[Leibniz, thought emergent or all the way down] - A New Kind of Science: The NKS ForumA New Kind of Science: The NKS Forum
Pages:1
Leibniz, thought emergent or all the way down
(Click here to view the original thread with full colors/images)
Posted by: Jason Cawley
A recent exchange between Stephen Wolfram and Neal Stephenson about Leibniz had me thinking about him again. Leibniz seems to come up fairly often in NKS circles. Gregory Chaitin quoted something of his with approval at the NKS 2003 conference. I happened to be reading the quoted text at the time, for quite different reasons.
He is certainly an interesting figure, historically and philosophically. Besides one of the inventors of calculus, he basically came up with binary as we know it today, and proposed the idea of a universal language. His main occupation was idealist philosophy, asserting this was the best of all possible worlds, and proposing his famous but often misunderstood monads. We can sometimes see parallels to modern ideas and too easily read them back into his system.
What I notice is that Leibniz did not believe that intelligence or something like it could be an emergent phenomenon. The primary motivation for the idea of monads is the notion that they are the only way to explain the phenomenal given, thought, because composition supposedly cannot be the origin of thought.
That is, the choice is between thought that goes “all the way down”, and thought that only arises at or above some level of aggregation. That there is thought in the world is data, given by experience, which we have to account for somehow. Either it can be made from components that do not themselves think, or there are “atomic”, underlying thought-entities from which the phenomenon of mind can arise. Leibniz clearly thought the latter, and did so precisely because he could not see any way to get thought as an emergent property.
This is clearest in one passage in the Monadology -
Supposing that there were a machine whose structure produced thought, sensation, and perception, we could conceive of it as increased in size with the same proportions until one was able to enter into its interior, as he would into a mill. Now, on going into it he would find only pieces working upon one another, but never would he find anything to explain perception. It is accordingly in the simple substance, and not in the composite nor in a machine that the perception is to be sought.
Leibniz, Monadology 17.
Up until the last sentence it is all true and a fascinating thought experiment. But the last sentence draws a conclusion from the previous that simply does not follow.
It is in fact simply an instance of the fallacy of composition. Meaning reasoning of the form: The parts of the whole X have characteristics A, B, C. Therefore the whole X must have characteristics A, B, C. You can easily see this is a fallacy by considering a common example. Sodium and chloride are each dangerous if ingested (true), therefore sodium chloride is dangerous if ingested (false).
Composition can modify attributes. There is no necessity that the attributes of the simple elements translate directly to the attributes of the composite. Applied to the case before us, the reasoning would run as follows. The parts of the machine are not capable of thought, therefore the machine is not capable of thought. There is no valid reason to believe the conclusion is true in that reasoning. It might be true for other reasons, but the particular argument advanced fails to support it.
While the conclusion does not follow, however, it was clearly Leibniz’s opinion in the matter. Arrangement can add something new. Leibniz’s “mill” thought experiment shows only that no special thought-thing is present inside the machine as its magic soul-source. It does not address the possibility that thought, sensation, or perception are emergent properties of the whole machine, not properties of its constituent parts. But he clearly thought it did. He thought it was not possible for thought or something like it to be an emergent property.
Leibniz did not see how something like thought could arise from arrangement alone. He had to explain the existence of thought some other way. He concluded, for the above unsound reason, that thought must reside in the ultimate constituent parts of the universe, the simple absolutely non-composite elements, which he called “monads”. Anyone who likes may agree with the conclusion and find his scheme attractive. But the reasoning advanced in favor of it by Leibniz was weak, and does nothing to dismiss the alternative that thought may be an emergent property of definite arrangements.
Leibniz explains inert things not being alive by positing that the thought-like elements of which they too are supposedly composed are just in a continual state of stupor. We might say, “simple” instead. To me the obvious weakness of this - it seems forced - serves to show that there is at least as much counter-intuitive difficulty in thought all the way down as in thought emergent schemes.
To me one of the clear results of NKS (indeed even there in precursors), is that universality is an emergent phenomenon, a matter of arrangements and not of special elements. The NKS book showed moreover that the threshold involved is rather low, low enough to imagine it being reached and surpassed by natural rather than engineered systems.
More may be required before one can properly speak of intelligence, but at least of intelligence in a high enough sense universality or something like it would seem to be a necessary, if not yet a sufficient, condition. To me this supports the emergent rather than the all the way down notion of where thought comes from. But some idealists might well read it differently.
There are other peculiarities in Leibniz sometimes overlooked by those eager to find signs of contemporary views in his work. When he calls his monads “ones” he is not contrasting them with “zeros” in some information theoretic sense. The reference is instead to Platonic unity. Leibniz explicitly says that no two monads are exactly alike. They are not part of a discrete model for him, as he subscribes to the doctrine of the “plenum” or absence of any real vacuum in nature, and also to the doctrine of real continuous infinities.
Because each portion of matter is not only, as the ancients recognized, infinitely divisible, but also because it is really divided without end, every part into other parts, each one of which has its own proper motion.
Ibid, 65. Which it seems to me corresponds more to the idea of a field than to a discrete 1 and 0 model.
For those interested in more Leibniz, you can find his Monadology in etext form here -
http://stripe.colorado.edu/~morristo/monadology.html
I hope this is interesting.
Posted by: Gunnar Tomasson
In an English-language book published posthumously, a friend of mine and author of an eleven-volume opus on ‘The Roots of Icelandic Culture’, the late Einar Pálsson (d. 1996), concluded that the ancient concept of Monad was central to the world-view of Iceland’s 9th century A.D. settlers, as later reflected in the Saga literature of 13th century Iceland.
In introducing an extract from The Theology of Arithmetic, attributed to Iamblichus and tentatively dated to the middle of the fourth century A.D., Einar sketched a bird’s-eye view of the Monad’s manifestation in Icelandic Saga culture as follows:
“Iamblichus begins his work on the ten first numbers (as used by Pythagoras) with the following statement: "The monad is the non-spatial source of number." This fits in with the decipherment of Icelandic numerical symbolism. Niall [title character of the Saga masterpiece ‘Brennu-Njalssaga’ (Saga of Burnt Niall)] is the embodiment of that point in space which is undivided; he himself is that which later is separated, the unity which becomes diversity or multiplicity....”
In General Relativity, the concept of “non-spatial source of number” would not seem to be problematic – presumably, it would translate into that of a Point in the Space-Time Continuum.
However, just as General Relativity and Quantum Mechanics have yet to be reconciled within a unitary conceptual framework, so the concept of a non-spatial Monad would seem to be problematic insofar as the pre-suppositions of QM are concerned.
The Iamblichus extract reads as follows:
“Just as without the monad there is in general no composition of anything, so also without it there is no knowledge of anything whatsoever, since it is pure light, most authoritative over everything in general, and it is sun-like and ruling, so that in each of these respects it resembles God, and especially because it has the power of making things cohere and combine, even when they are composed of many ingredients and are very different from one another, just as he made this universe harmonious and unified out of things which are likewise opposed.
“Furthermore [the Pythagoreans] say that in the middle of the four elements there lies a certain monadic fiery cube, whose central position they say Homer was aware of when he said: "As far beneath as is Hades, so far above the Earth are the heavens." In this context, it looks as though the disciples of Empedocles and Parmenides and just about the majority of the sages of old followed the Pythagoreans and declared that the principle of the monad is situated in the middle in the manner of the Hearth, and keeps its location because of being equilibrated, and Euripides too, who was a disciple of Anaxagoras, mentions the Earth as follows: "Those among mortals who are wise consider you to be the Hearth." Moreover, the Pythagoreans say that the right-angled triangle too was formed by Pythagoras when he regarded the numbers in the triangle monad by monad.”
Clearly, this is not the stuff of modern discourse upon Man and the World – a fact which is not necessarily a blemish on the ancient mode of thinking.
However, the ultimate test of the ancient mode of thinking, it seems to me, is whether or not the Pythagoreans turn out to have been right in proclaiming
(a) that All is Number; and
(b) that Man can, must, and will find his way back to his Creator through Number.
The jury is still out on that one.
Gunnar
Posted by: Gunnar Tomasson
Today, a friend forwarded an exchange on "Which came first - Mind or Matter?" - a subject matter akin to that of the present NKS thread.
In response, I forwarded my NKS reply to Jason's original post, underlining two sentences -
In General Relativity, the concept of “non-spatial source of number” would not seem to be problematic – presumably, it would translate into that of a Point in the Space-Time Continuum.
However, just as General Relativity and Quantum Mechanics have yet to be reconciled within a unitary conceptual framework, so the concept of a non-spatial Monad would seem to be problematic insofar as the pre-suppositions of QM are concerned.
- and commented as follows:
As indicated by the two underlined sentences, the point at issue - which came first, Mind or Matter - can be recast in terms of what Stephen W. Hawking termed "...the great intellectual achievements of the first half of this [20th] century. [Namely,] The general theory of relativity [which] describes the force of gravity and the large-scale structure of the universe...[and]...Quantum mechanics [which] deals with phenomena on extremely small scales...]."
"Unfortunately," he added, "these two theories are known to be inconsistent with each other - they cannot both be correct. One of the major endeavors in phyiscs today, and the major theme of this book, is the search for a new theory that will incorporate them both - a quantum theory of gravity. We do not yet have such a theory, and we may still be a long way from having one, but we do already know many of the properties that it must have. And we shall see, in later chapters, that we already know a fair amount about the predictions a quantum theory of gravity must make." (A Brief History of Time, Bantam Books, 1988, pp. 11-12)
Thus, "one of the major endeavors in physics today" is predicated on the implicit proposition that there is NO valid Mind-or-Matter Priority Question.
In turn, this casts new light on the "tragedy" (Max Born) of Einstein's decades-long intellectual isolation from his peers, one of whom (Oppenheimer) branded him "cuckoo" on grounds, which Hawking summarized as follows:
"...Einstein refused to believe in the reality of quantum mechanics, despite the important role he had played in its development. Yet it seems that the uncertainty principle is a fundamental feature of the universe we live in. A successful unified theory must therefore necessarily incorporate this principle." (Op. cit., pp. 155-156)
Bunk!
For, as indicated by Einstein in 1949 - a passage which I have cited before - the "tragedy" was rooted in his peers' epistemological confusion and intellectual hubris:
“Roughly stated [my] conclusion is this: Within the framework of statistical quantum theory there is no such thing as a complete description of the individual system. More cautiously it might be put as follows: The attempts to conceive the quantum-theoretical description as the complete description of the individual systems leads to unnatural theoretical interpretations, which become immediately unnecessary if one accepts the interpretation that the description refers to ensembles of systems and not to individual systems. In that case the whole “egg-walking” performed in order to avoid the “physically real” becomes superfluous. There exists, howerver, a simple psychological reason for the fact that this most nearly obvious interpretation is being shunned. For if the statistical quantum theory does not pretend to describe the individual system (and its development in time) completely, it appears unavoidable to look elsewhere for a complete description of the individual system; in doing so it would be clear from the very beginning that the elements of such a description are not contained within the conceptual scheme of the statistical quantum theory. With this one would admit that, in principle, this scheme could not serve as the basis of theoretical physics.” (‘Albert Einstein, Philosopher-Scientist’, The Library of Living Philosophers, Cambridge University Press, Third Edition, 1988, pp. 671-672)
Gunnar
Posted by: Jason Cawley
Some further comments on all of this, prompted by a private exchange with Neal Stephenson on Leibniz and monads compared to CAs.
Neal wrote "It's important to be clear on exactly how they differ. A monad has the ability to perceive all of the other monads in
the universe. The sum of its perceptions defines its state. Its state serves as the input to an internal rule, characteristic of that monad. This rule determines how the monad will act in response to a given state. When the monad acts, then, its actions are perceived by all of the other monads, causing changes in their internal states, etc., etc."
Right. What this means is that for every monad, there is for every state of the universe, some new internal state of the initial monad. Since "state of the universe" is a summation over all monads, it means the "complexity" of the input to the rule for each smidgen of the universe is the power set of the universe. Since Leibniz allows all of these rules to differ, you get another power again - though as Neal notices, since you already have the
power set of the universe, this additional "degree of freedom" is not strictly necessary.
But it is Neal that notices, not Leibniz. Leibniz explicitly says "there are never in nature two beings which are exactly alike, and which it is not possible to find a difference either internal or based on an intrinsic property" - Monadology 9. He does not say "or on extrinsic relations". Leibniz's statement would, moreover, come as news to laser engineers (quantum indistinguishability being fundamental to that real physical effect). That is, it may have seemed a reasonable philosophic principle but it just happens to be wrong.
There is also something a little unsatisfying about it as a supposed rule specification. There is nothing miraculous that one could imagine that would not be allowed as a supposedly "necessary" consequence of an input the size of the power set of the universe. (Especially since, for Leibniz, each of these is also strictly infinite, as I will show below).
While this serves Leibniz's purpose of obscuring the distinction between "spontaneous" and "foreknowable to God", it has little to do with the simple program idea. It is instead a "maximally complicated program" idea. If one likes, one can notice that all simple programs are proper subsets of, or restrictions on, a maximally complicated one. Not terribly meaningful to me, but true enough, fine.
CAs are certainly well defined with dependence on neighbors farther away than "immediate". In principle, there is nothing wrong with extending dependence to any arbitrary finite number of other states within the system. The CA idea does require them to be finite. Also, the simple program idea is that you really don't need to go all the way to the end of the universe to get interesting behavior. Which is after all comparatively trivial - you import complexity rather than explaining how it appears.
Neal wrote "In a typical CA scenario there is only one rule. So one could argue that monads are not CAs because they can have different rules. But it is easy to envision an experiment similar to the ones in NKS in which, however, the rule was allowed to vary from one cell to another."
True. Wouldn't be a CA as usually defined, would be a finite state machine (if the number of elements is finite), and as a generalization of a CA there is no reason to quibble about its title. I think it is more to the point to ask what more this additional degree of freedom actually gets one. It dissolves regularity. For any given previous state, it also is well defined only in one direction (casually "forward") - since you've got way more than enough to specify any output that can come next. A rule that ignores all previous states and just assigned new state X at cell A, and Y and cell B, qualifies. But doesn't correspond to our intuitive sense of a "rule", though.
The point is that overproviding degrees of freedom keeps only the names from the more restricted case of an actual rule. The actual rewrite rule can be anything, and it "fits the definition". That is a sign of a definition with weaker logical content.
(Neal asked for a reference to infinite divisibility in Leibniz).
Monadology 64-5. "The machines of nature, however, that is to say, living bodies, are still machines in their smallest parts ad infinitum. Such is the difference between nature and art, that is to say, between Divine art and ours. 65. The author of nature has been able to employ this divine and infinitely marvelous artifice, because each portion of matter is not only, as the ancients recognized, infinitely divisible, but also because it is
really divided without end, every part into other parts, each of which has its own proper motion. Otherwise it would be impossible for each portion of matter to express all of the universe." - Leibniz
He needs to project the entire universe into the microcosm, so he allows real infinite divisibility. In our modern terms, the allowed information content of a single monad must be as large as the allowed information content of the universe. Since the latter is specified by the power set of the universe, only continuous infinity will serve. (I suspect Leibniz noticed the need for "infinity", but not the distinction "continuous vs enumerable infinity". Whether he noticed the need for it or not, he provided it in his picture of things).
This is not classic atomism, but a conscious rejection of materialist
atomism. Atomism was associated with the materialism of Lucretius, with a finite number of atom types (Leibniz instead has an infinite number of different monads), with form (Leibniz's are formless), in void (Leibniz's are a "plenum" or continuum). Leibniz is not just "turning that around", because Lucretian materialism was not the dominant view - idealism was, and had been from Anaxagoras's first speculations on universal mind down to Descartes.
The model for monads is not atoms (as physical, extended, hard material objects of definite shape flying through void) but soul-like or mind-like mathematical points. They are no more required to be finite than the number of points in space are. Leibniz developed mathematical infinitesimals not as a means of reducing continuous quantities to atoms, but with the idea that infinite divisibility is a real phenomenon, and not simply a mathematically abstraction. That is why I said before that Leibniz's notions remind me of continual fields.
"you can't talk about them without granting that there is a limit to
divisibility."
Simple, he said. Leibniz regarded them as the ultimate constituent parts of the universe, as point like, but as without extension or materiality. They are atomic in the sense of simple, yes, and he calls them the "true atoms of the universe". But they are not finite, material, or extended. He explicitly says they are without form.
Is it incoherent to think of free will or intelligence as emergent properties, rather than going all the way down?
Whether free will and consciousness are emergent properties or elemental properties of the ultimate constituent parts, is the exact point in dispute. Leibniz rejected the "emergent" possible answer to that question simply because he could not see how something like consciousness could be an emergent property. Now this simply does not follow. An inability to see how to do something is not a proof that thing is impossible.
(You might semantically rest an "incoherent" on a definition of "free will". No theory has to match a definition, however, only the phenomenal givens. Which includes the apparent freedom of the will, but not some definition of "truly free" or what have you).
What is incoherent about consciousness as an emergent property? Leibniz's own thought experiment of the mill shows how it could be an emergent property. It simply resides in the arrangement of the parts, rather than within any of the parts themselves. Each of the parts can be as material as you like. Their arrangement is formal, and is not reducible to those parts.
To assume the whole can only have a property if that property is present in each of the parts is known in philosophy as the fallacy of composition. It is known that in general it does not follow. Some other reason must in each case be supplied, to make such an inference.
Arranged wholes can and do regularly have additional properties, due to the arranging alone, that are not found in any of the constituent parts, taken separately. This is beyond dispute. What may be disputed is whether consciousness is like that, or goes all the way down. But nothing has been offered here by Leibniz, to rule out the emergent hypothesis as one possible answer to that question.
It may make more sense to you. It may seem more plausible, or your hunch about how it really works. But this is quite different from claiming to have shown that the emergent hypothesis is "incoherent". Nobody has ever shown any such thing.
(The question arose of souls and minds in Leibniz, and what he really meant about them)
He distinguished souls from minds. Soul is a lesser thing, a criterion more easily met.
"While souls in general are living mirrors or images of the
universe of created things, minds are also images of the Deity himself or of the author of nature. They are capable of knowing the system of the universe, and to imitate it somewhat be means of architectonic patterns, each mind being like a small divinity in its own sphere... the totality of all spirits must compose the city of God". - Leibniz, Monadology 83
Even monadic pre-life is microcosmic in the sense that everything about the universe is projected down onto the smallest bit. But it is not microcosmic in the sense of "knowing the system of the universe". The data transmit, if you like, but not the laws. Finite mind has access, in addition to the data, to the laws or patterns in the data, and as such are microcosmic in the "knowing" sense, as well. Though with an element of abstraction or finiteness in that knowing. Whereas God tout court has the data and the laws infinitely. That is, his scheme is -
Infinite data - any monad
Finite laws - any mind
Infinite laws - infinite mind
(Neal mentioned Newton's discussion of "vegetable spirit")
The term is actually from Aristotle. He distinguished in De Anima between the vegetative soul - responsible for life as opposed to death - the sensuous soul - responsible for sensation or we might say consciousness - and the mind or intellect, we might say intelligence. Plants have the first but not the other two, animals have the first two but not the third, humans have the third - in the traditional view. Compared to this received picture, Leibniz is ascribing something akin to the first even to rocks (the idea of "stupor" as the analog or cause of apparently lifelessness).
As for death, Leibniz simply denies its reality - living things can be
"stunned" for long periods of time, but there is no real death in Leibniz. Since it is not the composition that is the source of life, but the simple constituent parts, this follows from the whole set up. Death is obviously the disarranging of a composition. If that composition is not essential, neither is its disarrangement.
Fun stuff. I do think that Leibniz was "weirder" by our typical present sense of things that Neal sometimes seems to read him, but I understand why he does so. It is perfectly normal to notice the suggestive similarities and points of convergence, or to take from a thinker the ideas that fit with our present ones from other sources. In the history of thought, highlighting peculiarities is usually the focus instead. It is about the variety seen across a large number of historical figures, so one harps on specific differences. Each can enrich our understanding of both present problems and previous thinkers.
Posted by: Ken Zweibel
It seem like one of the key views of NKS is that complexity can emerge from apparent simplicity. So the idea would be that consciousness emerges as a result of complexity and has no antecedents.
Maybe not. Why?
Emergent properties usually have some similarity; even in the sodium plus chlorine case, salt has chemical properties, as do the other 2. So maybe the lack of a spirit in the machine, a la Liebniz, is a telling criticism of consciousness as an emergent property. Maybe you need a 'kind of consciousness' at a basic level, to then lead to an emergent 'human consciousness' at the complex level.
That's why I like to think of each of the tiny CA boxes (or their network equivalent) as an on/off of being/nothingness, with our various emulation modes producing what we perceive as a world and ourselves. Perhaps simultaneously, there are an infinite set of other modes, producing some rather unusual other set of perceived worlds.
So the question becomes: how do these emulations arise? what do they mean? Are we the programmer who sets up our emulation (from some external level), or are we the 'victim' simulation, doing some weird gyration for unknown purposes (or none)?
Posted by: Ken Zweibel
It seems that a key idea from NKS is that the world is like information - patterns are information, not particles. Why should we impose our 'old' ideas about particles on NKS? Why not reverse the process and impose NKS on particles? or waves? or strings? It seems pretty patently obvious to me that all physical interactions can be modelled as the evolution of patterns - both physical configurations and forces altering them. Our most determined ideas about the substance of particles doesn't even come from particles - it comes from our electric field configuration interacting with others, to prevent our bodies from merging with other patterns. What we have is an 'unknown' stubborness of information and an 'unknown' set of laws and actions of information - and these could all be NKS. Our brain? A pattern that models external patterns in a world we call awareness. An emulator within the larger NKS activity.
This is very close to the old Buddhist idea that all is mind - the universal mind, actually. But not 'my' mind or our physical world. Just some universal mind, which is perfectly similar to NKS.
Posted by: Jason Cawley
Sure, the idea that the basic constituent stuff of the universe is thought-like is philosophical idealism, which comes in many varieties. Leibniz was an idealist in this sense. His solution to the problem of mind was to put mind in "at the bottom", as the basic substance, and build everything else out of mind-like substance. This was not original, though his monads are, as a kind of cross between that idea and a kind of atomism (prior atomism had been materialistic e.g. the Epicureans).
Idealism does not have a problem explaining the phenomenal appearence of thought. It runs into plausibility problems explaining the "dead and simple" - also in optimistic versions (Leibniz's "best of all possible worlds") into the problem of evil. Leaving the second aside as remote from NKS issues, one might ask why the supposed universal mind is so fond of vaccum background, single proton-electron pairs, with a remainder for matched sets of two each proton neutron and electron, and some photons bouncing between the above. Which as far as we know is most of the universe. Are these just the cutest ideas you ever did see? Or is simplicity favored for some other, much less "mentalist" reason?
To my Platonist but realist rather than idealist outlook, form is the right intermediary concept - what you are calling pattern. Form is thought-like, yes. The reverse is also true - thoughts may be correspondance patterns. Equivalence is a relation having two sides. There is something thought-like in the simplest things, or abstractable from the simplest things, I quite agree. The question is whether something like thought can and does arise from form-like substance, that does not itself have all of the characteristics of thought.
As I put it in the thread title, either thought is an emergent property of what does not yet think, or goes all the way down. The latter is idealism. The view that the basic constituent stuff is form-like need not entail idealism. To me the superior attraction of a form-based realism to idealism is that it faces the problem of the specific difference of the thought-like, or preserves rather than effacing the common sense distinction between what thinks and what does not. This does mean, however, that my preferred form-based realism has an additional problem that idealism does not set for itself or purposefully avoids - explaining how something like thought can emerge as a global property of constituent parts, that do not each think.
Idealism doesn't have to explain how thinking can happen. It puts that in at the bottom and has nothing more to explain on the subject. To me that is something of a philosophic dodge, but I understand the consistency of the view. It does have to explain how something simple, dead, and unthinking can at least appear to happen. When Leibniz faced that issue he invoked "stupor" - rock monads are supposedly dazed and slow compared to live ones. I find this sort of argument a bit silly. I don't doubt that idealism can be improved post Leibniz, but it remains an issue.
Form based realism sets itself a harder explanatory task, one it is entirely fair to say it is farther from actually fufilling. It has to explain how simple forms can be compounded into something so special that it thinks - that is, something that not only has the minor thought-like-ness of "two" or "triangle," but the full thought-like-ness of an actual mental event. How thought emerges as special and different, from constituent elements that are "as thought-like as form", but not full blown idealist thoughts.
I don't think NKS decides between the two. It is compatible with either philosophic position. I think the special emergent characteristics it has noticed suggest - but do no more than suggest - that the emergent answer is more plausible than it might have seemed before. I mean, the cells in a CA are not universal. The overall behavior is. Universality is a pretty remarkable phenomenon, and to see how soon and how simply it can arise to me suggests our intuitions about how supposedly impossible it must be to get from elements of character A to wholes of character B can be very far from correct.
I think nature can do the trick, because I see it being done phenomenally, and actual proves possible. (I mean, to me humans think and they are made out of elements that do not). But I know I am begging the question when I say I see it being done ("elements that do not"). Since I don't know how nature does the trick, I recognize there is a gap, an unknown, in my view of the matter. To me that is not a significant objection. I'm just leaving room for a later discovery I expect, admitting in the meantime that there is something I don't know or can't fully explain. I think that accurately reflects our presently limited knowledge about the world and about what thinking is. But anybody might prefer the idealist resolution of the issue, because it does not leave such a gap.
Posted by: Ken Zweibel
very interesting.
Yes, I agree, universality coming from simplicity is a big result and suggests that thought could emerge in parallel just due to the surprising growth of (what?) physical complexity.
(By the way, I always felt that the best analogy of CAs in terms of complexity is to the number line; CAs create complexity because they are individual digits trying to do the job of a continuous real number line. For them to map to a real number line, they must mimic somehow its infinite complexity by never repeating over an infinite length of time/updates.)
I am taking the other tack - that the original forms of the CA are being/nothingness, and they naturally lead to intelligence and thought - with perhaps the complexity of 'self' emerging along the way, and perhaps rational thought. But awareness all the way down.
As to the issues of hydrogen and helium, etc., two responses: NKS has the same problem (why a 'few' simple forms that can be used as a foundation for our world) instead of the eye-jarring complexity of a universal CA landscape; and mentalism at the near-planck-length sizes of being/nothingness automata could yield some interesting competitive/collaborative tensions among these aware bits that would result (as they do, here, at larger sizes) in some larger, cooperative forms becoming attractive (like organs and bacteria and cats and people, at our sizes, but leptons and quarks and photons at their level). Sort of evolution at the on/off scale, evolving up through our scale, with various motivations/pressures/forces at each level.
I am not a fan of thinking rocks; but I see them much as NKS does - as a form of intelligence of their own sort; and given that, their ongoing dynamic status should be sufficient to describe their state of being/nothingness. For me, the problem resolves to 'why do I think I am this object'? and not some localized sum of objects, or all of them. This is an age old quandary, answered differently many times; my ansatz: because this object/body/brain/sensory system is much louder and speaks English.
Posted by: Ken Zweibel
and to answer my own question a little better: the on/off of being/nothingness at the CA bit level is awareness at its most primitive. As entities grow in complexity in the world (real or CA) these 'atomistic' being/nothingnesses become better organized, and nerve cells, senses, and brains evolve; as integral to these complex cognitive devices, the automata provide the awareness that these complex devices organize and manifest as memory, imagery, thought, thinking, and being self-aware (of this animal). The property of awareness is universal; the self-awareness emerges with the local complexity of the body and its parts.
So thinking may be emergent; but awareness (the automaton of being/nothingness) is the atom from which it emerges.
Posted by: Jason Cawley
One can have any variety of the idealist view, the question is what reason there is to favor any given one of them. If thought can be an emergent property, why can't awareness? To common sense, awareness seems a restricted attribute of only certain (rare, living) things, just like thought does - if a somewhat wider set of them. I am not saying it can't be the case, but I don't see the reason to suppose it is the case.
Leibniz tells us his reason, we don't have to guess why he thinks it has to be that way. His reason is the fallacy of composition - he doesn't see how some wholes can have the attributes of thought without each part having that attribute. But that just doesn't follow. It is his reason, it just isn't a good one.
Otherwise put, I understand the conception you are proposing. It is similar to others many idealists have proposed. Now, tell me why I am supposed to think it, why it is more plausible than the alternatives. (Particularly, in my case, a form-based realism i.e. the idea that real pattern-like or form-like things exist, independent of mind, ours or any other, and are the formal basis for thoughts on the one hand, and objects on the other).
Posted by: Ken Zweibel
thoughts or intelligence don't seem like the hard problem to me - consciousness or awareness do.
Your argument about Liebniz breaks down because you take salt to be so different from sodium and chlorine. Yes, some properties do emerge from priors that do not have those properties - salt has different properties from sodium or chlorine, separately. However, both sodium and chlorine have the basic properties of being atoms, having electron orbitals, having chemical properties in general. Try to produce 'chemical properties' where none exist before (and if you say: use subatomic particles to make atoms, I will say - they also have antecedent properties that lead to chemical properties when properly combined.). The point is: some sort of related properties are needed to make new properties occur like the old ones. Where is there a 'prior' of awareness or consciousness? There is always some 'prior' in all theories - and 'matter' or 'waves' is the prior of physics - whether we call them strings or particles or atoms.
Thought seems trivial to me because it is just sound-feelings associated with word-symbols; and these harken back to mental manipulations, most likely emerging with or without freewill from processes in the mind/brain - NKS or non-NKS. Our awareness gives them 'color' or presence, just like our awareness gives NaCl a 'salty' taste.
Why should you be convinced? You shouldn't. Just as a scientist needs always to be ready to revise their particle/wave/string/brane picture of the universe. No one can convince anyone of these things, because they are just models, with their own measures of efficacy and usefulness. I don't believe it; I just think about it as possible. But I do take your comment (as probably intended) to bolster my arguments and possible insights. Thanks.
Posted by: Jason Cawley
"Where is there a 'prior' of awareness or consciousness?"
Form. The basis of awareness is the quite general phenomenon of copying, or the transfer of an abstracted shape from one formal system to another without any need for comparable underlying components. Form is an abstract but perfectly real and transferable aspect of all systems. It is the underlying "thought-like-ness" that thoughts emerge from. Nature has simply solved the problem of getting copies of outside patterns inside. Structures of neurons capable of universal computation can mimic any possible external form. Patterns are already instrinsically "thought-like". Thought is a phenomenon made from such "abstract" properties - according to the hypothesis of the philosophy of form.
Consciousness is a continual representation of integrated abstract mimicry. Consciousness is not form, but it is of form and it is made out of form. Consciousness is not mimicry, but it is made from mimicry. That is the prior constituent character in the parts that is exploited by a structured whole, as salt gets its character from electrical interactions. Thought gets its abstract character, its "softwareness", its independence of particular underlying constituents, from formal interactions. The point of the salt example, incidentally, is simply to illustrate that the fallacy of composition is in fact a fallacy - a matter one would think needs little establishing.
It might be objected (not that you have, mind, I am simply meeting one obvious potential objection to the above idea), so and so defines consciousness as X, and form as a basic constituent element does not appear in his definition, so it can't be adequate as an underlying property. But we must beware of mistaking separations posited by our own distinctions for ways nature must do things. Light of a wavelength of 650 angstroms may be defined in terms of a meter stick and a diffraction grating. Red light may be defined in entirely different terms of subjective evidence. But we are not justified in regarding them as two different things in nature merely because our definitions approach it from two distinct subjective points.
The experience of reading the meter stick is just as subjective as the experience of the color red. The two experiences are nevertheless feeling parts of the same natural elephant, not of two different things. Similarly, there is no requirement that someone's definition of consciousness must be based on form, for nature to actually create consciousness out of a basis of form. How nature does it is an empirical not a definitional question. It is naturally the same elephant, or it isn't, and the hypothesis is correct or incorrect accordingly.
Of course it would be a far more convincing hypothesis if we knew how to reproduce nature's success in creating consciousness, which I acknowledge we don't know how to do at present. Which makes the idealist alternative seem plausible. I just don't think it will eventually turn out to be correct. I think it is an explanatory dodge that evades the difficult empirical question, how does consciousness arise from things that are not conscious? That consciousness does arise from things that are not conscious, I believe from common sense evidence. Knowing common sense is sometimes wrong and that my impression that rocks (and atoms) aren't conscious somewhat begs the question, or is no more than an intuition.
One further point. On the hypothesis that everything has awareness or awareness potential all the way down, bits of silicon have awareness or awareness potential all the way down. So on this view, why aren't computers fully conscious (or are they, supposedly)? If they are not, then some additional trick is required for consciousness to arise. Nothing additional is explained. An explanation for how consciousness specifically arises, such that it is present in us but not in our computers, is still wanting. Since the "all the way down" position does not on its own make a distinction between rocks and computers, nor between computers and us, it cannot help explain specific differences between each pair of them. It still needs the missing science of the arising of consciousness, to deal with all the phenomenal differences seen among them.
Good stuff, thanks for the discussion. I do think these are the two plausible alternatives on this question, and that we've explored them here pretty well.
Posted by: Ken Zweibel
I will spend some time thinking about copying before saying anything - interesting idea.
You pose a good question about computers - but wasn't it Wolfram who said that artificial intelligence will come from training, not just from making a fancier computer? I think the ideas is that everything is modestly aware, but not aware of their own awareness, let alone of themsleves as selves. Social interaction and training provide most of the motivation for those further edifications. I am saying that some sort of self-image software and training in various actions (including those that require innovation, imagination, and leaps of faith) would help make a computer more aware. But still not very much.
Perhaps the open-endedness of the CA is what is missing from a computer - it needs the ability to do things differently instead of the same every time. It's a good question: what is missing? (actually, CA does something the same every time that appears different...)
But I obviously have not absorbed your idea about copying form being related to awareness. The part about bringing form 'inside' makes sense to me (emulation, as our 'world of awareness' ; we step in our mind as a way to step in our kitchen); but can't emulation exist without awareness of it? We construct it without awareness - except sensory input, which could be unconscious. I assume a computer can emulate quite well, if programmed to do so. But it's not conscious of its emulation - yet we are, and it's a good tool of learning for us.
I'll stop rambling here. Thanks for the discussion. More later, if you can stand it.
Posted by: Jason Cawley
Just a related item for those interested in this subject - Gregory Chaitin gave a paper last year on Leibniz that others here might be interested in. It is on the web at his site, here -
http://www.cs.auckland.ac.nz/CDMTCS.../kirchberg.html
The focus is on the implications of Leibniz for algorthim thinking, not on the phenomenon of mind problem. They are somewhat related though.
Posted by: Jason Cawley
Back on the emergence of consciousness problem, I want to add one additional thought that I see as supporting the "form" side, though it doesn't necessarily contradict idealism.
It is the notion that human minds aren't simply limited to internal experience. Instead they make use of exterior aids, reading back patterns from them. As a recent writer dealing with a limited problem of his own in what amounts to philosophical anthropology puts it,
The human mind is a self-externalizing phenomenon that creates its places of retention... - RP Harrison
By retention he means a kind of real dependence on a prior cultural form, which need not rise to the level of actual memory.
So, I just add to that that this works, that exterior aids are sufficient. They share enough in common with the interior contents of mind, as simply formal patterns, that they allow thoughts to be at least approximately reproduced. Writing works. You don't have to "write in neurons" to be understood. And before the level of writing systems, there is language. If the metaphor is allowed, the "hard disk" doesn't need to be made of the same stuff as the "RAM".
A simple enough point on a practical level. To me it suggests the formal nature of systems on both sides is "mental enough" to start working up from.
Now I don't think that means we know how nature does it, simply. Formal patterns in exterior things are not consciousness. Consciousness might begin with a looping system "noticing" matches between internally stored patterns and sensory data, or perhaps universal level complexity in relations between both (rather than only "noticing" "exact matches"), or it might need considerably more than either. I don't think we know. On our salt analogy, I don't think we understand electronegativity and ionic bonds yet. But maybe we've got something like "particles" and "charge", in notions like "form" and "copying" (or "pattern matching").
Forum Sponsored by Wolfram Research
© 2004-2008 Wolfram Research, Inc. | Powered by vBulletin 2.3.0 © 2000-2002 Jelsoft Enterprises, Ltd. |
Disclaimer
vB Easy Archive Final - Created by Xenon and modified/released by SkuZZy from the Job Openings