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

Re: "running" elephants - locomotory analogs




Date: Wed, 9 Apr 2003 08:37:47 -0400
From: "Philidor" <philidor11@snet.net>

Seems obvious, but you'd be surprised at how many people think having a
computer program which processes massive amounts of information adds
credibility.

I wouldn't claim that they do add "credibility", whatever that really is. I think we should try to understand functional mechanisms at as deep a level as possible, with first-principles explanations as a goal. Peel back the layers of the onion, so to speak. With a cautious approach that uses sensitivity analysis to check how important unknown data are for the results in question, once can head in this direction even with extinct animals. Qualitative or intuitive approaches are seldom able to either delve toward first-principles or analyze the importance of their unknowns. They tend to stay on the outside of the onion. Computers happen to be tools that are useful for doing both things, or one can do the calculations by hand. As I've been emphasizing, apparently to no avail, the presence or absence of a computer simulation is irrelevant; they are just time-savers. The tyrannosaur paper we wrote could have been calculated solely with paper and pencil, and in fact during my thesis work I did that anyway just so I'd know how it was done; it wasn't that complex.


The holy grail sought by both biomechanical and qualitative/analogy-based/intuitive/beer-drinking armchair approaches is an understanding of how mechanisms work in animals, particularly how form and function are related, and how to apply that understanding comparatively or to a single animal. I prefer biomechanical approaches to the latter approaches for understanding functional mechanisms for the reasons already mentioned, among others. To me, biomechanics is simply a deeper, more satisfying and fertile approach. I have learned how approaches that infer function directly from form alone, without much understanding or even consideration of underlying mechanisms, are often wrong because they overlook complex interactions or emergent properties. Outside of the onion problem. Because organisms are very complex (many layers), all approaches suffer from this problem to some degree or another, of course, but not necessarily all to the same degree.

George Lauder wrote a wonderful (and healthily cynical) paper on this problem in the 1995 (JJ Thomason, ed) Functional Morphology in Vertebrate Paleontology volume; I recommend it to all. (Also see a great paper by Mimi Koehl called "When does morphology matter?" in Ann.Rev.Ecol.Syst 1996, iirc) I know some people that read Lauder's paper, then swore never to try doing functional analysis in paleontology, as they were left feeling hopeless. I have more hope than that, and biomechanics (with sensitivity analysis) is a big reason why I do. I use many different methods in addition to biomechanics in order to seek for the solution that available lines of evidence point toward, with the expectation that eventually the best answer will be found via consilience of independent inquiries. More of those inquiries shall come in some of my upcoming papers.


--John R Hutchinson