[lit-ideas] Re: Madness, Foucault, Nietzsche & Emerson

  • From: John McCreery <john.mccreery@xxxxxxxxx>
  • To: lit-ideas@xxxxxxxxxxxxx
  • Date: Fri, 9 Dec 2011 18:46:50 +0900

Eric,

I agree with the thrust of your conclusion. I am not, however, holding my
breath. I am personally involved in a major research project, using social
network analysis to motivate historical and ethnographic research on the
Tokyo ad world. The network analysis math works like a dream and the new
tools, software running on fast personal computers, allows me to do things
only barely imaginable when I was a graduate student in the sixties. I can
turn around in a couple of hours analyses that would have taken months of
key punching and humping card decks back and forth back then. At the same
time there remains a huge gap between the discovery of a steeply skewed
curve in the distribution of winning ads for which creators receive credits
(predictable with simple random graph models) and the results of interviews
with some of the top creators. The former tells me that someone must be the
top creator who has won far, far more awards than anyone else. It doesn't
tell me why his name is Hiroshi Sasaki or how the winning ads were created.

A truly lovely example is what is now one of Japan's longest running
campaigns, a travel destination ad for Kyoto whose copy reads simply *sou
da, Kyoto he ikou *(Why not? Let's go to Kyoto). Sasaki himself introduced
me to a TV commercial planner named Ansai Toshio who told me the following
story.

Sasaki and another famous copywriter had been beating their brains for
several months trying to do what copywriters do, come up with clever words
that suggest some new angle on a familiar message. Then, one day, Ansai,
who was eager to get the commercial made, said, "Guys, why not just say *Kyoto
he ikou *(Let's go to Kyoto). Sasaki replied, *Sou da *(Why not?). The same
line has been in use for eighteen years and, according to advertising
critic Amano Yukichi captures perfectly that moment of sudden realization
when a Japanese remembers something that should have been obvious but was
somehow forgotten.

The new mathematical models and the algorithms that implement them enable
us to learn a whole lot that we didn't use to know, because getting heads
and hands alone around that amount of data (in my case, 30,000
relationships linking 4000 ads and 8000 creators)  wasn't humanly possible.
There is still a long, long way to go before they predict the kinds of
events exemplified by the one that Ansai described to me.

I close with a remark from *Complex Adaptive Systems: An Introduction to
Computational Models of Social Life *by John H. Miller & Scott E. Page.
Miller and Page observe that most agent-based stochastic models developed
so far assume agents who are either too stupid (relying on a few simple
heuristics) or too sophisticated (constantly doing elaborate game-theoretic
calculations) to be valid representations of the muddle of feeling,
prejudice and only occasional clear thinking characteristic of human
beings. That's a problem that so far is way, way too complicated for anyone
to have solved it.

I lied. I'll add one more favorite reference, to Warren McCulloch's
*Embodiments
of Mind. *McCulloch was the father of automata theory, a man who devoted
his life to trying to design machines to simulate human neurons. In his
introduction to the book, he writes that he is a man who builds machines
that try to simulate human behavior. Whenever they fail, there are always
those who say, "See, it can't be done." There, however, people like him,
who go off to build a better machine.

Cheers,

John

On Fri, Dec 9, 2011 at 6:19 PM, Eric Yost <mr.eric.yost@xxxxxxxxx> wrote:

> John: In between we have a vast zone where things become too complex
> for straightforward explanation in either simple mechanical or
> simple statistical terms. <snip> The science zone is expanding and
> defenders of what they see as other types of knowing are once again
> on the defensive. Whether new claims on behalf of metaphysical
> barriers will last any longer than previous attempts is an
> interesting issue.
>
>
> In Newtonian physics, there is the old "double pendulum," which if
> the arms are extended in a large arc, produces unpredictable
> "chaotic" movements. Usually these movements group around ratios of
> Feigenbaum numbers but they are still as unpredictable as next
> month's weather. And incalculable to boot.
>
>  Irreducibility can be observed in all kinds of complex systems (the
> behavior of market sectors in economies, for example). Emergentist
> topics are not limited to biology.
>
> This may be what Poincare (an early groundbreaker in chaos theory)
> had in mind when he wrote that the theorems of geometry -- whether
> Euclidian, Lobachevskian, or Riemannian -- were merely conventions.
>
> Curiously these self-organizing complex systems we are part of, and
> of which we consist, may offer philosophers a way out of the
> Derridian rejection of lexical foundations, or the Rorty-inspired
> rejection of epistemology itself.
>
> ------------------------------------------------------------------
> To change your Lit-Ideas settings (subscribe/unsub, vacation on/off,
> digest on/off), visit www.andreas.com/faq-lit-ideas.html
>



-- 
John McCreery
The Word Works, Ltd., Yokohama, JAPAN
Tel. +81-45-314-9324
jlm@xxxxxxxxxxxx
http://www.wordworks.jp/

Other related posts:

  1. Home
  2. lit-ideas
  3. Archive
  4. 12-2011