calling it a "cheap trick" I was leaning on two separate characterizations from science and philosophy. First, I was quoting Bataille. Second, that use of the anthropic principle alone to 'solve' problems is a form of "science on the cheap" (I think Baez said that roughly) is well-known. Part of my point was to try and demonstrate that we actually use anthropic reasoning constantly. Anytime we reason about the past or future practically, we employ the consistency of the past and our own memory, but we only notice that this is problematic when we stretch it to it's limits, like if we say "of course the fine structure constant has to be about 1/137, if it were something else we wouln't be here." So the "cheap" part of self-awareness is that it really is an awfully large gain in computational complexity that comes with only larning one "fact about the world." I want to stress that this doesn't tell me anything about how to make something self-aware; it doesn't tell us anything about the efficient cause or mechanistic explanation. The answer to 'how to make it' will presumably come from people like yourself who know a lot about human anthropology and evolution collaborating with people who have a good imagination for creating machine learning environments. So that's the difference between a 'teleological' and an 'efficient' explanation, to me anyway. The latter tells you how to make something while the former is only possible because we can use our powers of anthropic reasoning to tell stories after the fact that "explain" how something led to the present, or in this case "why evolution kept it." I tried to be clear that I have no idea how to make something self-aware, but it was just that toss away comment about "implementation" and the vague paragraph mentioning hands and memory and writing, which is about the extent of my knowledge concerning human anthropology. I'd love to learn more though, because I do assume that the mechanistic explanation can be found in human evolution, and that assumption too is a product of my anthropic reasoning--I know our evoltion led to our intelligence.

          So for the record I don't think I've said anything in conflict with the Heidegger of Being and Time, though I certainly said many things he never would have. What I really want to retain from Being and Time are all the resources we need to combat all of this neo-dualism that has cropped up in phil mind in response to A.I. paranoia. Those resources are the phenomenological means of accessing the dimensional content of experience, things like its duration and intensity. The dimensional content is exactly what a physicist needs as well to begin thinking about the physical properties of a system, and the language of dualism intentionaly hides the dimensional content of experience. This is why physics and phenomenology need one another, in my opinion.

          If you still think I'm wrong about anthropic reasoning I'll read your essay and perhaps I'll change my mind! :)

          Joe

          Joe,

          Rereading my post, I think it sounds more critical than I meant it to be... I really was impressed by your essay, and it deserves a 10, a rating I don't give lightly. And I'm far from an expert in any field of science, though I try to make sure my writing is accurate. Heidegger I do know fairly well, since he was very important to me in grad school back in the 70's. I'm very glad you know his work, and delighted to find someone with that background doing biochemistry. The truth is I can hardly read "Philosophy" any more, since the world itself is so much more interesting. It often seems to me that while our knowledge has grown exponentially since 1900, we still seem to conceptualize the world largely in terms that were familiar in the 19th century.

          You say, "What I really want to retain from Being and Time are all the resources we need to combat all of this neo-dualism that has cropped up in phil mind" - Yes, this is where Being and Time really succeeds, as still the deepest critique of the Cartesian/Kantian tradition. And he did it by articulating the many "equiprimordial" elements involved in what seems like the perfectly simple idea of "self". I get your point about taking self-awareness as a "teleological" explanation of what makes us humans so different, and you're right that it's implicit in all our experience, from a very early age. And I'm intrigued by your comment about "the dimensional context of experience" that's hidden by dualism... I hope to discuss that further!

          Thanks - Conrad

          Conrad, Mark, Stefan, everyone else who expressed interest in further discussion--

          Thank you all for considering my ideas and it would be my pleasure. I can be reached at josephbrisendine@gmail.com and feel free to reach out anytime. Meanwhile I'll get back in touch once I have properly considered everyone else's entries.

          Best,

          Joe

          Hi, Joseph, congratulations, this is an excellent essay. Both for the ideas in it, and the writing. Many of the things you mention resonate with the ones I chose to focus in my essay (comments from you would be most appreciated). I still get wound up by the anthropic reasoning, though. I understand that conditioning on our existence rules outs a-priori possible evolutions of the universe that do not give rise to us. I also understand that perceiving ourselves makes all things related to ourselves interesting. I do not understand, however, in which way anthropic reasoning provides explanations - and I do care for explanations, whatever those may be!

          In any case, this is an old problem I've been having for ages, it's not your fault, I'll just have to keep thinking. In any case, thanks for the great read!

          inés = one more sign without meaning about to fall apart.

            Hi Ines that was a lovely compliment!

            I had noticed and enjoyed your essay a great deal as well, I will definitely share my thoughts on it with you on your page. But first allow me to say that there's a sense in which you're right, anthropic reasoning explains nothing. If we insist that an explanation must be a mechanistic explanation, which means that it should tell us clearly how to construct the phenomena in question, then we should conclude that anthropic reasoning doesn't furnish explanations. In science, we typically adhere to the spirit of Feynman's "what I cannot create I do not understand", and we don't consider a phenomenon explained until we have a mechanistic explanation. I didn't do the best job explaining this difference in the essay becuase, frankly, I have only just recently begun to piece these thoughts together myself, whereas when discussing thermodynamics and biochemistry I was just presenting things I've understood for years. But anthropic reasoning does provide a different kind of "explanation" for phenomena, although I would also be fine with using a different word for what we mean in this case. My idea was that the explanations it provides are the kinds of explanations contained in stories, which might also be considered what a "teleological explanation" does. Rather than tell you how a thing works and what conditions are required for its stability, a story provides some simplifying narrative that assigns a reason for the existence of the phenomena simply in terms of its continuity with the rest of reality that we perceive. In ordinary life, I think we accept these kinds of explanations all the time, and only when we have the circumspection required by science do we even notice that they aren't very good explanations, or more precisely that the sense of that word in science and in everyday life is not really the same. So I think your instincts are correct! As with almost any situation where the results of science conflict with our ordinary understanding, the problem arises because of ambiguities in language that we don't notice until they conflict with our capacity to interpret the results of experiments. Anyway I hope that helps and I'd be happpy to discuss this or anything else with you further in the future, because your essay also has some ideas in it which I have never considered before but which really impressed me. I'll mention them on your essay's page though.

            Joe

            I think you are right, I am expecting from anthropic reasoning something it cannot quite provide. I just wish I could benefit more from the things it can actually provide. I promise to work on it, and if needed, come back to you...

            Thanks for the great explanation!!

            inés.

            Gee I think it all depends on what you mean by that, and what you include or don't as part of the "spark."

            Dear Joseph,

            your essay is remarkable for the degree of personal participation that you put in your story-telling, both in tone and in contents.

            However, if I were to summarise some of the main points you make, I'd have some difficulty with one which is quite central: the value of the self. On one hand you attribute much higher sophistication and 'computational capability' to the robot with self-awareness (following mainly Aaronson?); on the other, you regard the pretended certainty and stability of the self as a "wonderful irony of the history of philosophy". Maybe the conflict is only apparent? I'd be curious about a final word from you on the issue. (I read your text twice, but didn't go through the comments in your blog. Apologies if you have already covered the issue.)

            Thank you and best regards!

            Tommaso

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

              Thanks for the kind words Tommaso!

              I'm happy to try to clarify my position. As you intuited, I think the tension is only apparent. I think the gain in computational complexity due to the ability to use "anthropic reasoning" has been demonstrated clearly by Aaronson, just as you say, and that this complexity gain is the "teleological" explanation for our self-awareness. This doesn't mean, of course, that this caused us to become self-aware, as this would be like saying that photosynthesis was discovered "so that orgaansms could use sunlight." only a mechanistic explanation actually tells you how to create something and allows you to infer it's actual cause for existence. A teleological explanation does tell you, however, why it is that once there were self-aware animals they quickly out-learned non self-aware animals, just s it tells you why, once there were photosynthetic organims, they quickly covered the earth. So that's the powerful part of our selfhood. My claim is that it is powerful but also unstable. The sense of its instability can be found everywhere in our experience if we pay attenntion to it, but then my contention is that it also follows logically from understanding how biology increases its thermodynamic efficiency in line with natural selection. If you want to compute things near the Landauer limit, your band gaps have to be as close to kT as they can get without being overcome by noise. Our self-awareness then is the source of our superior understanding of nature and our ability to share information so readily, but the "free energy of formation of self-awareness" appears to be very small, and easily overcome. I hope that all makes sense!

              Thanks also for taking the time to read and consider my ideas, it is very rewarding to know that they were seriously considered by another intellect!

              Joe