Darwin-L Message Log 1:173 (September 1993)

Academic Discussion on the History and Theory of the Historical Sciences

This is one message from the Archives of Darwin-L (1993–1997), a professional discussion group on the history and theory of the historical sciences.

Note: Additional publications on evolution and the historical sciences by the Darwin-L list owner are available on SSRN.

<1:173>From DARWIN@iris.uncg.edu  Sat Sep 18 16:50:25 1993

Date: Sat, 18 Sep 1993 17:21:06 -0400 (EDT)
From: DARWIN@iris.uncg.edu
Subject: Phylogeny (history) is important, classification is not
To: darwin-l@ukanaix.cc.ukans.edu
Organization: University of NC at Greensboro

We have had much discussion of classification.  Along with many other people in
evolutionary biology today, I don't find the topic of classification to be a
fruitful one, and I would like to make the strong claim that classification is
almost completely irrelevant to the contemporary practice of evolutionary
biology.  This is because much discussion of "classification" in fact fails to
make a very important distinction now made within evolutionary biology between
two different activities: (1) classifying (making groups), and (2)
reconstructing phylogeny (evolutionary history).  Until this distinction is
clearly made there really can't be any fruitful discussion of either
classification or phylogeny.  This whole subject has been _enormously_
clarified in the last 15 years or so, and all practicioners in the field now
make this distinction clearly.  Here's a sketch of my reading of the situation;
I'm sure my views are not universally held in all their details, but neither
are they unique.  Every paragraph below could easily be given monographic
treatment, and many have been; my aim here is just to give a broad sketch for
those who have an interest in the subject but have not followed the literature
closely.  We have among our members some leading historians of systematics and
some leading systematists, so I apologize if I do violence to any of their
views in the interest of brevity.  While my comments address evolutionary
biology, I like to think they have implications for historical linguistics and
stemmatics as well.

The last 25 years has seen an enormous growth in the study of phylogeny, the
evolutionary history of life.  This was one of the great fields of study in the
late 1800s, but for much of the early and mid-20th century phylogeny was
comparatively negleced in favor of studies of evolutionary mechanisms and
"species-level" problems.  (This species-level work was enormously important of
course, and it is still with us.)  Since the mid 1960s, however, interest in
phylogeny has grown enormously due to three factors: (1) the development of
computational methods for dealing with large quantities of data; (2) the
availability of data on comparative molecular anatomy, though this has not been
as important to the conceptual development of the phylogenetics revolution as
the popular press (and some molecular biologists) would have people believe;
and most importantly (3) a new conceptual understanding of the relationship
between the observed similarities and differences among organisms and the
histories that can be inferred from those similarities and differences.  This
last factor is behind the method of "cladistic analysis" which has effected a
revolution in the study of phylogeny, and which is now the standard method for
reconstructing evolutionary history.  The conceptual development of cladistic
analysis in the last 30 years or so has been as important to systematics as the
development of the law of superposition -- that upper strata are younger than
lower ones -- was long ago to geology.

The briefest sketch of the long view of this subject would go like this (many
of these statements could be qualified; there is an increasing body of
published literature on this).  Beginning in the 1600s and 1700s there was an
enormous increase in factual knowledge of natural diversity on the part of
European scientists.  The notion of a linear "chain of being" which had been
the principal organizing system for diversity up until that time came, as a
consequence, to be challenged in the late 1700s and early 1800s, and a variety
of other arrangements were conceived for the diversity of life, including maps,
stars, circles, nets, and trees.

Notice that I spoke here of _arrangements_.  It is entirely true that may
students of diversity engaged in classification and published classifications.
But many theorists even during this early period argued that "classification"
was not the right intellectual model for understanding and representing the
structure of living diversity.  In simplest terms, classifications are based on
group-within-group relations; _arrangements_ may be based on group-within-group
relations but also on some sort of positional relations in an abstract space:
taxa are not only contained within other taxa, they are also near and far,
between, above or below other taxa.  Think of the difference between a printed
listing of states, counties, and cities (group-within-group), and contrast that
with a map of those same places showing their positions in geographical space.
The chain of being is an example of a linear _arrangement_; it is not strictly
speaking a classification because it contains a linear axis along which taxa
must be placed.  The term "system" was often used as a synonym of
"arrangement", and people came to speak of "the natural system" -- that is, the
true arrangement of the diversity of life.  From this notion of system we
derive the term "systematics" which is used for the field today.  Many
important workers in the early 1800s directly contrasted classifications (which
they regard as inferior) with systems/arrangements, among them Macleay and
Alfred Russel Wallace.  One sometimes hears people say "a natural system of
classification", but to the workers in the past or the present who make the
distinction between classifications and systems/arrangements, a natural system
of classification is an oxymoron: the reason such people speak of natural
systems is because nature isn't arranged in classes.

A very special conflict arose once people accepted the notion of evolution and
it became clear that the true arrangement of living diversity (the natural
system) is a tree.  This conflict arose because there are two kinds of tree
diagrams: (1) "logical trees" representing classificatory relations, and (2)
historical, genealogical trees: "trees of history".  I can make a "logical
tree" showing the classification of furniture into chairs and tables, and then
into desk chairs, dining chairs, lounge chairs, etc.  Such a "tree" however is
purely a classificatory device that has nothing whatever to do with
evolutionary, genealogical, "trees of history", in which the root is an actual
organism or population that lived in the past.  One of the most troublesome
problems in the history of systematics has been the confusion of logical trees
and trees of history, that is, classifications and phylogenies.  It is an
empirical fact that people within and without the field of systematics have
found "group thinking" to be easy and intuitive, but "tree thinking"
(historical/genealogical tree thinking) to be extremely difficult.  It is
possible to see the seeds of this confusion in the classification chapter
(XIII) in the first edition of the _Origin of Species_ where Darwin
distinguishes precisely between classification and arrangement in several
places (and here he owes much to an earlier paper by Wallace, I believe) but
doesn't really develop all the consequences that result from this distinction,
especially in view of the fact that many traditional groups were clearly not
whole branches of the evolutionary tree (the logical and genealogical trees did
not match).  Because the full implications of the classification/phylogeny
distinction were not genuinely internalized in systematics until quite
recently, some people have spoken of the Darwinian revolution in systematics --
the idea that systematics is really about reconstructing evolutionary history,
and that the natural system is in fact the sequence of events (the "chronicle")
that make up the evolutionary past -- as being effected only within the last 20
years or so.  This is a view with which I agree.

Now, is classification per se a valuable thing to study from the point of view
of the history of systematics?  Absolutely it is, in precisely the same way
that phlogiston theory is important to study in the history of chemistry, and
astrology is important to the history of astronomy.  And it is certain that
there are elements of "group thinking" still present in systematics today that
need to be eliminated; study of classification will help us to understand why
these persist and how people regard them.  But theoretical considerations of
"classification" as a distinct intellectual practice are not relevant to the
contemporary _practice_ of systematics, which has as its task the
reconstruction of evolutionary history.  There is nothing that sounds more
utterly barren today than the debates of the 1970s over "my way of classifying
is better than your way of classifying."

Incidentally, I understood along with Margaret Winters that the famous "Chinese
classification" that has been posted here a couple times already, and which
comes from Foucault, was actually taken by Foucault from Borges.  I also
understood that Borges made the whole thing up (which makes attempts to
"analyze" it look pretty silly).  Whether or not Borges made it up, I have
always regarded that example as prima facie evidence that the whole subject of
classification is irrelevant to evolutionary biology, because that example has
nothing to do with reconstructing history.  Anyone who wants to look at a
literary example that _is_ relevant to historical reconstruction and
representation might try another Borges story, "The Garden of Forking Paths",
which is about branching histories and contingency.  It is anthologized in his
book _Labyrinths_, I believe, and should be available in most libraries.

Bob O'Hara, Darwin-L list owner

Robert J. O'Hara (darwin@iris.uncg.edu)
Center for Critical Inquiry and Department of Biology
100 Foust Building, University of North Carolina at Greensboro
Greensboro, North Carolina 27412 U.S.A.

Your Amazon purchases help support this website. Thank you!

© RJO 1995–2016