[Humanist] 28.74 powers and limits of modelling

Humanist Discussion Group willard.mccarty at mccarty.org.uk
Sun Jun 1 20:43:07 CEST 2014

                  Humanist Discussion Group, Vol. 28, No. 74.
            Department of Digital Humanities, King's College London
                Submit to: humanist at lists.digitalhumanities.org

        Date: Sun, 01 Jun 2014 10:31:40 +1000
        From: Willard McCarty <willard.mccarty at mccarty.org.uk>
        Subject: powers and limits of modelling

One of the clearest statements of the (potentially dangerous) power and 
utility of modelling that I have come across occurs in the Introduction 
to Jay W. Forrester's 1971 book World Systems. Forrester wrote it for 
the second meeting of the Club of Rome, which convened at MIT to 
consider Forrester's concept of "system dynamics". (The Club of Rome was 
founded in 1968 at the Accademia dei Lincei to consider the serious 
problems facing the world. Its 1972 report Limits to Growth, by Donnella 
and Dennis Meadows and Jorgen Randers, was based on a computer 
simulation of the world they built at Forrester's lab; for a 
presentation of it see http://www.clubofrome.org/?p=326.)

But what concerns me here is Forrester's explanation in sections 1.5 and 
1.6 (pp. 14-16) of World Dynamics. "There is nothing new in the use of 
models to represent social systems", he writes. "Everyone uses models 
all the time." But the models we use, "mental models", are "fuzzy... 
incomplete... imprecisely stated... [changing] with time and with the 
flow of conversation". Basic assumptions and goals are hidden, 
considerations partial. In sum "this process", i.e. human reasoning, "is 
often faulty". The human mind is good at observing and detecting 
patterns. "But human experience trains the mind only poorly for 
estimating dynamic consequences of how parts of a system will interact 
with one another."

"Until recently", he continues, "there has been no way to estimate the 
behavior of social systems except by contemplation, discussions, 
argument, and guesswork." Now, of course, we have the computer, and so 
the ability to simulate the world. Forrester argues for a combination of 
"the strength of the human mind with the strength of today's computers". 
Their strength depends this or that model given them. "Such a model is a 
simplification of an actual social system, but it can be far more 
comprehensive than the mental models we otherwise would use as the basis 
far debating social policy." Although no model (i.e. in 1971) "can be 
considered more than preliminary, many are now beginning to show the 
behavior characteristics of actual systems."

Seems all quite unremarkably reasonable? Perhaps a sliver of doubt 
creeps in if one considers Brian Cantwell Smith's analysis in his 1985 
paper, "The Limits of Correctness", at a conference on unintended nuclear 
warfare, that in principle no perfect system is possible. The clear, complete, 
precisely designed, utterly stable system turns out to be rather a problem -- 
perhaps not in the time-scale Forrester and company were considering, 
but when measured by the flight-time of an ICBM....

What bothers me, however, is what may seem at first more benign: the 
mirroring loop by which the conception of the world as a "system", which 
came out of systems thinking coeval with the development of computing, 
should find its marvellously fitting mate in computing, and that in the 
process of this mating, certain human characteristics are dispensed with 
because they are judged to be "fuzzy... incomplete... imprecisely 
stated... [changing] with time and with the flow of conversation" and so 
on. Computing progresses, so the number of these annoyingly imprecise 
human characteristics that the machinery can do without is bound to 
increase, yes?

Surely there was, and is, much to celebrate, and more as computing 
progresses, as it will. But where, Alan Liu has asked, is the cultural 
(where, I ask, any other kind of disciplinarily self-aware) criticism in 
all this celebration? However morally justified (as with the Club of 
Rome's concerns), the promotional and the critical make for a very 
unhappy couple.



Willard McCarty (www.mccarty.org.uk/), Professor, Department of Digital
Humanities, King's College London, and Research Group in Digital
Humanities, University of Western Sydney

More information about the Humanist mailing list