A person who calls him-or-her-self "Nikman" showed up with a Comment at A Previous Topic on this Blog, and had some great things to say about our regular participant, Prof. Howard Pattee.I think this deserves a new Topic. Here is what we have so far:
nikman said... I stumbled across this blog for the first time a few minutes ago and discovered that Howard Pattee is posting.
I don't know what more to say right now. Except that there should be a system in place enabling us to pick his brain regularly on a vast array of topics relating to the philosophy of science. If he doesn't want that he should have posted using a pseudonym. You reap what you sow, dude. [Nickman] February 6, 2011 9:23 PM
Ira Glickstein said... Thanks Nickman for your positive comments on a regular here, Prof. Howard Pattee. Howard was Chairman of my PhD committee at Binghamton University. I consider it a real privilege to have him as a continuing positive influence in my life.
Please let us know how you know Howard. We can always use intelligent and courteous Comments here, so please join us.
Ira Glickstein February 6, 2011 11:29 PM
nikman said... IIRC [If I Remember Correctly] it began with "Artificial Life Needs a Real Epistemology" which was almost literally a breath of fresh air. At last, something besides all those stultifying functionalists.
The best thing up till then had been another HP, Hilary Putnam, who's still not chopped liver. More recently, of course, the computational complexity folks (Scott Aaronson et al.) have weighed in to the advantage of sanity. Anyway, I particularly love this:
"The problem also poses an apparent paradox: All signs, symbols, and codes, all languages including formal mathematics are embodied as material physical structures and therefore must obey all the inexorable laws of physics. At the same time, the symbol vehicles like the bases in DNA, voltages representing bits in a computer, the text on this page, and the neuron firings in the brain do not appear to be limited by, or clearly related to, the very laws they must obey. Even the mathematical symbols that express these inexorable physical laws seem to be entirely free of these same laws."
You asked. [Nickman] February 7, 2011 11:15 AM
Howard Pattee said... nikman,[Emphasis and notes added by Ira] Please continue this Topic with Comments here.
Thanks for the complement. If time permits, I enjoy having my brain picked. That 1995 quote is indeed what I had thought about for many years. From childhood we are taught that every event has a cause, and that every event is the result of natural laws. In fact, the evidence for both beliefs is weak or mythical.
The evidence is that we (living systems) choose what we perceive as an event and we favor events that can be described as causal and that that can be predicted by laws. We survive by that choice because it is ultimately the result of natural selection. The evidence is that most of the structures in the universe including organism are undetermined by laws and are mostly selected frozen accidents.
We have also been saddled with Aristotle’s logic of causality that can prove the existence of God as the First and Final Cause or the uncaused cause. His logic also implies determinism.
In contrast, modern quantum cosmology sees causality and determinism as just the illusion of good statistics of primordial uncaused spontaneous creations and symmetry breaking. Of course we could still say that God is a (gratuitous) uncaused cause; but how do we know S/He is not just playing dice?
Howard February 8, 2011 9:20 PM
Ira Glickstein
Those interested in Howard's more recent work may want to read:
ReplyDeleteOn our TVPClub Blog
2008 - Biosemiotics the study of how symbol systems control living organisms and societies.
2008 - Has Language Become Parasitic? - From an evolutionary perspective the survival value of even culturally selected human information is not obvious. Our symbol-based technological culture has given humans power over natural forces that can cure disease, increase lifespan, counter the effects of genetic deficiencies, and unleash weapons of mass destruction. None of these achievements look good from an evolutionary perspective.
Scholarly Papers
2005 - The Physics and Metaphysics of Biosemiotics - The origin of life requires understanding the origin of symbolic control and how inanimate molecules become a message.
2001 - The Physics of Symbols: Bridging the Epistemic Cut - Evolution requires the genotype-phenotype distinction, a primeval epistemic cut that separates energy-degenerate, rate-independent genetic symbols from the rate-dependent dynamics of construction that they control.
Links to postings by and about Howard Pattee on our TVPClub Blog.
Happy reading!
Ira Glickstein
To get discussion going here, I've just read Howard's The physics of symbols. As Howard knows, while I respect his intellect far above most other academics, I view the Universe quite differently from him. The most satisfactory sections of his paper, for me, were:
ReplyDeleteSection 11 - Answering Pearson's Question - "It is not possible to distinguish the living from the lifeless by the most detailed 'motion of inorganic corpuscles' alone." [where I take motion of inorganic corpuscles alone to mean the Laws of Physics, at whatever microscopic level], and
Section 12 - The illusion of autonomous symbol systems - "There is a real conceptual roadblock here. In our normal everyday use of languages the very concept of a 'physics of symbols' is completely foreign."
And for good reason (I would say) because symbols are mere, totally physical neuronal configurations in our physical brains, or physical marks on paper or voltages in computer devices, etc., that have no intrinsic meaning other than what we assign to them. A hoofprint in the snow is merely physical, even if a skilled hunter can interpret it to reveal that a deer of a particular gender and weight and velocity and direction passed a certain time ago.
Thought Experiment:
I write a chess program and set it running in a reliable computer. As we know, as soon as we specify the rules of chess, a finite (but very large) number of possible games are immediately specified. So, I set the computer to play each possible game in sequence and then I destroy myself and all else in the Universe.
Imagine that, unknown to me, someone has programmed each chess piece to be sentient, with the ability to observe the activity a few squares away, and come up with theories and predictions based on those observations. Of course the pieces are constrained to move according to the rules of chess, and only as directed by my chess program.
OK, so a given piece plays in and observes several games, during which he may or may not be captured, and he figures out some of the rules of chess and is even able to predict some of the moves of himself and others.
My chess program is playing every possible game, in sequence, so some games are "logical" in terms of player strategy, while others are "stupid", but all follow a fixed set of the rules of chess. Thus, the sentient chess pieces invent a language and symbol system, and argue with each other about the origin and meaning of life, prove and disprove theories, and so on. As my chess program plays each game, the theories improve, but the "illogical" games throw them off the track of the real situation.
I think this is the situation we humans find ourselves in. We are mere pawns in a finite, discrete, and deterministic Universe of unknown (and unknowable) origin and purpose (if it has such). We are observant and sentient and have figured a lot of it out, but are sometimes thrown for a loop by the occasional "illogical" quantum wierdness and spooky action at a distance.
Ira Glickstein
nikman is a Nick, although must we concern ourselves with that? I'll check out the TVPClub links.
ReplyDeleteBy the way I should've made it clear that the quote I threw in was from "The Physics and Metaphysics of Biosemiotics" not "Artificial Life Needs a Real Epistemology". What's important of course is that it's jaw-dropping in its incisiveness. (Infer "profundity" if you wish.) Descartes, eat your heart out.
What I'd be interested in seeing Howard comment on (putting entirely aside computer simulation virtualism on the order of that Singularity nuttery) is whether he sees a future for the Informational approach in general. (Me, I like it, but you have these crises of faith.) Brukner and Zeilinger's attempt to map QM in information space started off with a bang but seems to have trailed off into the mists. What's Jan Kåhre accomplished recently? (Does Howard believe there's much connection between his own work and that stuff in the first place?) What has semiotics itself ever done for us except provided employment to academics and helped give (say) Clifford Geertz, Roland Barthes and Umberto Eco motivation for distinguished careers?
Ira: "And for good reason (I would say) because symbols are mere, totally physical neuronal configurations in our physical brains, or physical marks on paper or voltages in computer devices, etc., that have no intrinsic meaning other than what we assign to them. A hoofprint in the snow is merely physical, even if a skilled hunter can interpret it to reveal that a deer of a particular gender and weight and velocity and direction passed a certain time ago."
ReplyDeleteMaybe one way out of this impasse is to think of information as an analog of energy (Kåhre does this in his own way).
Energy always comes coded. You can't run an electrical device directly from thermal or kinetic energy. You need a process of transduction. Information always comes coded too. If you can't read the code you don't get the message. Energy "qua energy" is invisible, of course, as is information "qua information".
Lost in the mists again.
Thanks Nick (nickman) for your continuing Comments. I'll give Howard a chance to reply before I get back to you with any detailed reply.
ReplyDeleteIf you would like to become an Authorized Author on this Blog, which will allow your Comments to appear immediately and save me the trouble of Moderating them, please send an email to me, ira@techie.com, with your name, educational background, etc.
All I know about you so far is that you are into Howard's work as well as several other great thinkers and you seem to be capable of courteous discussion of serious topics. Your location appears to be in the Trenton, NJ area, you are using Firefox 3.6, Windows XP, and you may have alerted someone in NY via Facebook to this Topic :^)
Ira Glickstein
Er ... the San Francisco Bay Area, Windows Vista or Mac OSX depending, Safari in both cases (so I can save webpages as cross-platform single files) and so far I've avoided involvement with Facebook.
ReplyDeleteOther than that, bang on. I'll think about upping my status.
Nick: OK now I see you, in the Oakland CA area, with Safari, but my tracking software seems to have confused Vista and XP.
ReplyDeleteWho was that guy in Trenton and his Facebook friend in NY? I guess there are lots of folks who are interested in Howard. If you guys are reading this, please come forward and post a Comment!
We are all looking forward to more participation on this Blog. Thanks for your interest.
Ira Glickstein
Nick asks my opinion of the “informational approach to physics” which I think was popularized by Wheeler’s “It from Bit” phrase. I find it confusing and no help with the measurement-type problems. It is a metaphysical view, largely a matter of style and language, and therefore a waste of time arguing against it.
ReplyDeleteNick also asks what I think of semiotics. I am not a semiotician, and it is not clear to me what the main contributions of biosemiotics will turn out to be. For me, the concept of information is enough of a problem. The analogies of genetic memory and natural language may not carry beyond the fact that they both use discrete, linear strings from a small arbitrary alphabet. So far, we have found nothing in how the brain interprets symbol strings (natural language) anything like how the cell interprets genes.
I take the epistemology of physics and empirical physical theory as fundamental, and my problem is how symbolic activity is related to physics. Here is a discussion about physics and symbols with a leading biosemiotician, Kalevi Kull, that touches the basic issues.
Howard
Ira’s thought experiment about how the universe plays chess takes a good imagination, which Ira has, but I think it is too complicated to clarify the problem. A satisfying explanation is simpler than what it explains. A myth takes a simple event and elaborates on it by adding imaginary features whose existence is empirically unsupportable. Still, myths can be useful for creating testable models.
ReplyDeleteEinstein made the point: “Make things as simple as possible, but no simpler.” You all probably know that this epistemological principle was first stated by the medieval scholar William of Ockham and is called Ockham’s razor “entities must not be multiplied beyond necessity.”
He had other great ideas. Some consider him the father of my favorite subject, epistemology, because he argued that only individuals exist, rather than types, essences, or forms, and that these universals are the products of abstraction from individual brains and have no extra-mental existence. Most brain scientists now accept this as an experimental fact, but philosophers still argue about it.
Howard
The Kull dialog is good stuff and I want to chew on it for a while.
ReplyDeleteIt's interesting that one topic not mentioned is symbol grounding, which may point to the essence of the symbol-physics or mind-matter interaction. (Not define it, just point to it.) I follow Harnad in seeing a child's early experience in the physical world -- identifying objects, making connections, in effect writing his or her own code, which nobody but he or she can do -- as crucial. After that you start committing "symbolic theft" as Harnad says. But what actually happens at that moment when a physical existent in the world -- a rock, a bug -- maps as a symbol in the mind and becomes representation, information? Is it objective "information" undergoing an analog of transduction or will we always need to view the process in complementarity terms? Sort of seems not unlikely.
"It from Bit" by the way shouldn't imply that Reality is binary. It's the human mind that constructs a world from binary propositions because that's how our minds work. Maybe it's our basic code. I believe Wheeler understood that and was being Bohrian: Science is the study of what we can say about the world. Anyway, people who ask questions like "How many bits does the universe contain?" aren't talking about the universe.
Nick and Howard: I too read the Pattee/Kull dialog with great interest. I think Howard's moderate take on biosemiotics is well-founded.
ReplyDeleteThanks Nick for introducing Harnad and "symbol grounding" into the conversation. We should all agree that signs are grounded (for example an aggressive animal showing its teeth as a sign of a threat to bite, or a person holding his or her hand up in the "stop" sign gesture, etc.)
On the other hand, Howard often says that symbols are arbitrary, and thus by implication NOT grounded. I have never agreed totally with that opinion.
Certainly, the symbol characters in pictographic- and ideographic languages (such as ancient Egyptian and early Chinese) are grounded in physical things and real ideas. The characters in alphabetic languages are also grounded in physical things and sounds.
Yes, as these languages developed, and adapted to metaphoric expression, the symbols became more and more stylized and thus more distant from their origins, but the connections, however tenuous, may still be traced to their origins.
Even spoken words in every natural language have semantic origins that may be traced and thus ground them.
The only truely arbitrary symbols are encrypted computer codes, and that may partially explain the difficulties we have had with artificial inteligence, a field I was very sanguine about when I first met Howard, but he, in his gentle way and over much of a decade, managed to pry me away from my confidence that programmed computers could actually be intelligent at the human level.
Nick writes: "'It from Bit' by the way shouldn't imply that Reality is binary ..." That reminds me of a colleague of Howard's at Binghamton U (Walter Lowen) who loved dichotomies. "There are only two types of people in the world," he would say, "Those who believe there are only two types and those who do not. QED."
Ira Glickstein
I suspect you and Howard aren't talking about the same thing. Symbols are arbitrarily selected but once selected they REFER and do so specifically and consistently. Otherwise we wouldn't have language.
ReplyDeleteThe old school (Putnam calls it "magical thinking") of thought on this matter held (or, apparently, still holds) that meaning is somehow inherent in a word per se. But what if there's another planet where people speak Nenglish instead of English and call what we call "peaches" not that but "pears" and vice versa? That's not inconceivable. If you visit there you'll either have to beat the natives into submission or adapt to local custom.
Symbolic representations (other than pictorial or pictographic) are arbitrary, chance choices that become conventions. We tend to think of the moon as feminine but there are cultures where it's masculine. Macho Moon. In Chinese culture turtles are symbols of evil (marketing the VW beetle in China was a serious problem for quite a while). So there you are.
Apparently not a symbol of evil so far as I've been able to google (I thought I remembered that) but to call someone a turtle (or especially a turtle egg) is a serious insult. Apparently it rubbed off on das Auto. I'm positive I read that in Lévi-Strauss and he was making the same point as Howard.
ReplyDeleteNot sure if I actually posted a correction I meant to post. Turtles apparently aren't symbols of evil, but to call someone a turtle or especially a turtle egg is a serious insult. It rubbed off on das Auto.
ReplyDeleteNick wrote: "Symbols are arbitrarily selected but once selected they REFER and do so specifically and consistently.". How can you be so confident that symbols in natural languages or the genetic code have been arbitrarily selected?
ReplyDeleteI mean arbitrary in its sense as a random choice. How did the animals get their names? Did someone put a bunch of random names into a hat and pull them out as each animal came by?
According to Genesis 2:19-20: "And out of the ground the Lord God formed every beast of the field, and every fowl of the air; and brought them unto Adam to see what he would call them: and whatsoever Adam called every living creature, that was the name thereof. And Adam gave names to all cattle, and to the fowl of the air, and to every beast of the field;"
Well, according to this site the Hebrew name for horse has a literal meaning "as leaping", for sheep "as grazing", for eagle "to lacerate" and so on. Of course these animals have different names in different languages and likely different literal meanings in those languages, but they are far from arbitrary in the sense I make of that word.
If you (and Howard) mean by arbitrary that any sound would do for any word associated with a given thing, idea or action, in any language, so long as all people who speak that language agree on the meaning, I would concur, to that extent.
Is the meaning of arbitrary, arbitrary? (Paraphrasing former President Clinton's what the meaning of is, is...:^)
I think an even stronger case may be made for the nucleotides in DNA which code for proteins. Each of the four bases (A, G, C, and T) has a specific structure and triplets of the bases (codons) code for amino acids, and sequences of amino acids code for proteins.
We call these symbols and Howard speaks of open-ended evolution, but there are obviously physical or energy constraints. A machine that cranks out some physical product is made out of gears and levers and bolts and other standard mechanical parts, but we would not call any of these components symbols (except when symbols representing them appear on blueprints of the machine), nor would we consider any of them, or the arrangement of symbols representing them, to be of arbitrary form, would we? So, how can we assume the physical parts (molecules) that code for and make up the physical structures of proteins arbitrary?
A million monkeys on typewriters could put letters together in arbitrary order, and some would form words or even sentences, but this is totally different from how a native speaker puts symbols together in his or her language.
Ira Glickstein
There's syntax and then there's semantics. It's fairly generally accepted thanks to Chomsky and structural linguistics that there's a fundamental, hard-wired syntactical structure of human language and it's universal. It explains why any reasonably intelligent person has a shot at learning the basics of any other language.
ReplyDeleteBut then there's semantics. How do individual words and their meanings emerge? It's fundamentally a game (I hardly invented that idea). It's contextual. Wittgenstein has a riff about carpenters constructing a building, but without going there how do people agree on names for things? Somehow they do. They arrive at consensus. Somebody throws something into the mix and the others either accept or reject it and once it's accepted it stays accepted until something more productive comes along. It's "what works," it's natural selection, which as we know has no natural teleology of any kind behind it. And there's genetic drift. Species can degenerate as easily as they can advance. French say "oui" and Germans say "Ja" and Greeks say "nai" and where's the logic there? But even so it's understood what they mean.
Ira says, I think an even stronger case [for non-arbitrariness] may be made for the nucleotides in DNA which code for proteins. Each of the four bases (A, G, C, and T) has a specific structure and triplets of the bases (codons) code for amino acids, and sequences of amino acids code for protein.
ReplyDeleteHaving a specific structure does not imply non-arbitrariness. Arbitrary in the case of rules, codes, and symbols means that a choice of functionally similar alternative structures exists with no known logical or empirical basis other than chance that one alternative prevails. This is also called a “frozen accident.” As far as what is known of their functions, driving on the right side of the road is arbitrary, your speaking English is arbitrary, and the genetic code is arbitrary. This does not mean that the alternatives are equiprobable or exactly functionally equivalent. Natural selection will decide whether the accident is functionally good enough to be “frozen.”
Howard
Donald Knuth: “There are 10 types of people ― those who know binary and those who don’t.”
ReplyDeleteThanks Nick, and I agree with almost all you wrote about syntax and semantics, particularly the idea that words are naturally selected.
ReplyDeleteWhile memes are not genes, differing greatly in their time horizons and memory mechanisms, there are conceptual and functional similarities worth taking note of.
I totally agree with this part: "Somebody throws [a word for] something into the mix and the others either accept or reject it and once it's accepted it stays accepted until something more productive comes along. It's 'what works,' ..."
But we diverge when you add: "... it's natural selection, which as we know has no natural teleology of any kind behind it."
Teleology has to do with goal- or purpose-driven behaviors and I would argue that the purpose of language is more efficient communications between people, which leads them to naturally select (and reject) new words and other memes. If the selection process is not driven by a purpose (better communications), then it must be totally random, which makes no sense to me.
What do we mean by teleology? An organism (e.g., you or I) makes decisions for the sake of an end, namely maintaining or improving our lot in life. We observe the situation, process it in our brains based on previous experiences and current goals, consider alternatives, and, by consensus of groups of neurons, choose a given action.
Conjoined people (e.g., a tribe or a language group) may be considered a super-organism, and their interaction is very much like the interaction between groups of neurons in a human brain. The situation (new thing that needs a name) is observed, processed by people suggesting names for it based on previous experiences (established semanitcs and syntax), and a decision is made (people accept or reject alternatives until one emerges as the winner).
Even genetic evolution, I would argue, is teleological. Perhaps the origin of life was due to true "random" mixing of long-chain molecules (RNA World), but, once the genetic code was established (probably multiple competing codes), and reproducing prokaryotic cells evolved, each of the successful ones had a purpose, namely survival and reproduction.
Ira Glickstein
Natural Selection certainly has no goal. If enough people end up believing that then the late Steve Gould lived his life in vain. (He had a goal, of course, which was to stop people from believing stuff like that.)
ReplyDeleteI mean species have no goals. What is the goal of homo sapiens sapiens? People live their lives. They make their own history, as Marx said, but not as they choose. Now, as individuals they certainly have goals, or at least needs, as has just been dramatically demonstrated in Egypt, and when those individual sets of needs and goals converge sometimes history happens in a purposeful way. Sometimes being part of a movement helps you define your needs and goals. But movements as such don't have goals even though often the members get together and issue policy statements.
I'm at home today and my wife, who's kind of a masochist, is watching the CPAC conference on CSPAN. A few minutes ago she told me that some ignorant woman had just said, in re the POTUS, that "the world doesn't need a president of Islam." I don't need to join a movement in order to desire to neutralize people like that, but if you want to organize a terrorist death squad count me in.
I don't want to get involved in discussing memetics. Been there too often, done that too much. I'll believe they exist when somebody shows me one.
Well, natural selection per se, may have no goal, but, if you and I (and your wife) have goals, why could we not say that a sports team, as a super-organism of several players, has a goal? Indeed, that is what we call it in soccer, they score a goal, and in football, we have goalposts. Marvin Minsky wrote a book called Society of Mind in which he likens groups of neurons in the brain to layers of agents who, acting cooperatively and competitively, creat our mind.
ReplyDeleteIra Glickstein
Ira, you are not making the crucial distinction between genes and memes. The differences between genetic and human language are as fundamental as their similarities. The defining difference is this: Variation in genetic information must be expressed before selection can begin because natural selection operates only through the phenotype, not directly on the genotype.
ReplyDeleteHuman brains, in contrast, have the enormous advantage of being able to acquire, evaluate, and select information before expression. In other words, humans can think before they act. Evolution by natural selection cannot do this. Teleology implies a goal before acting. Therefore memes can be teleological but natural selection cannot have a purpose as it is normally defined.
Howard
Further comment: Teleology is a concept fraught with ambiguities. The concept arose at human thoughtful planning level where the concepts of choice and purpose acquired their conventional literal meaning. That is why applying concepts like choice and purpose to evolution, even as a metaphor, is confusing. Evolutionists know that selection of heritable information occurs after its expression. To call this teleological would be putting the effect before the cause.
ReplyDeleteTo reduce confusion Ernst Mayr popularized Pittendrigh’s term Teleonomy the apparent quality purposefulness and of goal-directedness of structures and functions in living organisms that derived from their evolutionary adaptation and reproductive success.
Howard
Thanks, Howard, stating that "...memes can be teleological..." Under what conditions can memes be goal-directed? (So we can tell Nick, and also so I will understand why you think memes can sometimes not be goal-directed.)
ReplyDelete***
You state that there is a crucial distinction between memes and genes, which is that gene must be expressed in a phenotype before selection can begin. I concur.
But memes must also be expressed by a person before they can be selected by his or her group. Say Albert looks at a newly-discovered animal that is leaping, and he names it "as leaping" in his language. Then Betty happens to look at the same animal as it is running and she calls it "graceful runner". Neither of these are memes until they are expressed and one or the other catches on among a group of people.
According to Wikipedia "A meme acts as a unit for carrying cultural ideas symbols or practices, which can be transmitted from one mind to another through writing, speech, gestures, rituals or other imitable phenomena. Supporters of the concept regard memes as cultural analogues to genes, in that they self-replicate, mutate and respond to selective pressures."
Meme comes from Ancient Greek μίμημα (mimēma), "something imitated", so it is not a meme until it is expressed and imitated. Thus a meme is a concept or word or song that is shared by a society of people. One person's internal thought is not (yet) a meme.
***
You write "Human brains, in contrast, have the enormous advantage of being able to acquire, evaluate, and select information before expression. In other words, humans can think before they act."
Isn't thinking, per se, even if you do not say or write anything or move a muscle, an action? As you say we "acquire, evaluate, and select information before [external]expression" but is not that internal conversation a form of private expression that is not so private anymore because it can be monitored via brain scan, as millions of neurons fire electrical pulses and shoot neuro-chemicals at each other?
How different is that from what is happening in Egypt right now? The people observed what happened in Tunisia, and some took to the streets while others opposed them and. After a few weeks, a decision was taken to embark in a different direction. Other than the scale difference in time and space, is that not similar to the process an individual human brain follows to make a crucial decision? For example, we observe another person taking a risky action, and succeeding. So, one group of neurons within us wants to follow suit, while another wants to take the less risky path. We may vacillate back and forth before we act.
Ira Glickstein
Ira, it is true that memes to be selected by a group must also be expressed, but unlike gene mutations that are initially meaningless and survive only by slow and inefficient natural selection, memes are usually first created and internally selected by rapid and efficient thoughtful evaluation before they are communicated for group selection.
ReplyDeleteYou state that, “One person's internal thought is not (yet) a meme.” Is a mutation in a gene then not (yet) a gene? You are quibbling over definition. Neural selection (Edelman) may be Darwinian, but its enormously greater speed and efficiency is why model-building brains evolved.
You ask, “Isn't thinking, per se, even if you do not say or write anything or move a muscle, an action?” No, it is not in any conventional (conservative) or legal sense. If you can’t distinguish your thoughts from your actions you know you will get in real trouble.
What’s happening in Egypt is too complex to model or predict. It involves of lot of gene-based emotions, historical and modern memes, religious and political ideologies, individual thinking, and media technology.
Howard
I'd like to ask Howard if he believes in the objective, literal existence of memes. That is, in the way Dawkins and Dennett clearly do.
ReplyDeleteIn "The Selfish Gene" Dawks approvingly quotes Nick Humphrey (who I think has since modified his opinion) to wit that memes "literally parasitize the brain" (boy, that one sure sticks in the mind) which is kind of melodramatic for what's supposedly serious science ....
It's been a couple days since Nick asked for Howard's opinion on "..the objective, literal existence of memes..." and I hope he answers soon.
ReplyDeleteAccording to Wikipedia a gene is "a locatable region of genomic sequence, corresponding to a unit of inheritance, which is associated with regulatory regions, transcribed regions, and or other functional sequence regions."
According to WIkipedia "A meme acts as a unit for carrying cultural ideas symbols or practices, which can be transmitted from one mind to another through writing, speech, gestures, rituals or other imitable phenomena. Supporters of the concept regard memes as cultural analogues to genes, in that they self-replicate, mutate and respond to selective pressures."
IMHO, both genes and memes are abstractions that describe units of heredity passed on to future generations by past generations. Genes are inherited from your mother and father (and the genetic engineering labs :^) via DNA and RNA, while memes are inherited from your early care-givers (usually your parents and relatives) as well as the larger society.
When we say "Jane has the gene for tallness" we are actually generalizing several sections of her DNA that code for a variety of proteins that, given normal fetal development, nourishment, and care, will result in Jane being taller than average. I call this an abstraction because Jane may have more of her height in her trunk and Joan, who is the same overall height, may have more in her legs. (At water aerobics, some women can easily touch their noses to their knees while others, of the same overall height, cannot.)
When we say "Ira has the meme for English" we are again generalizing a subset of words in Ira's active and passive vocabulary, and we are implying that Ira can understand and be understood by most people who are said to share that meme. My (originally) Brooklyn English is not identical to the British or Indian versions, but, again as an abstraction, such as "tallness", the term "English-speaking" has a real meaning.
Any trait or characteristic or behavior, however specific or general, that is mostly inherited via genetic (DNA, RNA) processes, may properly be called a gene.
Any trait or characteristic or behavior, however specific or general, that is mostly inherited via memetic (imitation) processes, may properly be called a meme.
Both, IMHO, have "objective, literal existence".
What say you Howard?
Ira Glickstein
Has anyone detected such a memetic structure in a brain? It's physical, after all. What does "life after death" LOOK like? I really, sincerely, honestly want to know!
ReplyDeleteTSG, Chapter 11 (I couldn't make this stuff up):
"Just as genes propagate themselves in the gene pool by leaping from body to body via sperms or eggs, so memes propagate themselves in the meme pool by leaping from brain to brain via a process which, in the broad sense, can be called imitation. ... As my colleague N.K. Humphrey neatly summed up an earlier draft of this chapter: '... memes should be regarded as living structures, not just metaphorically but technically.(3) When you plant a fertile meme in my mind you literally parasitize my brain, turning it into a vehicle for the meme's propagation in just the way that a virus may parasitize the genetic mechanism of a host cell. And this isn't just a way of talking -- the meme for, say, "belief in life after death" is actually realized physically, millions of times over, as a structure in the nervous systems of individual men the world over.' "
Nick and Ira,
ReplyDeleteYour question is: Do I believe in the objective, literal existence of memes? That is, in the way Dawkins and Dennett clearly do.
I prefer to avoid speculating on what Dawkins and Dennett believe. In my opinion they are both smooth polemicists (as popular philosophers must be) but they demonstrate the saying: It is easier to be clever than clear.
The issue is the meaning of “objective” and “literal.” I use the physicist’s concept of objective which means as independent of all conceivable observers (subjects) as possible. So there are degrees of objectivity depending on how much subjective influence there is on the concept. For example, flat Euclidean space is not as objective as curved Riemannian space-time. However, in normal life Euclidean space is more useful.
I would say that genes are more objective than memes because there is less subjective influence on the meaning of gene than the meaning of meme. Test: Which concept generates more subjective arguments? The concept of gene has proven to be essential for evolution theory, even though it is not entirely objective. So far the concept of meme is not essential for any theory I know, but I think it is stimulating for the imagination.
I have a problem with literal as opposed to metaphorical. The problem is that in physics, the most objective concepts are also largely metaphorical. When someone says, “It’s only a metaphor,” they mean, “Don’t take it too seriously.” Not so in physics. Forces, fields, mass, action, charge, spin, strangeness, color are serious metaphors.
In this sense “literal existence” is an oxymoron.
Howard
Nick asks: "Has anyone detected such a memetic structure in a brain? It's physical, after all. ..."
ReplyDeleteAre you familiar with imprinting made popular by Konrad Lorenz working with geese. During the critical period approximately 13 to 16 hours after hatching, whatever the goose sees becomes imprinted as an image of "mother" and the goose, for the rest of its life, will follow that figure. In the wild, it is the goose's actual mother, but, in the lab, geese have been imprinted by a person in a lab coat.
Thus, at least in some birds, there is a particular structure in their brains that becomes active at a certain time after hatching, records an image that I would say is the meme "mother", and then the recorded image can no longer be changed. I call it a meme because the bird will follow and imitate that "mother" image, which is how they learn to fly, feed, and migrate.
Songbirds are born with the ability to learn a local version of their species song. If, in a lab, a baby bird is not exposed to the sound of the song, it will sing a genetically-coded primitive version of the song. On the other hand, if it is exposed to a given rich version of that song, it will learn to sing that version. That is how the birds of a given species who live in one neighborhood learn to distinguish there own from those of the same species but from a foreign neighborhood who must be chased off if they hapen to invade our turf. That songbird brain structure for learning the song meme is active for only a short period and then it remains fixed.
A baby human can learn any human language, or set of languages, to which it is exposed. Language(s)learned (indoctrinated) prior to puberty may be indistinguishable from a native speaker. If learned after puberty, the person will always have a "foreign" accent! Thus, similar to birds, but of course much more complex and less sharp-edged, there is a limited time period for learning to speak any language as a native speaker, which implies a specific brain structure for that meme. Center-city gangs have colors and signs they use to protect their turf. Some of our fellow humans become enraged when they hear foreign accents and languages that sound like "bar, bar" to their untrained ears and they regard foreigners as "bar-bar-ians"
Nick asks about the "life after death" memetic structure in humans. Well, I think it is pretty clear all (or most) humans are "wired" to believe in something greater than themselves. This has traditionally been expressed in the acceptance of spiritual or religious memes, which have been key features of all long-lived societies. For some humans, this brain structure is hijacked by concepts like Extreme Nationalism or Racism, or Humanism, or Communism or Social Justice, or Save the World Ecology (Global Warming activism), etc., but, whether we seek to live forever after mortal death walking hand in hand with God, or by saving planet Earth for future generations, it is the same brain structure seeking after something greater than a single individual.
Ira Glickstein
Nicholas Kristof told a funny story in the NY Times about when he was a kid on the family farm in Oregon and he took a chicken egg and placed it in a duck's nest. Chick hatched, believed itself a duckling, mother duck had no problem either, but then one day she took her brood to the pond ...
ReplyDeleteMoral: Imprinting doesn't prove memetics.
We can formulate Ira's argument as a syllogism:
(1.) Ira believes memes exist and possess properties X, Y and Z.
(2.) Properties X, Y and Z are observed in the world.
(3.) Therefore memes exist.
Anton Zeilinger, my favorite physicist, has a neat definition of photons. "A photon is a click in a photon counter." I'm sure Howard can appreciate that. Another physicist, David Mermin, then asked: "Is Anton a click in an Anton counter?" The joke might not have have worked as well if his name had been Dieter.
But at least you get that click. And it's a physical click. It's a measurement, an observation. No click in the old meme counter yet, however, and we've been waiting a while now.
As I see it, a “meme” is just a vague new word for many types of learned memory. Instead of asking if memes “literally exist,” which is largely a matter of definition, it is better to look at how the many levels of memory evolved, which can be answered empirically.
ReplyDeleteThe oldest heritable memory (genetic memory) is more than a billion years (~10^17 sec.) old and is still expressed in present organisms including humans. Non-heritable memory has a relatively short span and covers time scales from thousands of years (~10^11 sec.) from the origin of writing to individual human memory span that includes long term (~10^9 sec.) and short term time scales, which is in milliseconds for sensorimotor control. Artificial control time scales are now in femtoseconds (10^–15 sec.).
The evolutionary fact is that development of all levels of nervous system activity is dependent on genetic memory structures. That is, what we call the ability to learn (like imprinting) requires genetic competence. So does human memory and language. The distinction between inherited (genetic) and acquired (learned) memory is a classical problem that can only be resolved by experiments. At present no one even knows how to do the experiments, because inherited and learned information are intimately related and cover time scales of more than 20 orders of magnitude.
All acquired or learned memory can be communicated (“replicated”), can be modified (“mutated”), and is evaluated by individuals and groups (“selected”). The only reason I see to use “meme” is to call attention to some similarities with genetic memory. Most scientific interest is in the differences between the two. The question is whether the use of meme clarifies or confuses the empirically decidable issues.
Howard
Howard says non-heritable memory dates from only thousands of years ago ("non-heritable" defined as memory not inherited via genetic evolution). When did the first animals with centralized brains evolve? Probably 300 to 500 Million years ago! We know that insects, such as ants and bees have the ability to learn and communicate, by signs and imitation, the location and quantity of food, which requires non-heritable memory.
ReplyDeleteSo, if genetically-inherited memory (prokaryotic life) originated about 3.5 Billion years ago, that would put the ratio of genetic/memetic at around 10:1, which is smaller than the ratio of the origin of human farmers to the origin of human physicists, and thousands of times smaller than the ratio of origin of farmers/computer programmers!
Kristof's duck, chicken and egg story, recounted by Nick, demonstrates that there is a brain structure in the newly-hatched chicken that accepts the meme "mother" and even a brain structure in the mama duck that accepts the meme "my egg, my chick". The failing attempt of the chick to swim in the pond, despite its best effort and urging by the mama duck, demonstrates the power memes hold over belief, even, as in this case, a false belief.
The power of memes over genes is demonstrated by their cross-species applicability (duck-chicken). Domesticated wolf pups (dogs) accept a human as qualifying for the "leader of the pack" meme, another cross-species demonstration.
Nick believes nothing has real, objective, literal existence unless it can register, like a photon, in a measurement device of some sort. I think there are lots of things, such as love, that cannot be measured quantitively, yet have real, objective, literal existence.
Ira Glickstein
Ira is right about the evolution of memory in brains. It is 100s of millions of year old. I have read about birds that can remember for months hundreds of exact locations where they hid seeds. This is better than most human memory.
ReplyDeleteMy point was that learned memory has all the qualities of memes. The use of meme would appear to be of value primarily to emphasize the analogy with genes (variation, repeatability, selection).
Memory structures and learning in brains is still mysterious. They are somehow distributed over millions of synapses, and so far only “detected” by recall or an action. Nick wants a “meme detector” to demonstrate its literal existence; but would he not accept the literal existence of memory just by recalling it?
Incidentally, there are amazing studies of how early infant learn language sounds at a
TED talk by Kuhl.
Howard
My point was that memory is in fact (in all likelihood) distributed. I'd seriously demand to see a meme only if I believed in the possibility of their existence. Memetics is pure reductionism and that approach may not work any longer even in genetics (see: epigenetics).
ReplyDeleteA paper I like is Alva Noë's and Evan Thompson's "Are There Neural Correlates of Consciousness?".
http://socrates.berkeley.edu/~noe/NCC.pdf
BTW Howard is mentioned in a paper of Thompson's from the last century ("Symbol Grounding: A Bridge from Artificial Life to Artificial Intelligence"):
http://individual.utoronto.ca/evant/SymbolGrounding.pdf
It's pretty good too.
hi Howard et al .. glad to have a moment to post.
ReplyDeleteI've worked independently on the relationship between symbols and physics, and my results may be summarized as follows.
Before one runs a program on a digital computer, it must be plugged into an electrical outlet, because any physical implementation of algorithm consumes energy.
Moreover, if the algorithm is 'exponential', so will be its energy consumption. Further, if either is a polynomial function of the problem size, so will be the other. Hence an isomorphism of sorts.
Cutting to the chase here, we have:
Data:Algorithm:ComputationalSpace:ComputationalTime
=
Matter:Energy:PhysicalSpace:PhysicalTime
There is much to be said supporting such a relation, though here I will offer a bare-bones argument.
Moreover, if the asymptotic computing time of an abstract algorithm is an exponential function of 'problem size' (e.g. number of cities in the traveling salesman problem), then the asymptotic physical time of the physical implementation on digital hardware will be an exponential function of the problem size.
Summarizing, we have four parameters that can be expressed as a function of the problem size:
Energy consumption
Number of algorithmic steps
Physical time
Computational time
If any of the four is exponential, then they all are. If any is polynomial, they all are.
This analysis takes no account of matter, data, and space, but hopefully it gives the flavor of a way of understanding physics and computation that involves a salient structural similarity. Hopefully, it offers a glimmer of insight into the relationship between symbol and matter.
Saliently, this same model can be used to understand other 'domains,' including consciousness and biology. The generalized model depends on the ubiquity of the 'passive/active duality' (coinciding with the space/time duality).
When dealing with (discrete) information, its passive and active forms are, respectively, data and algorithm - which taken together comprise computation, or a computational event.
Th-th-that's all folks ..
Eric
Eric, thanks for your comment. I hope Howard sees it and replies but he has been away from this blog for some time because of some deadlines on a book and some professional papers he is working on.
ReplyDeleteI find your analogy between algorithm (symbols) and the time and energy it takes for a digital computer (matter) to process it interesting. So, we have symbol/matter/energy and time/space (though how does space fit into the digital computer analogy?)
Howard and I went round and round about the duality of symbol/matter. I take the view that there is no such thing as a pure symbol.
For example, a trained human hunter can look at the impression of a deer hoof in the snow (which is certainly totally matter) and tell you the approximate age and size of the animal, the direction and how fast it was running, and perhaps, given several hoof prints, how long ago the animal passed by, the gender and health of the animal. Similarly, a dog can sniff the hoof marks and learn much more information.
So, to you and me, hoof prints are mere matter, but to a hunter and a dog they are rich in symbology.
On the other hand, you and I can (to some extent) understand Howard's academic papers as rich in symbolic meaning, while to a typical hunter and his dog, they are mere paper and ink with no symbolic meaning at all.
At one point, I drew symbol and matter as orthogonal axes, with matter "X" and symbol "Y".
Everything has a material content and therefore some X coordinate value. (Even an algorithm must be made up of ink and paper or chalk and chalkboard, or, if in a brain of neurons and connected axons and dendrites and chemical pathways, or, if in a computer, states of memory locations in a silicon chip or magnetic disk or, if traveling on a wire or wireless medium, of electrons or photons, etc.)
The symbolic content "Y" coordinate is dependent upon the observer or user of the physical content. For example, the hoof print would have very little symbolic content to me (an animal was here some time ago), but a lot of "Y" for the dog and hunter. Howard's latest paper (or this comment) a lot of "Y" for you, and perhaps no "Y" at all for most other people and animals.
Howard's semantic closure (now he calls it semiotic I believe) depends upon the receiver of the symbol carrier having some special ability to decode the symbolic content.
For example, a strand of DNA is merely a long-chain molecule that may or may not be tasty to some other long-chain molecules or prokaryotic or eukaryotic cells. But, to the proper parts of the genetic system, that DNA may be the code for the proteins that make a person or a chimp or a bug or plant.
I hope Howard joins this conversation and that you continue to comment on it.
Ira Glickstein
Ira .. thanx much 4 ur thoughtful reply, & please pardon my delayed response. I also hope Howard will comment. (Howard?) Unfortunately, 'Howard' is a mere symbol (string), whereas Howard is more of a material instantiation of genetic and cosmological happenstance. Though you and I may connect 'Howard' with Howard, there is no assurance that Howard himself does, or if he does, that he will be reading this post - without. conflating 'Howard' with Howard.
ReplyDeleteAs far as the "isomorphism" (loosely defined) between physical and symbol systems is concerned, you ask (I think) what is a digital computer's analogue to space. It is memory - because data resides in memory as matter resides in space. (If this seems fuzzy or imprecise, it's at least partly because the 'official' version is too big for a comment box.)
Now, as regards whether symbols exist, there is something of a linguistic issue here - as there is with other questions regarding the existance of time, god, transcendence, and whaddeva.
The language evolved without reference to these relatively recent novelties of abstraction, so the predicate "exists" doesn't quite cover it. But pick a verb - any verb - coinage or otherwise, and bingo! - you have the requisite predicate. Now with the use of this device, we can all agree now on the ontological status of symbols. Or not.
Thoughts?
Thanx again. It's a pleasure to have a forum for discussing these ideas.
E
Eric and Howard, I am in Brugge, Belgium towards the end of a bicycle barge trip from Brussels to here and do not have time right now for detailed comment on you thoughtful ideas. Keep up the good word. ira
ReplyDeleteIra .. have fun .. sounds pretty exciting ..
ReplyDeleteHoward .. I just found this paragraph from one if ur posts:
"I prefer to avoid speculating on what Dawkins and Dennett believe. In my opinion they are both smooth polemicists (as popular philosophers must be) but they demonstrate the saying: It is easier to be clever than clear."
This captures something I felt for a long time but didnt quite know how to put.
Honestly, a weight is lifted from my chest. Perhaps I could have seen myself that it was the popular nature of these guys' philosophy that prevents them from coming any closer than a rhetorical flourish to the truth.
Perhaps it's why I like it here, on this site, where no agendas exist (that I can see) but sharing & anayzing ideas. Then discussion takes on a special flavor - something akin to intellectual honesty - if that's even possible.
I've thought for some time re: evolutionary epistemology .. - ie, we're not made for truth, though we are capable of it in certain instances.
Bukowski (poet) talks about the inexorable guy at the top with a gun .. I think that's a more realistic model of social organization than the democratic institutions - including open dialogue - that supposedly offer a more civilized alternative to our more primal inclinations.
Cheers -
E
Eric, Ira, and others,
ReplyDeleteI’m sorry I am not doing blogs right now, although I would like to respond. As I explained to Ira, I am recovering from heart surgery (aortic valve replacement and a by-pass). I’m doing fine, thanks to great surgery and cardiac rehab, but I am behind in other commitments. Along with normal daily activities, I have a book contract with the manuscript due Dec. 1.
If anyone is interested, you can Google “Pattee symbol-matter” and download the second entry: EVOLVING SELF-REFERENCE: MATTER, Symbols, And Semantic Closure (from Communication and Cognition - Artificial Intelligence, vol. 12, nos. 1-2, pp.9-28, 1995.)
Eric (and Howard and Eric and Howard :^) - I've been back from my bike/barge adventure in Belgium for several weeks and about caught up with my U. Maryland online class and activities here in our retirement community. (I gave a talk on Hume's History of England a couple days ago and am scheduled to talk on Clean Coal to two different clubs later this week. Being retired means you never get a day off :^).
ReplyDeleteEric raises some great issues:
(1) The relationship between a material instantiation and the symbol that represents it.
OK, let us agree that Howard actually exists in Time and Space as a localized, identifiable bundle of Matter and Energy.
So what about the existence of the symbol (character string or phoneme pressure pattern) Howard? Well, Howard exists as an array of physical neuronal connections in my physical brain (and yours) as a result of the interaction of Ira and Howard when he was chairman of my PhD committee and now as a presence on my Blog.
Howard's ideas, expressed as symbol strings in his talks and papers and Blog postings, have a physical existence in the form of ink and paper, bits in computer memory, electrical and magnetic vibrations as they fly over our WiFi links to the Internet, reside in our brains, and so on.
None of these symbolic representations of Howard carry exactly the same meaning. Eric, Nickman, Ira, and other students of his have each observed different aspects of Howard and we have distorted these images according to our peculiar viewpoints and prejudices. Yet, we each associate our Howard with the "real" Howard as well as the somewhat different symbolic Howard in each of his fan's brains and his papers and Blog posts.
Thus, IMHO, each of the multiple Howards have as real a material instantiation as the one and only Howard.
(2) Eric writes "you [Ira] ask (I think)what is a digital computer's analogue to space." and answers "It is memory - because data resides in memory as matter resides in space."
OK as far as it goes. In the simplest, idealized computer (Turing machine), memory consists of symbols printed on a long physical paper tape, and those symbols are read, altered, and re-written according to a fixed set of rules in the machine. In the real physical world, Matter (and Energy) exists, is encountered by other Matter/Energy, interacts and is altered, and remains in existence in a framework of Space (and Time).
But, the same is true of symbols and the ideas and physical constructs they represent. Symbols exist in Time and Space as instantiations of Matter and Energy, and, as such, they interact, are altered, and remain in existence.
Ira Glickstein