ASBS Newsletter – Book Review
Cladistics. A Practical Course in Systematics
written by Peter L. Forey, Christopher J. Humphries, Ian J. Kitching, Robert W. Scotland, Darrell J. Siebert, and David M. Williams
(From ASBS Newsletter Number 79, June 1994)
Publisher: The Systematics Association Publication No. 10. Clarendon Press / Oxford University Press, Oxford. 1992. xi+191 pp. ISBN 0-19-857767-2.
As some of you will have noticed by now, I have been looking for a good introductory book about cladistics. I have, in fact, been looking for some years now, and it's been a pretty depressing process. While I don't claim to have any moral relationship to either King Arthur or Don Quixote, the zetetic legends surrounding these two figures are beginning to seem very realistic to me. Unfortunately, after reading this book, I have realized that my quest has not yet come to an end.
Most of the systematics books that I've encountered so far in my search have either been written by someone who claims little expertise in cladistics, and who therefore misrepresents it out of ignorance, or they have been written by experts who get so tied up in the minutiae that they can't see the wood for the trees. The first of these books make good compost, while the second type are only of practical use to the cognoscenti.
This is not to say that there aren't some good general introductions out there. For example, the chapter by Michael G. Simpson in Fundamentals of Plant Systematics (by Albert E. Radford) is getting dated (1986) but is still readable, as is the section in the more recent (1990) Plant Taxonomy: The Systematic Evaluation of Comparative Data by Tod. F. Stuessy. There is also a good introduction in The Insects of Australia, Second edition, Volume 1. The chapter by David Swofford and Gary Olsen in Molecular Systematics (edited by David Hillis and Craig Moritz) is the best discussion of the actual analysis techniques, and Wayne and David Maddison's MacClade: Analysis of Phylogeny and Character Evolution provides an excellent introduction to an explicitly phylogenetic perspective on biology.
However, the problem with these works is that they are either general introductions that don't go into enough detail to be useful to someone who wishes to explore the subject seriously, or they only cover one part of the topic. What the systematic world still needs is an introductory text that strikes a balance between the generalities and the details; and this is, of course, what Cladistics intends to do. It falls short of this desideratum, however.
The book had its genesis as the course material for an intensive one-week workshop (called "Cladistics: Theory and Practice") sponsored by the Systematics Association in 1991. The course manual was revised for the repeat of the workshop in 1992, and was then published in its current form. The intent of the workshop was to cover modern cladistics as taught by expert practising cladists, and in this I'm sure that it succeeded. However, the idea of publishing the workshop manual directly as a book is little short of ludicrous, for three reasons.
Firstly, the book reads like a set of lecture notes - the style is very abrupt, with very little development of any one concept, and little flow between concepts. This does nothing to enhance the reader's understanding of the ideas being discussed, nor of how these ideas relate to one another. As an exercise in scientific communication this is not up to par, and I'm sure that the authors would like to have extensively re-written their material if they had been given the chance.
Secondly, the way in which ideas are discussed in person and the way in which they are communicated in books is very different. This is why personal contact and interaction are so important in the workshop environment, both between the instructor and the participants and among the active participants. This is why intensive workshops can be so successful. In this context, the lecture notes only fulfil the role of reminders of the topics covered, not as the sole means of communication, as they must be in a book. So, the course manual is part of a larger package; and the reader of this book is being short-changed, as they are only getting part of this package. For a book to be comprehensible, the material needs to be completely re-written from what is appropriate for a course manual.
Thirdly, there are so many mistakes in this book that it is positively misleading for novices. Consider this as a partial list:- Figure 1.4 only shows four ways that two characters can relate to one another, whereas the text claims that there are five such ways; Figure 3.5b has a line joining the wrong character states; Figure 4.3a shows taxon D having character state 2 instead of 0; Figure 4.12d has taxon C having character states 11 instead of 01; the legend for Figure 6.2a refers to the left-hand diagram rather than the right-hand one; the example calculations in Table 7.1 do not follow the formula given; Figure 7.12 seems to have a spurious character change included; Figure 8.6b disagrees with 8.6a about whether Cycads have pinnate leaves or not; section 9.2.2 is referred to repeatedly in Chapter 9 when it apparently should be section 188.8.131.52; on page 151 the reference to section 184.108.40.206 should be to 220.127.116.11; page 158 refers to an asterisk in Figure 9.18, which isn't there; Chapter 10 has several references to Nelson (1973) when there are three such references in the References; Figure 10.3 has variant spellings of the name of taxon 18.104.22.168.4.6. Some of these problems are trivial, but some actually change the interpretation of the information being communicated, which is unacceptable. There are also many typographical errors in the text, figure legends and table titles, including unnumbered pages, mis-spellings, wrong tenses, transposed words, etc.
So, even from the start the authors have an uphill battle to make this book worthwhile. Nevertheless, it's still advisable to have a look at the content of the book, to see what topics they consider worth covering as an introduction to cladistics. The workshop consisted of about 10 hours of lectures, and the book consists of 10 chapters, presumably following a 1:1 relationship. (The workshop also had about 30 hours of practicals, devoted to using the Hennig86, PAUP, Phylip, and Component computer packages - however, it's much easier to teach cladistics using MacClade and PAUP on a Macintosh).
Chapter 1 (11 pages; by Robert Scotland) covers Cladistic Theory, and does so quite well. However, there is no balancing alternative viewpoint, as phenetics and phyletics only get a paragraph each. More to the point, these alternatives are only criticized, while cladistics is only lauded. A slightly more objective perspective would not be hard to achieve, but this seems to be beyond the capabilities of many practising cladists.
Chapter 2 (8 pages; by Robert Scotland) discusses Character Coding, while Chapter 3 (22 pages; by Ian Kitching) describes The Determination of Character Polarity. These seem to be quite well balanced and comprehensive coverages, and they get the book off to a good start about the techniques.
Chapter 4 (28 pages; by Ian Kitching) supposedly covers Tree-building Techniques, but it rather surprisingly never makes clear how a tree is actually constructed. This chapter is actually about tree evaluation, rather than tree construction. Tree evaluation is considered by default to be based on a parsimony criterion, and what we actually learn about here is how character reconstruction is carried out on a pre-existing tree. While this is useful, even essential, it relegates actual tree-building to a black box. Perhaps a change of title is needed.
Chapter 5 (17 pages; by Darrell Siebert) has the unwieldy title of Tree Statistics; Trees and 'Confidence'; Consensus Trees; Alternatives to Parsimony; Character Weighting; Character Conflict and its Resolution. As you will gather from this amazing list, this covers everything that didn't go into the previous chapter; this makes it a bit eclectic, to say the least. However, there is definitely a single thread that runs through it - parsimony is good, and everything else is to be criticized. Consequently, we learn very little about the ideas of those people who believe that phylogeny reconstruction should be based on an explicit evolutionary model; we are clearly under the influence of the pattern cladists here. Furthermore, we never learn that ordinations are supposed to summarize non-hierarchical patterns, unlike dendrograms - they are solely criticized for their inability to reflect trees.
Chapter 6 (13 pages) and Chapter 7 (22 pages; both by David Williams) form a pair covering DNA Analysis, Theory and Methods respectively. These are both quite good, although Chapter 6 is a bit erratic, sometimes repeating things covered in earlier chapters and sometimes assuming a fair bit of unexplained background. Chapter 7 says very little about either distance methods such as neighbour-joining (other distance methods are covered in detail) or about maximum likelihood methods, which is unfortunate, as these are very commonly used in molecular studies.
Chapter 8 (13 pages; by Peter Forey) is an eclectic, but good, discussion of Fossils and Cladistic Analysis. It almost convinced me that they have some use after all.
Chapter 9 (23 pages; by Christopher Humphries) covers Cladistic Biogeography, and it's certainly the hardest chapter to come to grips with. It's a very convoluted introduction to the topic, covering more detailed historical background than do the other chapters in the book. In fact, it reads very much like a cut-down version of something that was much longer. Consequently, the ideas are very hard to follow (I spent a long time on the example in section 9.3.3, and I thought that I already understood it); and I have a horrible feeling that people are more likely to be put off by this chapter, rather than encouraged to try some biogeographic analyses for themselves. This is a pity.
Chapter 10 (10 pages; by Peter Forey) brings us to Formal Classification, and to the most controversial aspect of cladistics - cladistic classification. The desire to have only monophyletic taxa creates all sorts of practical difficulties. Unfortunately, the existence of paraphyletic groups in our current classifications is not explained at all well in this chapter, and the concept of metaphyletic taxa is not discussed all. Furthermore, the supposed chapter summary actually incorporates four paragraphs of ideas about ranks that appear nowhere else in the chapter. This hampers what is otherwise a good end to the book.
So, what do we have here? A pretty brave attempt, I guess, but no cigar (for those of you who remember Groucho Marx's quiz show). We definitely need a book like this one, but this one isn't really worth it. It may very well be the best book of its kind, but this is only because of lack of competition rather than because of the intrinsic quality of the book itself. Until something better comes along, we may well have to recommend this book to our friends as an introduction to cladistics. However, someone, somewhere will write a good introduction to cladistics someday, I'm sure of it.
Reviewer: David Morrison