• Adam Smith Institute

    Adam Smith Institute place holder
  • Philosophy & Logic

    Philosophy and Logic
  • Cambridge

    Cambridge
  • Children’s SF

    Children's Science Fiction
  • Enter your email address to follow this blog and receive notifications of new posts by email.

    Join 418 other followers

Mathematical models do not seem to enlighten us any more about the past than they do about the future

mounted archerI have expressed doubt before about the use of data about the past to predict the future.  In my book the future is inherently unpredictable.  We might project forward the big numbers such as life expectancy, age of marriage, and population size.  We can add likely economic trends, but even if we get all of these right it does not actually tell us what the future will be like, and even less, what it will feel like to live in.

I have flagged up my skepticism concerning ‘cliodynamics,’ with its use of mathematical modelling to predict the grand movements of the past, present and future.  I compared it to the fictional ‘psychohistory’ used by Hari Seldon in Isaac Asimov’s “Foundation” series.  Now my skepticism has been reinforced. A new paper suggests that war, rather than agriculture, led to the formation of complex societies, and uses mathematical models and statistics in an attempt to validate this claim.  Some of the points made by Peter Turchin in support seem very naïve.

“You cannot have a large state without bureaucrats, but bureaucrats are expensive. You have to pay them,” he says. “So the big question is how do complex societies evolve when they are so expensive?”

The notion that war enabled societies to become complex and employ bureaucrats seems convoluted.  We all know that human societies have engaged in war, just as tribal ones do today.  War was assumed to be the norm between Greek city states.  Agriculture is also a factor, in that people no longer depended on what they could catch or could find.  Land became important, and peoples occasionally fought over it.  Agriculture generated surpluses and people fought over those.  All of this is mainstream,

“To test the two competing theories, Turchin and company designed two mathematical models for predicting the spread of complex societies. One based only on agriculture, ecology and geography. The other included those three factors, plus warfare. Then, they used data from historical atlases to determine whether these models matched up with the way the different states and empires actually evolved.”

Lo and behold the model that included warfare explained 65% of the variance, whereas the agriculture model only managed 16%.  But as I said, no-one has supposed that agricultural societies did not engage in war.  I note that their models did not include population data, and I have yet to see precisely how they defined the ‘complexity’ of different societies.

It is when Turchin talks about more recent times that my disbelief rises even more.  Talking about competition between societies, he tells us that workers were paid more in the 1920s because US companies voluntarily gave them more money out of a fear that Soviet communism might spread.  Oh dear.  I wonder if all the bosses met in a room and all agreed to do this by a binding majority vote?  I think I prefer the more conventional account that companies competed for workers, and such things as mechanization and electrification made workers more productive and therefore worth more to their employers.  I rather think that if the mathematical models are constructed using facile assumptions, then their findings are not going to be all that worthwhile.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: