Transcript Review of the Cohesion Concept of Species

Review of the Cohesion
Concept of Species
John Collier
University of KwaZulu-Natal
www.ukzn.ac.za/undphil/collier
Phylogenetic species
Phylogenetics: The Theory and Practice of Phylogenetic
Systematics, E.O. Wiley, 1981
•
•
•
•
Species determined by phylogeny. Historical individuals.
not phenotypic similarity
traits are signs of lineage
diagnostic traits determined by considerations of evolutionary
history, derivation of traits, making some traits more reliable for
diagnosis than others
• evolutionary processes constrained by synchronic and diachronic
constraints Wiley called horizontal and vertical cohesion
• cohesion determines species identity
Information flow
Evolution as Entropy, Brooks and Wiley, 1986
• Phylogeny as a process of information flow.
• Historical constraints (vertical cohesion) and ecological constraints
(horizontal cohesion) plus self organization within information flow
govern speciation.
• Cohesion determines species identity
• Speciation is a disruption of species cohesion
• Species cohesion determines species macrostates, with variation
within species determining microstates
• Entropy of species (as determined by informational
macrostate/microstate relation) is typically not maximal
• This allows self-organization and disruption of cohesion by selforganizing bifurcation.
Information Flow 2
Brooks and Wiley, 1986 (continued)
• Focus of information is on genes (and chemical DNA), but this is not
required, and some exceptions are noted.
• Distinction between internal and external (within and without the flow
of genetic information) noted. Sexual selection a case of an internal
process, natural selection an external constraint. Developmental
constraints are also assumed to be internal, though development is
also treated as a case of information flow with vertical and horizontal
constraints. External constraints (e.g., selection) said to be purely
rate-determining
• Restrictions on gene flow and equilibrium with constraints are
assumed and treated statistically with the entropy concept.
• Kinetics (both internal and external) largely ignored (mistakenly:
Harrison)
Cohesion concept of Species
The meaning of species and speciation, Templeton, 1989
• Explicit use of cohesion to define species
• Fully gene centred: species determined by gene flow (“intrinsic
cohesion mechanisms”)
• Process oriented, but no explicit distinction between vertical and
horizontal cohesion.
• No theoretical unification, just a collection of mechanisms (next
slide) that confine and separate gene flow
• Cohesion both separates species (isolation) and unifies species
(facilitation of gene flow)
• Speciation as disruption of cohesion
Cohesion concept of Species 2
Cohesion mechanisms (Templeton 1989)
A.
B.
Genetic exchangeability
1. Promotion (facilitation)
2. Isolation
Demographic exchangeability (constraints from selection and drift)
1. Replaceability (genetic drift promotes genetic identity)
2. Displaceability
i. Selective fixation
ii. Adaptive transitions (natural selection)
a. constraints on the origin of heritable variation
b. constraints on the fate of heritable variation
ecological, developmental, historical, population
genetic (all “extrinsic” – to genes, presumably)
Some remarks on the phylogenetic
and cohesion approaches
• Wiley’s approach is more general because it does not
specify heredity mechanisms
• Templeton’s approach is more operational, but difficult to
generalize
• Brooks and Wiley regard sexual selection and
developmental constraints as internal, whereas
Templeton regards sexual selection as a form of natural
selection (intrinsic to gene flow), but developmental
constraints are extrinsic
• The last is a significant difference following from
Templeton’s focus on genes alone (and assuming
implicitly that selection is on genes, thus intrinsic to gene
flow), and Brooks and Wiley’s focus on information flow.
We need a general account of cohesion to
give generality to accommodate nongenetic heredity and the formal parity of
constraints with information and cohesion
in order to evaluate the views and extend
them.
This requires an analysis of the concept of
cohesion.
The cohesion concept
•
•
First of all, cohesion is an identity concept, so
we start there.
Identity, A = B
•
•
•
•
Logical condition, same for all things
Equivalence relation: symmetric, transitive, closed
A = B implies that B has every property that A has, and
vice versa
This tells us virtually nothing, since it is a
purely logical relation, but it does put these
logical constraints on any concept of
autonomy. The next move is to look at what
makes parts of something parts of that thing.
This is provided by the unity relation.
Unity
Unity, U(A)
•
Unity is the relation among the parts of a
thing A such that:
•
•
•
•
•
•
If a and b are parts of A, then aUb, and bUa (symmetric)
If a and b are parts of A, then aUb and bUc implies aUb
(transitive)
By a. and b., U is an equivalence relation
U(A) is the closure of U, given any initial part.
By a. to d., U(A) contains all and only the parts of A.
It is empirical question what satisfies U(A) for
a given A. Typically the type of unity relation
will depend on the sort of thing A is.
Cohesion, the dividing glue
Cohesion, C(A), dynamical unity
•
Cohesion is the unity relation for dynamical objects,
such that:
•
•
All parts aCb are dynamical
C is dynamical
Simple examples of cohesion:
a quartz crystal
a gas in a box
•
Note that in each case the cohesion is not absolute; it
is a matter of degree.
•
We should expect difficult intermediate cases.
•
Cohesion can differ in strength in different dimensions
(factors)
The Tautology Problem
Gold (2001)
• Hey decries this popularly acclaimed definition.
He unmasks it as a tautology framed in the
ostensibly palpable, yet suffering a lack of utility.
Hey urges us to substitute the word “cohesion”
with any other word and suggests we will find
equal meaning. (Ghiselin later made efforts to
repair Templeton’s mis-construction by
identifying a species as that which is by what it
does, specifically defining species as the
products of speciation.)
The tautology non-problem
• The cohesion concept has alternatives
that might be correct, e.g., phenetic
concepts or essential properties concepts
• The cohesion concept neither endorses or
prohibits particular contributions to
cohesion. This is an empirical matter.
• This empirical matter is testable.
Phylogenetic species are cohesion
species
• Though tautology is a non-problem, if we
assume a phylogenetic account of species as
historical entities, it is necessary that cohesion is
the determiner of species identity.
• If we use the cohesion account of species, then
the phylogenetic view of species follows
• So the accounts are equivalent if not identical
Pluralism?
• There are many contributors to cohesion, so we
might take it that we need a pluralistic account of
species to accommodate all of these.
• This seems plausible given Templeton’s menu of
cohesion factors.
• On the Wiley account, however, various factors
contribute to overall species cohesion, giving a
net effect, as in forces in Newtonian mechanics.
• We don’t think that Newtonian systems are
pluralistic, so by parity we should not consider
species concepts to be pluralistic; there are just
different aspects of species identity
Pluralism? 2
• However, if we have different loci of action then
the argument is not so simple.
• The genetic view of Templeton provides a single
locus of action, arguably.
• But if extra-genetic factors are in play, it is not
obvious that there is a single locus of action.
• I have not yet worked this out fully.
• It would interesting if there are speciation
events that do not involve genetic differences.
Thank you