[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Subject Index][Author Index]

Life Beyond the Cladogram



With grudging consent of the slave-masters, I've secured a whole hour of
freedom for Christmas. This allows me to respond to a comment recently made
by Paul Willis [Re: various & sundry. Tue, 22 Dec 1998 08:39:52] .. .. ..

[Quote]  "Essentially, if you do not have a logically falsifiable phylogeny
for a group of animals, then you do not have a logical framework for
investigating their palaeobiology."  [End quote]

This statement is doubtless appropriate for the special context in which
Paul made it (appraisal of crocodilian palaeoecology paper by P.J. Markwick,
Palaeobiology, 1998).  But I fear that many people - perhaps even Paul
himself (?) - would regard it as a valid stand-alone generalisation.  Such a
generalisation about palaeobiology is utterly wrong.  It encapsulates some
misunderstandings which cause a lot of grief, both in primary scientific
work and in the correspondence on this list.

I don't want to clutter my comments with definitions and quasi-legal
wrangling over terminology.  To keep it brief, I'd hope that no one would
object to defining 'Palaeobiology' so as to include the 'natural history' of
ancient organisms (their ancient ecology or, more crudely, the way in which
they made a living).  This field of interest contrasts with Phylogeny and
Systematics - the natural context of the organisms, and the formal
expression of that context (more crudely, how and where the organisms 'fit
into the pattern' of the natural world, regardless of how you choose to
uncover and describe that pattern, and how you choose to explain its
origin).  Some people might classify phylogeny and systematics as components
of 'Palaeobiology' sensu lato, but that's immaterial to my comments here.
(This proviso also covers those surprisingly numerous cases where an article
with 'palaeobiology' in the title transpires to be  largely phylogenetic in
content.)

To put it as simply as possible (and, I hope, in undistorted form), Paul's
aphorism says - you can't hope to understand the 'natural history' of
organisms without first understanding where they 'fit into the pattern' of
the natural world (your understanding of that pattern being, of course,
soundly logical, provisional and falsifiable).

How do we uncover and describe the pattern?  Easy - use the well-articulated
principles of comparative anatomy, founded on the concept of homology.
There are dozens of "How to do it" manuals on the market, along with
computer programs to handle a lot of the spade-work for you.

How do we get information about the 'natural history'?  Not so easy.  There
are three routes to enlightenment.  (1) Trace fossils, which can tell you
how the organisms moved around, what they ate, how they reproduced, and so
on - and in what environments they did so.   (2) Reductionist approaches -
interpret jaws, teeth, limbs and so on as bits of machinery, and predict how
they could (and couldn't) operate in mechanical terms.  Organisms, or bits
of them, are treated as raw materials for an exercise in physics and
mathematics.  (Similarly, they can be treated as raw materials for an
exercise in chemistry.)    (3) Use the principles of functional anatomy,
founded on the concept of analogy.  Here, alas, the principles are poorly
articulated (if understood at all), there are no "How to do it" manuals, and
no helpful computer programs.

The first pursuit (comparative anatomy) yields a phylogenetic-systematic
hypothesis; the second set of approaches (trace fossils, reductionism,
functional anatomy) furnishes a hypothesis of 'natural history'.  The two
hypotheses are obtained by different procedures, with different sets of
assumptions, starting (sometimes) from different raw materials.   Obviously
the two hypotheses are complementary, and they can be mutually enlightening.
But there is no logical necessity to bring the two hypotheses into
conjunction.  The two hypotheses can stand independently.  If you do choose
to bring them into conjunction (and that is an option, NOT an inescapable
necessity), then neither of the hypotheses takes priority.  They are two
independent and equally valid scientific hypotheses which may, if you
choose, be amalgamated into a bigger picture. 

Much of this seems to have been evident to many anatomists in the past, from
Aristotle to Cuvier, and beyond.  And many of them were inclined to accord
greater scientific importance to the 'functional' or 'natural history'
hypothesis rather than the phylogenetic-systematic hypothesis.  (Though, of
course, they used very different language.  The early part of E.S. Russell's
history of animal morphology (1916) is an illuminating source of information
about the concepts; see reference at end.)  Such liberal approach to biology
still persists in some European cultures and (to me, a bit surprisingly) in
some Asian cultures.

Typical 'western' science has gone the other way:  the
phylogenetic-systematic hypothesis is accorded central importance, whereas
the natural history hypothesis is regarded as an 'optional extra' that is
necessarily dependent on, and shaped by, the prerequisite phylogeny.  This
is exactly the view expressed by Paul Willis, and by many others.  Niles
Eldredge, for example, described an evolutionary tree as a phylogenetic
hypothesis (cladogram) with an 'overlay of adaptational narrative' (i.e. the
'natural history' hypothesis superimposed as an optional extra).  [I can't
recall the exact words he used, but these are pretty close.]   Here the
phylogenetic hypothesis is perceived as commendably objective, testable and
falsifiable science.  (Good stuff!)  By contrast, the 'adaptational
narrative' (natural history hypothesis) tends to be portrayed as subjective,
untestable, unfalsifiable... little better than guesswork and speculation.
(Miserable garbage!)

Those perceptions are deeply ingrained.  They pervade the literature, they
are put forth by teachers, learned by students, virtually accepted as axioms
- self-evident truths that no one is encouraged to question.  They also
happen to be wrong and, hence, the cause of serious misunderstandings.

The 'natural history' hypothesis can be constructed on sound scientific
principles; it can be objective, testable and falsifiable.   (Yes, there is
a note of caution: many such hypotheses aren't rigorously constructed and
aren't presented in a form that would render them testable and falsifiable.
But that doesn't mean that such arguments CAN'T be constructed and presented
as valid scientific hypotheses.)  The natural history hypothesis and the
phylogenetic-systematic hypothesis can be independent and of equal
scientific status. 

The real problems arise partly from our ignorance and partly from the depth
and extent of our brainwashing.   Ignorance?  Yes, as I mentioned earlier,
the principles of functional anatomy, grounded in analogy, are not well
understood and have never been fully explored and clearly articulated.  We
just don't have explicit and well-tested 'rules of the game' for
investigating the natural history of extinct organisms (though we do have
very elaborate and well-understood rules for investigating their phylogeny).
Brainwashing?  Yes, even if someone presents us with a soundly-formulated
hypothesis of 'natural history', we may fail to recognise it as such
(because it's so alien to our everyday conception of science) and may be
inclined to dismiss it as untestable speculation or guesswork.  And, so, the
misunderstanding is perpetuated.

Let me make it a bit more pertinent to the interests of this list (if anyone
has bothered to read this far).   We all skim through, sometimes stop to
read, and sometimes contribute to, this list because we share common
interest in T. rex, or in creatures more or less closely related to T. rex.
As someone with an interest in 'natural history', I don't really want to
know if T. rex is more closely related to Xosaurus than Yosaurus (or vice
versa), or if it belongs in clade A rather than clade B.  Instead, I want to
know what it ate for breakfast and whether or not it brushed its teeth
afterwards.  And I suspect that a lot of other people subscribe to this list
in the hope of gaining information on similar questions of natural history.
These are valid scientific questions, but a lot of scientists won't attempt
to answer them - because the questions are perceived to fall outside the
boundaries of 'real' palaeontological science (concerned with
phylogenetic-systematic hypotheses, or with the quantifiable data of
biostratigraphy, biomechanics, and so on) - or they will answer them with
safe generalisations (culminating, of course, with the safest of all -
'Don't know' and 'Your guess is as good as anyone's').  Some will even
dismiss such questions as frivolous, not worthy of serious scientific attention.

But what happens if someone DOES try to answer such questions with logically
structured arguments culminating in hypotheses that are testable and
potentially falsifiable?   Whenever I've tried to do so in the past, the
responses of reviewers have been mighty revealing.   They include comments
such as the following:

- 'curious' - [i.e. I'm not sure if this really qualifies as science...]

- 'lacking in phylogenetic perspective' - [i.e. this may not be entirely
adequate, because it's not phylogenetic or, at least, not tied into a
reassuringly familiar phylogenetic context...]

- 'intuitive' - [i.e. the principles underlying this hypothesis are so weird
and unfamiliar that I failed to recognise them as a scientific argument...
or else it's a fancy form of guesswork]

- 'untestable' - Ii.e. the argument is disturbingly unfamiliar in structure
and outside my conception of science, so it can't be scientific...]

The comments in square brackets are my attempts to second-guess the thought
processes of others, and I admit that they may be so far from the truth as
to be very unfair.  Also, I apologise if anyone recognises their own
handiwork in there!  But I suspect that some, at least, are reasonably
representative of reactions to a scientific hypothesis that concerns natural
history (grounded in analogy) rather than phylogeny and/or systematics
(grounded in homology).

Before finishing (do I hear cries of relief?) I'll mention one other type of
response.  If it's proposed (as testable scientific hypothesis) that
organism A might have performed function B, and this hypothesis survives
attempts at falsification, then a last-ditch objection might be -  'Alright,
organism A might have performed function B... but you can't prove that it
actually did so.'   My counter-response would be -  'Alright, your
phylogenetic diagram proposes that clade X might be the sister group of
clade Y... but you can't prove that it actually was.'   And, in reality, I
suspect that my 'natural history' hypothesis is closer to the truth than is
your phylogenetic hypothesis.  There can be only one true (real)
phylogeny... but organs and anatomical structures are often multifunctional.

The bottom line.  Unless we make a conscious effort to break free from the
stranglehold of phylogenetic-systematic hypotheses, we won't make any major
progress in developing palaeobiology (in the sense of 'natural history') as
an autonomous science, with its own legitimate scientific 'rules of the
game'.  Loosen up the preconceptions.  There is, and was, life beyond the
cladogram.   And some of those 'untestable speculations' might transpire to
be a good, if unfamiliar, brand of science.

Back to the swamp....

Tony Thulborn

REFERENCE:  Russell, E.S.  1916.  Form and Function : a contribution to the
history of animal morphology.  Murray, London, 390pp.  [reprinted 1982, with
new introduction by G.V. Lauder, by University of Chicago Press, Chicago,
383pp.]