Re: Genes and Memes

From: Matt Mahoney (
Date: Fri Apr 25 2008 - 09:49:42 MDT

--- Nick Tarleton <> wrote:

> On Thu, Apr 24, 2008 at 2:21 PM, Matt Mahoney <>
> wrote:
> > Memetic fidelity can be estimated from the evolution of language.
> After 100
> > generations, a language bears little resemblance to its original
> form. If we
> > assume that cultural beliefs are inherited in the same way as
> language, then
> > the error rate is about 10^-2. Therefore our cultural beliefs, to
> the extent
> > that they are heritable, must have a complexity of less than 100
> bits.
> At least, any that last more than 100 generations; but even this is
> patently false. The works of Homer, say, are >>100 bits and have been
> transmitted faithfully. Many religions are that old or nearly so, and
> although they have changed, >>100 bits have been preserved in each.
> Perhaps I misunderstand what you mean by "cultural beliefs".

I see the problem. By my argument, language should have a complexity
of 100 bits, rather than 10^9 bits. I need to be more precise in what
I am measuring. I should say that the complexity of the features or
constraints of culture that are stable from one generation to the next
is 100 bits. For example, a stable features of religion is the worship
of one or more gods, but evolution does not favor one god over another.

-- Matt Mahoney,

This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:01:02 MDT