Maximisation Principles in Evolutionary Biology 343used to develop population genetics is that it has the same mathematical form
as the Reimannian distance metric mentioned in the last section, but Edwards
[2002c] has argued that this is only of superficial interest.
7 RECONSTRUCTING EVOLUTION: THE PRINCIPLE OF PARSIMONY
AND OCKHAM’S RAZORWe now turn away from principles that emulate the techniques of mathematical
physics to one that involves a minimisation principle which ultimately takes its
justification from the theory of statistics and in particular the method of maxi-
mum likelihood. The problem it addresses is that of reconstructing the course of
evolution from a knowledge of present-day characters — be they morphological
traits amongst species, blood-group frequencies amongst populations or DNA se-
quences amongst individuals. The heuristic foundation is that similar things are
closely related. But why?
InThe Origin of SpeciesDarwin [1859] replaced the hypothesis of the individual
creation of each species with the hypothesis of a single origin of life followed by
the evolution of new species. He did not appeal to Ockham’s razor in support
of the change, and would probably have been surprised to have been told that
his achievement was an example — perhaps the example par excellence — of the
application of Ockham’s razor. His argument, rather, was one of analogy with the
way lineages in genealogies could be traced back to common ancestors.
Ockham’s razor, sometimes now calledthe principle of parsimony, is tradition-
ally associated with William of Ockham (ca. 1280-1349):
Pluralitas non est ponenda sine necessitate(multiplicity ought not to
be posited unnecessarily).In recent years this principle has sometimes been invoked by systematists to
justify particular methods of phylogenetic systematics (for reviews see [Sober,
1988], and [Stewart, 1993]). In this context the principle manifests itself as what
we might call ‘the hypothesis of the single origin’, whether of life itself or of a
particular character. Darwin drew his first evolutionary tree with a single root
in 1837, and the principle that the hypothesis of a single origin is to be preferred
where possible can be calledthe Darwin principle. Thus if a particular character
appears the same, or very similar, in two species, we are to prefer the hypothesis of
a common origin for this character over the hypothesis of two independent origins
and parallel selection, unless, of course, other evidence overwhelms it. ‘Perhaps
the correct way of viewing the whole subject’, wrote Darwin inThe Origin, ‘would
be, to look at the inheritance of every character whatever as the rule, and non-
inheritance as the anomaly.’ Again, in 1862 he wrote ‘Is it not a more simple and
intelligible view that all the Orchideæ owe what they have in common, to descent
from some monocotylodenous plant...?’ [Darwin,1862].
Ockham’s razor invites us to prefer the hypothesis of the single origin, but the
rise of probabilistic thinking in the eighteenth century rendered this explanation