Memes Are Science

Having socked it to memes some time back (MemesShmemes) I'd like to propose, or at least borrow, a suggestion permitting us to say that MemesAreScience.

Simply, let's define a Meme as a logical mapping from a pattern of stimulus to a pattern of response.

Religion, for example, is held up by the memeticists as a meme. Okay, so let's quantify that. Distinguish the pattern of stimuli, the pattern of reponse, and the mapping from one to the other. Do it mathematically. Then you can start to taxonomify religions, cults, fads, slogans, and so on. Trace memes in non-human animals and plants too. If a meme can be distinguished, then replication of same, and MetaMeme?s and all that, could be described scientifically. Theories concerning the necessary attributes of the patterns of stimulus and response, and the necessary logic of the mappings between them, could be quantified, tested and falsified.

To describe the patterns here we could use, for example, GraphIsomorphisms?. And we could quantize GraphDiffeomorphisms? to describe the mappings. Base the whole thing on the LeibnizianDefinitionOfConsciousness and you'd really have something.

Perhaps, given the apparent feasibility of this, we should not say that MemesAreNotScience, but that MemeticistsAreNotScientists? ... --PeterMerel

If you could actually create those graph isomorphisms then those graphs would be memes and you would be their discoverer. There's no a priori reason to believe this is possible so until it actually happens MemesDontExist. Memeticists are all just bullshiters without a single original thought between them and intent on debasing the living mind. In a just world, the kind of persistent lack of creativity and antisocial destructive desire they demonstrate would get them all lined up and shot. -- RK

Graphs are not memes. The process of making a paper airplane is a meme. Calling the process of making a paper airplane a meme doesn't debase the "living mind", nor is it "antisocial". You seem to have memetics confused with something else. -- EH

The process of making a paper airplane is a behaviour. 'behaviour' is a term with a very narrow and well-defined meaning. Behaviours are not "memes" and if you deny that memes are digital encodings of ideas and behaviours (or here graphs) then you have hollowed out the term and destroyed any meaning it could possibly have. Bravo, you've just engaged in language abuse. There's no point discussing what memetics is anymore since you've just destroyed any possible basis for communication. Also, it figures that you would buy into memetics since your mind is a black hole for creative thought; witness your destruction of PM's idea. As for antisocial? The difference between you and me is that I recognize my antisocial behaviour for what it is and I understand what the term means; you do not. -- RK

Mellow out for a moment. You are talking about a kind of graph, not of all graphs. Graphs per se are just as abstract as are numbers. Graphs certainly can model memes (assuming they exist, for the sake of the argument), but they can also model a vast number of other things. In that interpretation, it is true that "[not all] graphs are memes", but in a different sense, yes, "graphs are [i.e. model/represent] memes". -- Doug

Who are you responding to, Doug?

Memes are digital encodings of concepts. Currently there is no known digital encoding of concepts and no reason to believe such a thing is possible. Hence memes do not exist. If graphs can model concepts + value (the things that define stimulus / response pairs) then those graphs are memes. The graphs wouldn't model memes, they would be memes. Exactly the same way that some sets + binary operations are rings, and those sets/rings may model something else. And I'd really like to see something on continuous graphs before accepting that graphs are just as powerful as numbers. Abstract doesn't really give you anything without power. -- RK

Just for the moment, let us skip, for the sake of argument, the question of whether memes exist, because other issues have come up here.

You are of course correct that "there is no known digital encoding of concepts", in the broad all-inclusive sense that would be required for strong AI, but in a weaker sense, digital encodings of concepts are what we do every day when we program. Sure, this is relatively trivial, but if I write a program where 0..6 represents Monday..Sunday, I have a simple digital encoding of a very tiny concept.

Weak AI is successful and highly commercialized precisely because we do know how to create digital encodings for a lot of very simple concepts. Where we have gotten stymied is with the complex concepts.

Graphs have been profitably used to represent more complex, but still simple, concepts, by using graph edges to represent simple relations between graph nodes (which are simpler digital encodings of simple concepts, such as "Sunday"). So we could have a graph represent that concept "appointment sometime before Sunday" as nodes: UNKNOWN, SUNDAY and edges: {from: UNKNOWN to: SUNDAY nature-of-edge: BEFORE}

...And that gives us a graph of the concept "a date before Sunday". (I probably made mistakes in the above, please bear with me.)

Representations of meaning can be built from bases like that to recursively complex levels, encompassing whatever we understand thoroughly enough to encode in graph nodes and edges like that.

There are, of course, vastly more things that we have not figured out how to thus encode, than the things we have figured out how to handle, but still... -- DougMerritt

P.S. When I said graphs are just as abstract as numbers, and you responded with the issue of continuous...(1) I should have said "integers", perhaps, because that's all I had in mind at the time, but (2) I think that there is such a thing as "continuous graphs" in purely abstract mathematical theory, but I am not familiar with that -- but I would like to learn, if anyone has a comment or URL on that. -- Doug

Do you know who was it recently that had an argument about the meaning of meaning? Ahhh, yes it was a discussion from XmlSucks, XmlIsJustDumbText. Anyways, I'm a strong believer in strong AI; I believe that it's possible to construct a digital AI (frames, graphs, it's all the same) which proposition goes beyond mere strong AI. But memetics goes far, far beyond even that.

Memetics claims that human concepts are digital. Considering the complete and utter dearth of supporting evidence that more than a decade of hype has produced in this area, this isn't a proposition that can be left unchallenged. People who persist in passing around this proposition as if it were some kind of obvious fact instead of the unsupported wild speculation it's been explained to them it really is, they deserve to be abused. -- RK

I'm afraid I'm not following RK's thinking. He said "I'm a strong believer in strong AI; I believe that it's possible to construct a digital AI" (which I, too, believe), but then he said "Memetics claims that human concepts are digital. Considering the complete and utter dearth of supporting evidence that more than a decade of hype has produced in this area, this isn't a proposition that can be left unchallenged".

That [RK] confuses me, even when we exclude the subject of memes, for a later date. You believe in strong AI, but not that "human concepts are digital"? To me that sounds like a contradiction, but obviously it is not a contradiction to you, therefore I am misunderstanding you. Please clarify. -- Doug

The strong AI proposition is that it's possible to construct an intelligence that can learn about and function in the world as we do. The digital AI proposition is that we can do this without using quantum, continuous or even pseudo-continuous phenomena; just plain old symbolic logic, or symbolic logic augmented only with blah, or only with blah and blah. The memetics proposition is that each and every concept which occurs in human minds is reimplementable in a digital AI; and if we treat the gene analogy strictly then we use the term 'digital AI' in the strictest possible sense of digital. What I'm doing is not discounting the possibility that there are some meaningful, though on the whole unimportant, differences between a digital intelligence and a biological intelligence. If concepts in biological minds are not digital, it may not be possible to crack them and that's a rather important theoretical consequence of a practically unimportant difference. It's the difference between SyntheticBiology being able to create life using engineering and being able to reverse-engineer evolved life. -- RK

"In that interpretation, it is true that '[not all] graphs are memes', but in a different sense, yes, 'graphs are [i.e. model/represent] memes'."

RK said "[i]f you could actually create those graph isomorphisms then those graphs would be memes", not that those graphs would model or represent memes. I still don't think RK understand what memes are. He's confusing levels of abstraction. -- EH

Since it's obvious you're incapable of following along, which requires creative interpretation of what people say, I'll do the breakdown for you:

  1. Peter observed in MemesShmemes that "unit of culture" is a meaningless tautology saying that the Joule is "a unit of work" doesn't define what the hell a Joule is. Newtons * meters does.
  2. I observed that the meaning of 'meme' has actual content radically different from the purported definition
  3. I observed that this meaning which always lurks in the background involves digital encoding
  4. I analyzed exactly what this means
  5. I observed that MemesDontExist
  6. I acknowledged Peter's graph proposal as an example of memes if it's successful
  7. Doug pointed out that graphs are completely general and, only the discovery of a subset of them as memes would be meaningful

You of course prefer the meaningless propaganda non-definition of the term 'meme', the definition that doesn't enable any meaningful communication and that in fact impedes any possible meaningful communication. Why do you do this? Because you use words as a game which people can win or lose, and you observe that the memeticists have also been playing that game "successfully" so you respect them for it. -- RK

I urge you not to flame EH so quickly; he strikes me as being in a relatively mellow and open-minded frame of mind. Yes, he criticized you, but it struck me as if he was simply trying to understand you despite his logic on the subject, this time. I have criticized EH for not doing that in certain past circumstances, so I must applaud him for doing so this time, if you see what i mean. -- Doug

Dealing with Eric is energy-sapping and annoying. He may be open-minded about this one particular subject but he's not open-minded about the particular issues that make it so energy-sapping and annoying to deal with him. His creative center is utterly barren and his overall attitude to others is lousy. Until he confronts those issues, he will continue to be consistently and unremittingly energy-sapping and annoying. Which doesn't mean it's always unrewarding to deal with him, but it almost always is. Note also that I used a double negative (not unrewarding) deliberately. -- RK

Um, imho it's more productive and less unfriendly to give all comers, no matter what their previous frippery, sufficient rope to tie their shoelaces together. --Pete.

I've known Eric off and on for years now. He's not some random stranger. He's acted in a consistently bad way from the very first conversation I ever had with him. I'm a long, long way past the point of having any doubt, reasonable or unreasonable, which he may benefit from. -- RK

And let me take the opportunity here to say 'I told you so' because I told you so. Eric has managed to lower himself to meet my rock bottom expectations. I have higher expectations of toddlers. -- RK

Well, I remain hopeful. Stop whacking him on the top of the head with the telephone book and let's hear his story. --GoodCop?

He doesn't have a story, just a repetition of the standard dogma about memes. To wit, that memes are the equivalent of Mendelian traits (a concept of genes nobody uses in public anymore, for good or ill) and that we should wait a few centuries to find the equivalent of DNA and RNA for "memetics" but we should still accord it all the respect of genetics. This con game, ploy, and power grab, is something we both rejected years ago.

Memetics only gets the benefit of the doubt so long as its inherent equivocation isn't understood. Even then any intelligent person should be suspicious of its less than stellar academic reputation and its transparent association with a winner. Eric is not only not suspicious of the power grab but even after the equivocation is explained to him, still demands we pay allegiance to memetics.

Eric doesn't need to be indulged, he needs to be whacked on the nose and told BAD DOG in a stern voice until he learns not to piss on the carpet. Maybe he never will and when you get tired of having a carpet that looks like it's been tie-dyed by someone with a fetish for yellow you'll take him back to the pound and they'll put him to sleep ^W^W^W^W give him to a nice family out in the country with a big yard. -- RK

JanuaryZeroSix ThreadMess


View edit of January 10, 2006 or FindPage with title or text search