Re: memetics-digest V1 #1464

Date: Tue 27 Jan 2004 - 20:18:17 GMT

  • Next message: Douglas Brooker: "Re: memetics/memics/mimetics"

    [Kenneth van Oost:]
    > Not only he [Bruce] is chopping populations (where a memetic
    > process could emerge) into credible... and non-credible
    > individuals, ...he argues that their behaviour should
    > (NEED to) be credible in the first place! That troubles me
    > What is a credible individual_ an intellectual or someone
    > working the rice- fields of Cambodia !?

    There are many different definitions of "credible". In this case, I think Bruce is bang-on. Let me explain why.

    In the stock market, there are "sophisticated" and "naive" traders. Sophisticates follow the justification for values, whereas the naive follow "apparent" trends (and by their actions create them). The distinction in the description of trend-following is important: only those trends that are supported by more than belief persist for longer than the short-term. Credibility in the creation of market movements need only be apparent, not real. If you bet on the winning side, and you made a lot of money, your comments have credibility in the future. I agree, therefore, based on evidence from the stock market, that successful memes are -- under the definition -- necessarily generated by credible sources.

    (Note that this ought to be compared with sustainable memes, which are generated by sources with "real" and "justifiable" credibility. Interesting that mutual funds always caveat that their past performance is not an indication of future success.)

    There are lots of relevant and useful examples for an application of market theory to the study of memes. The one I'll use, since I can draw on something I wrote previously when trying to explain another concept, comes from a recent issue of the New York Times magazine. "Futures Markets in Everything", by Noam Scheiber, describes the failure of a Pentagon initiative to gather more timely data about national security threats:

    > The idea was to
    > find an efficient way of collecting a wide range of expert
    > opinion about threats to national security. To do this, the
    > Pentagon created a futures market -- a cross between a
    > betting pool and a stock market -- in which government
    > officials and other experts would buy and sell "assets,"
    > or futures, linked to developments in the Middle East. An
    > asset might be, say, an imminent terrorist attack in
    > Jordan. If experts believed that such an event was likely,
    > they'd snatch up the asset, and its value would rise.

    Theoretically, this probably would have been the "most efficient" way to collect and analyze the likelihood of terrorist events... but there was a problem in the implementation: They didn't adequately restrict the market to protect its efficiency, instead allowing irrationally exuberant (i.e., "naive") traders to take over and move the market away from "reality". The resulting intelligence data was not "real" or "justifiable" or "true" or "valid", despite its inherent popularity and apparent "credibility" at the time. (Remember, the US and Britain went to war over this data, which meant it had to have apparent credibility... but now Tony Blair may lose his job as a result.)

    > The original model for PAM [Policy Analysis Market]
    > was limited to people with access
    > to the best available information. But thanks to some
    > bureaucratic red tape, the idea of using government
    > analysts was junked in favor of opening the market to the
    > general public.

    No wonder it didn't work... you can't open this kind of thing up to the general public, the proverbial Cambodian rice-farmer, and still expect it to accurately reflect reality. It would be like asking a random person on the street, "How many words were in the second draft of Richard Dawkins' "The Selfish Gene", including the footnotes and photo captions?" Nobody would know, although, before its publication, if you had asked enough people you would still get a distribution around some centre mean. This would NOT be a reflection of the true number of pages, and therefore it would not be a "valid" predictor of some future event (say, a range of reasonable prices for the hardcover edition of TSG), but rather it would be an indication of what the average person thought a reasonable length might be. (Given what Hussein said in 1991, was it not reasonable to expect to find WMD in Iraq? Maybe... but even I predicted that it was a bluff aimed at protecting Iraqi sovereignty.)

    Blaming Admiral Poindexter for PAM's failure is ridiculous; whoever pushed for the inclusion of "non-experts" was ignorant of the sources of inefficiency for markets, which I would argue may be the trouble here too. What you absolutely do not want in a system for assessing the value (as likelihood) of "national securities" are bubbles of the kind that occurred in 1633-1637 (Tulipomania), 1720 (Mississippi land boom), 1850s (railway mania), 1925 (Florida land boom), and 2000 (Tech). These all involved people who lost sight of "real" and "justifiable" value, choosing to focus instead of dreams of an indeterminate future; they involved people who were infected by, and then -- because it was rational to do so -- chased the implications of, a particularly virulent meme that wasn't grounded in reality.

    If your purpose is to lead people down the garden path, perhaps because you're selling an anti-tobacco campaign, then by all means include the Cambodian rice- farmer. But an "efficient" market in the case of Poindexter's PAM would not have included the maximum number of traders, but rather the maximum number of knowledgeable traders. There's a big difference. This goes to the "apparent credibility" vs. "actual validity" issue.

    > Congress didn't see it that way. This summer, shortly after
    > an enterprising Senate staff member came across a sample
    > bet -- on the assassination of Yasir Arafat -- on a
    > Pentagon contractor's Web site, two Democratic senators
    > held a press conference to denounce the idea as ''stupid''
    > and "morally wrong." PAM and its chief Pentagon advocate,
    > Adm. John Poindexter, were promptly put out to pasture.

    Perhaps it was "morally wrong" to bet on the assassination of a world leader. But it only would have been "stupid" if the resulting analysis were based on the random noise of traders without better information than what's provided by CNN. The key question for assessing the value of PAM is not to hear that assassination of Arafat was included, but rather to ask about what ranked higher than his possible assassination. What was more thought to be more likely? (This, after all, was the point.)

    Unfortunately, by including "everyone" in the decision-making process, the resulting rankings would have been worth as much as a Gallup Poll-driven "death pool". It would have told decision-makers what "the market" was concerned about, which they ought to highlight during the media coverage of Shock and Awe, but it would not provide a detailed outline of what they ought to blow up while on that campaign. See the difference?

    Who but experts would know about some horrifying yet unreported new threat? Including "everyone" means that those few voices with relevant information would get drowned out by all those with a "credible" perspective... but isn't that always the way? Someone always knows that something awful is going to happen (e.g., Mimi Swartz & Enron) but without "true" and "valid" information from people who also happen to be "credible", publicly unknowable events (e.g., 9/11) will be impossible to predict. In this sense, the existence of memes is a real drag. I would imagine, however, that such an observation would also demonstrate a clear and present need for us to understand memes, and therefore suggest a useful platform from which one might seek funding.

    [Kenneth van Oost:]
    > If he was a credible individual, than his
    > social/ cultural heritage will set in, and will
    > influence the outcome.

    That's why I think the earlier warning about the decline of social sciences as "science" is so important to pay attention to: if the social/cultural heritage values "truth" above all else, including ego, then the outcome should be representative of reality rather than some weird but persuasive memetic message. If your social/cultural heritage forces you to align with reality, then will your communicated outcomes not also be aligned with reality?

    [Kenneth van Oost:]
    > If I sit down on the bench marked Whites Only, by the
    > act of sitting on it, I propagated, replicated racistic memes.
    > The memeplex, so to speak, is in the bench, not in my head.

    Ah ha! Not necessarily so. The nature of the meme (and memeplex), like "credibility", is defined in the interaction. Someone has to see and understand and be infected (or at least affected) by the racist meme for it to spread.

    Let's simplify so we can go beyond the contextual cues that Lissack has already outlined: What if it were a racist newspaper you were looking at, instead of a park bench? Where would the "meaning" be?

    Yes -- the potential meaning is inherent to the newspaper, just as it is to the park bench. Yet, the "decoding" and "interpretation" of this meaning -- indeed, the very nature of the newspaper's "newspaper-ness" -- requires that a person with the relevant knowledge, skills, and abilities exist to read it.

    If there exists someone who can perceive the relationship between paper and ink, where the memes live, then a newspaper has meaning-as-experience. However, this felt-meaning is only a "potential", just as the "experience of sound" if a tree falls (in the woods when nobody is there to hear) is also only a potential. Someone who can "understand" -- who is capable of being infected by the meme -- has to be there for the experience to occur. What you feel is necessarily tied to what you are equipped to feel. (In this way, to extend Dawkins' comment about cultural evolution from The Blind Watchmaker, memes qua memes don't exist for books or computers but are embodied and carried by them as potential experiences for people.)

    To a blind person, a newspaper has no meaning except its functional use as a door-stop or a firelog. Only when someone else reads *aloud* from an individual newspaper, thereby translating its meme-ness into another medium, would it become distinguishable from any of the set of existing newspapers with similar tactile characteristics. A newspaper on its own therefore has no intrinsic meaning: while you wouldn't be reading the news without the paper and ink, you also wouldn't be "reading" without the equipment to delimit the embedded symbols and the "reference book" (to paraphrase Searle) to decode them and give them value. This is both biology *and* social context; being stuck in the middle isn't such a bad thing.

    [Scott Chase:]
    > What sort of criterion is hits on google beyond being
    > a popularity contest? Does popularity translate to
    > validity?

    As a tool to measure the "influence" or "health" or "selectedness" or "success" of a meme, I think Google's index is our most powerful tool. However, I also think you're right: popularity doesn't mean what the meme describes is true. A high PageRank doesn't indicate "truth" or "validity".

    To update Keith Henson's Google numbers:
    "(memetic OR memetics)" yields 173,000 vs. 137,000 for "memetics" alone
    "(memetic OR memetics) AND meme" yields 30,500 vs. 21,700
    "(memetic OR memetics) AND Dawkins" yields 10,200
    "(memic OR memics)" yields 5,390
    "(memic OR memics) AND meme" yields 227
    "(memic OR memics) AND Dawkins" yields 117
    "(mimetic OR mimetics)" yields 83,600 vs. 18,400
    "(mimetic OR mimetics) AND meme" yields 2,040 (or 7% of "e" version) vs. 274
    "(mimetic OR mimetics) AND Dawkins" yields 730 (or 7%)

    Something interesting...
    "(memetic OR memetics) AND meme site:edu" yields 589
    "(mimetic OR mimetics) AND meme site:edu" yields 175 (or 30%)
    "(memetic OR memetics) AND Dawkins site:edu" yields 466
    "(mimetic OR mimetics) AND Dawkins site:edu" yields 70 (or 15%)

    From the difference in the prevalence between citations for our keywords by generic and academic sources, we could say that the "e vs. i meme" in spelling memetics/mimetics has spread further outside of academe than inside. Is this because academicians are more sophisticated? Or is it something about the intuitive bastardization of the word "meme" by popular culture without reference to the original process that created the word, including ignorance about the meaning of the word "mimetic" and what I believe to be Dawkins' intent in alluding to it?

    I would argue that *this* is certainly a valid use of Google for the scientific measurement of memes...

    Jeremy Burman

    =============================================================== This was distributed via the memetics list associated with the Journal of Memetics - Evolutionary Models of Information Transmission For information about the journal and the list (e.g. unsubscribing) see:

    This archive was generated by hypermail 2.1.5 : Tue 27 Jan 2004 - 20:30:11 GMT