ON THE PRESENT REVOLUTION
Though this be
madness, yet there is method in't.
It is certainly
not the least charm of a theory that it is refutable.
False facts are
highly injurious to the progress of science,
sways humanity by...dinning it into our ears that this
Scientific explanations are accepted consensually as they explain observed facts in an elegant, explicit manner with the proviso that they continue to be of value predictively as new data arrives. Cladistics has apparently succeeded in explaining complexly associated facts elegantly and explicitly, and apparently predicts similar results with different data. This essay examines cladistics in the light of scientific modeling, and presents a general perspective for the cladists' demonization of New Systematics and taxonomy in general.
THE SCIENTIFIC REVOLUTION
Positivists have put forward the
concept of a purely factual science, without presuppositions. The
"transformed" school of cladism modified this viewpoint to promote
a pure methodological science (Scott-Ram, 1990). Regarding scientific
revolutionaries, "Once science is viewed as a matter of
straightforwardly building theory from fact..., it may seem to make sense to
imagine a presuppositionless science and pejoratively to label those tenets
that are no longer deemed factual as 'metaphysical presuppositions'"
(Hatfield, 1990), i.e. unnecessary, biased, restrictive first principles. On
the other hand, other schools of cladism have a distinct metaphysics,
evolution being "a central tenet from which the principles and methods
of taxonomy are to be deduced" (de Queiroz, 1992). Assuming that
assumptions, paradigms and theory are unnecessary for the advance of
knowledge is the called Baconian fallacy (Fisher 1970: 4). Postmodern distain
for science as a problem-solver is incorporated into our everyday lives. For
instance, the rhetoric is full-blown in the mouth of a fuzzy chaoticist
character in best-seller
Truth, as given by the supposedly failed modernists, is said to be historically deliberately constructed as the great "meta-narratives" of western culture, these including Science, Christianity, Democracy, Communism, and Progress. Such constructions require exposure as self-perpetuating, aggrandizing shams, systems that commit acts of cultural tyranny by promoting the fiction that all knowledge may be reduced to an absolute, unified, rational system. In the field of textual criticism (e.g. hermeneutics, the rule-based science of interpretation) deconstruction of works, paragraphs, phrases and names reveals imbedded connotations that signify hidden presumptions of ideologies and cultural biases. "...With the collapse of the modern 'metanarrative' of reality, science is sustained by the 'performability' of theories, their ability to generate more scientific work" (Lyotard, 1986: 46). According to Golinski (1990), an understanding of scientific discourse in the era of the Scientific Revolution "requires attention not merely to the language of that discourse but also to the political functions of language in expressing and mediating social relations."
The postmodernist is committed to modes of thought that emphasize fragmentations, discontinuities and incommensurable aspects of any given intellectual system. Certainly most of us would agree with Barzun (1958: 360) that equality is "a concept for dealing out justice among incommensurable human beings." The postmodern result of this emphasis would be a "Rediscovery of Man" (C. Smith, 1975). Kuhn (1970: 198), a postmodern philosopher of science, considers revolutionary paradigm shifts in science a human reaction to communications breakdowns because of relative incommensurability of theories and experimental and observational positions. Words mean something different to different generations of scientists. Without "translation," a supplanting naturally follows. Kuhn recommends "going native" and learning that new foreign language but warns that the learner will at some point slip completely into that language. There is no middle ground, according to Kuhn, no real building of theory upon theory.
The postmodernist ideology requires "self-referencing" in that the ideas of the artist or writer should always be self-conscious and should not assume, as modernists are apparently wont to do, cultural homogeneity on the part of the viewers or readers. Definitions and discussions of postmodernism are admittedly various and may be found, e.g., in Jary and Jary (1991), Wynne-Davies (1990), and O'Sullivan & al. (1994). Major writers in a vast literature include Friedrich Nietzsche, Jacques Derrida, Michel Foucault, Jean-François Lyotard and Gilles Deleuze (see Piercy, 1995).
There is much to value in the writings of the some postmodernist thinkers, and deconstructionist methods are, of course, attractive to the literary critic, but postmodernist observations are often merely bandied about "playfully" and are not intended to be placed in the marketplace of ideas. It would be valuable, for instance, to engage in discourse leading to some consensus on solutions to the problem of protecting electronic publication of scientific papers in an chaotic cyberspace where attribution is quickly lost, but postmodernists simply reject this concern as the "cult of originality": the idea that everything we write, say or think is a pastiche or "cut-ups" of previous persons' efforts; intellectual autonomy is a mere fantasy (cf. Eliot & Cantsin, 1996), while we all live in an artificial reality of layers of meaning, connotations, signs and symbols where computerized virtual reality is actually a relief from illusion, being consciously self-referencing; "our notion of the natural and the real is really a highly political construction, a product of history..." (Woolley, 1992). Even in science, "each viewpoint creates its own reality" (Scheffler, 1982: 19 in a critique of Kuhn). The darling of postmodernism, F. Nietzsche, "rediscovered and restated the romanticist relativity of all knowledge, the futility of applying scientific cause and effect to life, and the primacy of faith" (Barzun, 1958: 305). "All universals accessible to the finite being are necessarily, in the words of Nietzsche, fictions--acts of mind. But without them we could not continue the game. Play is life itself, says Derrida" (Merrell 1985: 136). Perhaps cyber-oriented postmodernists may arrive at a workable solution to the problems of electronic publication of scientific information in an age of chaos and complexity; one cannot dismiss enthusiasts out of hand. There are, however, many critics of postmodernism, e.g. Neville (1992); Norris (1990), at least in those areas of discourse that may be reached by reason.
Kuhn (1970: 193), in an influential
work, pointed out that, because seeing is "theory-laden" and words
mean different things to different people, scientists "in some sense
live in different worlds" and each is "isolated in their own
systems of meanings as well as within his own universe of observed
things" (Sheffler, 1982: 16) and to that extent they cannot judge one
theory as superior to the other, an entirely postmodern viewpoint; this
incommensurability leads to abrupt, major paradigm shifts--scientific
revolutions. Many systematists have experienced postmodernist arguments and
world views, though probably not recognizing these as such. Some of us have
puzzled over recent books that announce an end to Science (e.g. Horgan,
1996). Science, apparently, has been negated because quantum mechanics has
shown that reality is composed of contradictory truths, indeterminacy, and
action at a distance, which "support the notion that reality is brought
into existence and maintained by the action of consciousness" (
Gerald Holton (1995), in a celebrated ringing damnation of postmodernism in science, warned that "today's version of the Romantic Rebellion, while strong in other fields, represents still only a seductive minority view among analysts and science policy makers, coming not up from the grass roots but down from the treetops. However, while it is held among prominent persons who can indeed influence the direction of a cultural shift, the scientists at large, and especially the scientific establishment, have chosen to respond so far merely with quiet acquiescence. If those trends should continue, and the self-designated postmodernists rise to controlling force, the new sensibility in the era to come will be very different indeed from the recently dominant one." Raloff (1996) quotes Holton as saying that "a 1992 draft of the new National Science Education Standards announced that they would be 'based on the post-modernist view'--one that questions the objectivity of observation and the truth of scientific knowledge.'" Though this was happily excised, the published National Science Education Standards (National Research Council, 1996: 201) now explains that "science is not separate from society but rather science is a part of society," covering itself, however, by adding that science is not based on "personal beliefs, religious values, mystical inspiration, superstition, or authority" though these may have "personal value."
It is the thesis of this writer that cladism--the ideology, and cladistics--the method, share many features with postmodernism and have in fact a common ancestry with that same movement, as have other recent "revolutions" in the physical sciences, including virtual reality, fuzzy logic and chaos theory. Cladistic partisans denigrate New Systematics in much the same way as those of postmodernism deprecate scientism, to wit, that both are authoritarian nonsense. What unites the revolutionary thrust of postmodernism with its scientific aspects is intensive use of the computer: postmodernists have discovered that it provides instant democracy (Terdeman, 1996), while scientists find it essential in dealing with complexity, including virtual modeling, fuzzy neural networks, strange attractors, and parsimony algorithms. Complexity is the new preoccupation of cutting-edge science (Gell-Man, 1995). Postmodernism, however, fuels the scientific revolutionaries in formalizing the deconstruction of past theory.
Curiously, most of the facies of revolution listed above have their emblems: postmodernism has the muted cornet symbol common in Thomas Pynchon's fiction, virtual reality has its glove, chaos theory has the owl's eyes of the Edward Lorenz attractor, and cladistics has the branching tree (fuzzy logic's emblem is unclear). Ashworth (1990) echoed a postmodern preoccupation in pointing out that "The emblematic world view is...the single most important factor in determining late Renaissance attitudes towards the natural world... Every kind of thing in the cosmos has myriad hidden meanings and that knowledge consists of an attempt to comprehend as many of these as possible." According to Ashworth, this is not a Medieval outlook, but something more evolved; it ended in the mid-1600's with a "decontextualization" of the world, a "desymbolization of nature," a sudden death of "animal semantics" (Ashworth, 1990). To post-modernists, this time was a catastrophe of knowing when words were separated from things, both nature and words suddenly lacked hidden meanings, and scientists became cautious about imposing patterns on nature.
The metanarrative is a particular focus of postmodernist ire; in the case of cladism, the New Systematics is found particularly evil. According to Simpson (1953), the term was first used by Hubbs (1934), and is a reaction to idealism and typology; what modern systematists of this school "classify is not a specimen but a population or series of populations existing in nature, of which the specimens are samples. The characteristics of a population are those of all its members collectively with all their resemblances and differences. Such a population has an average condition, but it has no single, crystallized, idealized pattern or morphotype." This was not a revolution in itself (Mayr, 1980: 43) but is a synthesis (Huxley, 1940) of taxonomic thought prevalent between 1920 and the present in which taxonomy was encouraged to reflect the results of studies in population biology, genetics, anatomy, ecology, cytology, etc. Macroevolutionary theory was minimized and the phrase "adaptive landscape" (Wright, 1932) was a shibboleth: species and supraspecies concepts, phylogeny and classification were to be determined by hypothesized evolutionary scenarios associated with adaptive zones, niches and major radiation (Simpson, 1953: 156).
The cladist has abandoned the second, theoretical aspect of the "dual nature of evolutionary change" (Mayr, 1976: 448). Eldredge and Cracraft (1980: 312) wrote, deconstructively, that "the expression 'adaptive radiation' is itself a value-laden description, hinting at the causes underlying the particular pattern." Stuessy (1990) has attempted to bring the syncretic method up to date as the New Phyletics, maximizing numerical contributions from cladistic, patristic and phenetic study in an effort to be more explicit and to some extent minimize theory. But being only partly numerical, with no defined single procedure or methodology, taxonomies generated under the New Systematics (or population systematics) never achieved classifications with the fine resolution and coherent methodological justification of the cladograms of evolutionary relationships based on the Hennig's synapomorphy criterion. The numerical taxonomists Sokal and Sneath (1973) were harsh: "...the New Systematics contributed little to our understanding of the nature and evolution of the higher categories and of taxonomic structure in general. Books...[they include Hennig, 1966] deal with the latter topics but they contain little more than descriptive generalizations." Yet the field of phenetics, championed by Sokal and Sneath is now in the doghouse. Clearly, it is not only theory, the wondrous paladin of progress, that is now considered counter-revolutionary.
It is clear, to this writer at least, that scientists use intuition and induction to generate hypotheses, and use a loose form of the hypotheticodeductive method (as discussed by McMullin, 1990) to find support for theories, as when manifold results of experiment or multiple observations agree with theory and successful exact predictions can be made. Karl Popper's (1965) special form of the latter method, involving complete rejection of induction and the restriction of the testing of a hypothesis exclusively to its empirical consequences gives a kind of philosophical credence to those who reject theorization in context of the New Systematics, no matter what experimental evidence that theory is based on. This is a kind of narrow operationism that enhances the chance of overlooking internal invalidity (Neale & Liebert, 1973). As Chesterton (1956: 156) solved the problem for his fellow True Believers, "true premises produce a true conclusion. What is called induction seems simply to mean collecting a larger number of true premises.... What is pompously called induction is simply collecting more of the data," and scientists know "that their conclusions would not be true unless their premises are true.... But in the final process of truth there is nothing else except a good syllogism"; much of induction is simply wishing, according to this religious apologist. Building on facts is good science, and building on theory is problematic, but good theory built on facts can guide science. To the extent that the New Systematics is built on facts, basing classification on such theory is defensible. Surely restricting evolutionary studies to an algorithm is reprehensibly simplistic. Hedberg (1995, 1996) expressed well an apprehension that cladistics makes assumptions that are not met with in the real world.
Wiley (1981: 138) lamented the low resolving power of theoretical definitions of homology as a dilemma that "is the direct result of the exclusive use of induction in framing hypotheses of homology and hypotheses of phylogenetic relationships.... The 'problem of homology' does not exist, however, if we admit phylogenetic relationships based on other characters as a major criterion in the hypotheticodeductive mode of hypothesis testing." Hennig (1966) presented series of logical deductions as a scheme of argumentation leading to a correct phylogenetic tree. Wiley, as others, ducked the problem of probability that any hypothesis is correct by referring to the fact that "the one true phylogeny for any given set of organisms will never be known" while at the same time restricting testing to "the congruence or incongruence of a particular hypothesis of synapomorphy with other hypotheses of synapomorphy in an open system of the testing of competing phylogenetic hypotheses." Popperian refutation of a hypothesis is limited to parsimony, i.e. a shorter tree.
Eldredge and Cracraft (1980: 248) discussed at length that "nearly all attempts by synthesists to deal with macroevolution, at least until recently, fail even to mention species, much less to incorporate hypotheses pertaining to species into the analysis" pointing out that principles based on a well-corroborated within-population phenomenon (fitness differentials of natural selection) are applied wholesale across the taxonomic hierarchy, there being "no necessary relationship between natural selection and speciation, at least in terms of allopatric speciation." Whether they are right or not, the drubbing that New Systematics has received is most significant as an extravagant gesture of political polarization designed to highlight the new method of cladistics.
Comparing pre-Hennigian systematics with cladistics is comparing a straw dog to a bloodhound. According to Crowe (1994), "the New Systematics...was riddled with subjective approaches, e.g. arbitrary a priori character weighting and models of evolutionary process, which could be used to distort phylogenetic hypotheses almost at will by invoking preferred adaptational models or evolutionary just-so stories." Postmodernists would agree: "the writer's most effective weapons against [the silence imposed by expertise and complexity] have always been simplicity and common sense" (Saul, 1992: 7). Cladistics, of course, does not say genetics, population biology, cytology, etc. are not valuable fields of study, just that there is no relevance of these fields to uncovering phylogenetic relationships. In this total rejection, a "fact-proof shield" protects the True Believer (Hoffer, 1951). This is protection from the uncomfortable fact that the bloodhound of cladistics finds trees bearing characters at the ends of branches not taxa.
The "just-so stories" (Kipling, 1966--"How the Leopard Got His Spots" etc.) referred to by Crowe (1994) above and in the Hennig Society Homepage are the baleful metanarratives of the "theory laden" New Systematics. Neo-Darwinism and the modern synthesis are not all bad, having been shown to be strong in some respects and shaky in others by modern writers (e.g. K. Smith, 1992). Besides, according to the historian Fisher (1970:131) "All scientists use story schemas some of the time...." The above ridicule may be seem justified, however, by what appear to be the striking results of cladistics. The Process Structuralists "paint neo-Darwinism in the blackest terms without offering anything very concrete as an alternative" (K. Smith, 1992), but cladistics appears to provide a major new way of approaching evolutionary analysis. In fact, the astoundingly high resolution of the parsimony method cannot be fully generalized to a modeling of ancestors, and we are actually all in the situation of struggling to infer on the basis of poor data "how the leopard got his spots."
Postmodernists have criticized Kepler's astronomy because it came out of his medieval religious views, Darwin's evolution because it was motivated by his desire to perpetuate his privileged social class, and Newton's rejection of Descartes' philosophy because it might challenge conventional religion (Sagan, 1995). Similar ad hominum or otherwise off-center accusations about New Systematics or pre-Hennigian taxonomy in general abound in the cladistic literature; Crowe (1994), for instance, complained that limited funds for systematics research should not be spent on "attempts to impose underlying and unjustifiable process assumptions to historical phenomena," these being "much to do about nothing." Clearly purveyors of traditional taxonomy, as postmodernists feel is the case with all scientists, sustain their funding with nonsense. There has not been an attempt at a real comparison of scientific value between the taxonomic results of New Systematics and cladistics; there should be and it should not be judged circularly.
The Hennig Society summary goes on to say that: "Cladistics produces hypotheses about the relationships of organisms in a way that, unlike other systems, predicts properties of the organisms." And what are these other non-predictive systems? The Homepage provides the following headings in its Education section: "The Paradiso Tour: what is cladistics? The Purgatorio Tour: other so-called phylogenetic methods. The Inferno Tour: phenetics, synthetics and related sins." The demonization of other methods and fields of study militates against consilience--the support of other methods of investigation, through which reasoned discourse promotes a general consensus on what are scientific facts (cf. Hull, 1983: 184).
One might ignore the implications of the Hennig Society Homepage, attributing its excesses in cladistic propaganda to youthful exuberance or to something akin to the teleological explanations we all sometimes indulge in--but for the fact that cladistics is flawed, and its practitioners are now generating students. Unlike Lysenkoism, cladistics is based on a definite advance. Compare a cladogram, for all its problems, with Simpson's (1953: 387) uncomfortable concept of an evolutionary or phyletic species that transforms from one species over time into another, anagenetically, crossing at some point an arbitrary boundary; this does not seem useful even in paleontology even if "the early and the late forms differ as much as species usually do, and if taxonomy is to be significant and useful they must be distinguished." The Hennigian algorithm really does reflect a Darwinian evolutionary theory of simple relationships and has considerable resolving power given phylogenetically informative data. It is the unskeptical emphasis on the straight Hennigian method and that method alone that makes cladism an ideology, and it is the modeling of hypothetical ancestors on the basis of insufficient evidence (character distribution on a cladogram) that makes the cladogram no ultimate solution. To the extent that one stands on one's own shoulders, one finds one's head in a narrow place.
Promoting polarization helps define and enliven a political or ideological movement; any resistance empowers the new ideas. Many of the self-referencing and deconstructionist faults of cladism noted above have been pointed out before but not in the context of their common origin in a widespread and robust Romantic rebellion against 19th and 20th Century Enlightenment values, billed by partisans as a genuine "paradigm shift" happening in most fields of science and art. Although rebellion and rejection are characteristic of this shift, modern systematists should eschew participating in a polarizing reaction because a useful classification based at least in part on an evolutionary framework requires both Apollonian syncretism and Nietzschian methodological fervor.
That cladistics have long been of interest to postmodernists for use in textual criticism (reciprocal illumination" is much like the "hermeneutic circle" according to M. Schmitt, email@example.com) and linguistics (Platnick & Cameron, 1977; Discussion Group PMC-Talk, 1992) is not in itself problematic. But, important to systematists, the cladistic method is much the same as Rawlsian proceduralism. Richard Rorty (1991), the neopragmatist, has explained J. Rawls' (1971, 1985) method as follows: "Rawls thinks that 'Philosophy as the search for truth about an independent metaphysical and moral order cannot...provide a workable and shared basis for a political conception of justice in a democratic society.' So he suggests that we confine ourselves to collecting 'such settled convictions as the belief in religious toleration and the rejection of slavery' and then 'try to organize the basic intuitive ideas and principles implicit into a coherent conception of justice.'" Rorty indicated that one cannot free oneself from history and tradition by an appeal to Nature or Reason. In that we are all different, we should abandon Enlightenment values, such as making everyone accept some global moral objective. We should just stay in our moral clubs and come out only to "haggle" with others, otherwise ignoring people unlike ourselves culturally. Rawlsian procedural justice will suffice for all problems.
Rawls' proceduralism hits uncomfortably home in its pragmatic reflection in systematics today; I suggest scientists read "scientific" for "moral" or "cultural" above. The proper, required proceduralism of systematics is, according to cladists, a judgmental paradigm involving a simple, generally agreed upon version of evolution (much like Darwin's own idealized concept as opposed to "overall similarity" fide Stuessy, 1990: 57).
THE OBSCURE AND MYSTIFYING
Examples of hubris in the revolutionary literature are many. According to the cladist Brooks (1981), "the only necessary and sufficient criterion for recognizing that evolution has occurred is the discovery of synapomorphies." "Cladistics are the analytical process by which phylogenetic histories are inferred" (Snow, 1996). According to the fuzzy logician Kosko (1993): "One day I learned that science was not true... They [scientists] said that all things were true or false..." but "All facts were matters of degree." The virtual realist Woolley (1992) wrote: "Science only admits into its canon those theories that can be rejected by it [!]" therefore self-centeredly refusing to investigating astrology, aromapathy, divination, ESP and telekinesis, while he fratricidically decried "postmodern autism" because virtual realists such as himself are "upbeat and forward-looking." Gleick (1987: 3, 38, 39) described chaos theory: "...Chaos has become a shorthand name for a fast growing movement that is reshaping the fabric of the scientific establishment" and "chaos has become not just theory but also a method, not just a canon of beliefs but also a way of doing science.... To chaos researchers, mathematics has become an experimental science, with the computer replacing laboratories full of test tubes and microscopes," also "Stylistically, early chaos papers recollect the Benjamin Franklin era in the way they went back to first principles...." also "Revolutions do not come piecemeal. One account of nature replaces another." Gleick quoted Joseph Ford: "Evolution is chaos with feedback." Like cladists, chaoticists reject theory in favor of the simplicity and commonsense of their synthetic system: "Natural selection operates not on genes and enzymes but on the final product. So an adaptationist explanation for the shape of an organism or the function of an organ always looks to its cause, not its physical cause but its final cause. Final cause survives in science wherever Darwinian thinking has become habitual... It became a rich and fruitful problem to explain a leaf in terms of how natural selection shaped such an effective solar panel [but] leaves come in just a few shapes, of all the shapes imaginable; and the shape of a leaf is not dictated by its function" (Gleick, 1987: 201).
These neo-romantic monomethodical systems are actually only incremental or at most situational alternative advances hidden behind a buzz of appeals to an irrational, ideological revolution involving cant about paradigm shifts. Polly (1993) agreed with Hull (1988) that "science is a process more like evolution than revolution... The synthesis view of [G. G.] Simpson was the product of more than sixty years of experimentation in portraying evolution in classification after 1859. It is equally the product of historical precedent, advances in scientific method, individual influence, and chance. The current cladistic 'revolution' has been grinding on for more than thirty years at this point and has had a similar history...."
Mayr (1976), in a thorough criticism of cladistics, puzzled over "no hope for a meeting of minds" while at the same time being largely dismissed as a contributor to evolutionary systematics by Eldredge (1982) in the introduction to the reprinted edition. Writing of Darwinian, Marxian and Wagnerian "revolutions," Barzun (1958: 323) concluded their "form was System, a clear claim to the title of ism. From each of these isms the public learned that the riddle of the Sphinx had been solved, but that the solution was somewhat technical and intricate; a new language was to be mastered, whose universal use would revolutionize the world. Yet at bottom lay a simple principle [for each system]." As Morison (1979) has pointed out, as the wheel of history turns, ironically "the current tendency of those with high scores in quantitative studies to dismiss the literary arts as "trivial" is simply a distant echo from the masters of the medieval quadrivium similarly derisive of the merely verbal studies of the trivium."
The syncretic struggle to find a more complex analytical approach that matches a more complex, global understanding of nature is found to be anathema (a "sin," above) by cladist partisans. This is not a minor point. The paradigm shift is supposedly upon us; enthusiastic new doctorates trained in cladistics are now replacing syncretic taxonomists. Cladist zealots are now recommending major changes in the rules of nomenclature to better reflect classifications based solely on parsimony analysis. Mickevich (1983) unambiguously equated classification theory with cladogenesis. As it turns out, the International Code of Botanical Nomenclature (as well as its faunal counterpart) is an internally consistent (e.g. the burden of autonyms!) and supposedly scientific procedure with an associated vast cryptic literature that desperately needs streamlining and reality-testing. On the other hand, a completely new code would serve only to fix cladogenesis as the only way to group taxa. Because nodes on cladograms are not completely generalizable as ancestral species, and thus cladograms actually have far less resolution than they are presented as having, the phylogenetic relationships of a large proportion of taxa cannot be established in any detail. [See also my recent critique of statistical phylogenetics: Zander 2001.] The old ranking system works fine for the groupings of the hierarchical results of cladistics and remains of value precisely because it is old and well-entrenched. We, as taxonomists, have a duty to make the present Code less legalistic and more accessible by the uninitiated (e.g. Names in Current Use and the proposed Biocode--Greuter, 1996; Greuter & al., 1996), or a revolutionary with an axe to grind will do it for us.
Postmodernism shares with New Age partisans some disconcerting attitudes about objective truth: "if an experience seems real, it is real" (Schick & Vaughn, 1995). The monomethodical quick fix to intransigent problems in science (virtual reality, fuzzy logic, chaos, parsimony analysis) is quite similar to the isolated, circular, irrational cultural artifact (like language, mores, rituals and fashions) that must be respected by self-referencing postmodernists. To many of us when first confronted with cladistics, it seemed magical the way nicely resolved cladograms of phylogenetic relationships could be generated from large, complex data sets. A. C. Clarke is credited with the observation that "a sufficiently advanced science is indistinguishable from magic." The reverse is also true.
CLADISTICS AND FINE STRUCTURE
The cladogram inspires respect because of its apparent high resolution of evolutionary relationships, especially when a fully resolved cladogram is presented. A series of problems having to do with evolutionary theory as over-simplified in cladistics are reviewed by various authors (e.g. Felsenstein, 1978, 1983; Stuessy, 1990: 126-128; Zander, 1993), but these papers alike suggest no numerical solution, no attractive alternative to the elegant cladistic method.
Saether (1986) pointed out that because parallel changes often exceed unparallel changes in the real world, much outside information and theory in addition to the parsimony analysis is required for analysis below the family level. He reiterated Hennig's (1966) insistence on the importance of intuition and subjective evaluation in systematics, and concluded his treatment of convergence with the observation that objectivity in parsimony analysis is a myth. Apropos of this, Medawar (1984) ably defends intuition in science.
In a study using artificial phylogenies mimicking actual morphological data sets, Lamboy (1994) found that maximum parsimony is poor at finding the true phylogenetic tree: "for simulated data sets having values of the consistency index in the range reported in the literature, maximum parsimony found the true phylogenetic tree only from 0.7% to 27.8% of the time...." and he cautioned against "using phylogenies reconstructed from morphological data to make inferences about biological processes or patterns..." because of the apparent high frequency of multiple character state changes per speciation event, parallelisms, reversals, extinctions and phyletic changes. Convergence may also easily obtain by chance alone as discussed by Armbruster (1992), or genetic polymorphism persisting through speciation events, producing identical morphological variants at places on the tree (discussion by Felsenstein, 1983). Mayr (1976) noted rightly that Hennig was incorrect in his claim that determining whether a particular character state is present in a stem species will distinguish between corresponding character states in several related species as synapomorphies, convergences, homologies, or parallelisms; actually only apomorphy as against the other three are decided.
The autapomorphies of the terminal taxa will become the synapomorphies of their derived taxa; so modeling is enabled to the degree to which parsimony analysis constructs ancestral nodes with their autapomorphies retained as synapomorphies. If, on the other hand, the average number of autapomorphies of the terminal taxa subtending a node is more than the number of synapomorphies at that node, then the algorithm becomes top heavy with inferences from too few characters (fallacy of false interpolation, which requires full information to avoid, Fischer, 1970: 122). In practice (Zander, in ms) one might expect about half the nodes of the cladogram to actually model ancestral species. Cladograms therefore usually do not completely model phylogenetic trees.
I submit that, while cladists may feel the systematic bottle is finally nearly full, and others that it is emptying fast, that the bottle is about half full. The probability that a cladistic solution is correct is determined solely by how well the selected terminal taxa retain phylogenetic information, while the shortest tree maximizes possible synapomorphies as possible solutions. Nodes lacking a number of synapomorphies equal to the average number of autapomorphies of that branch must collapse. Requiring generalizable models of ancestors indicates that much of the highly resolved fine structure of most published cladograms is altogether too sanguine and logically insupportable.
Plank (1949) remarked that there exist phantom problems that are insoluble, either because no indisputable method of solution exists or because the question is unreasonable, and that one should ascertain at the outset whether the problem is genuine and a solution expected. Mayr (1976: 431) was of the opinion that "the idea of one ideal natural system is a phantom." It is probable that much information about phylogenetic relationships is lost and that many potential shared ancestors may never be modeled on firm data.
NEW SYSTEMATICS DA CAPO
The New Systematics emphasized grouping by as many characters as possible, often polythetically, and usually resulted in only a comparatively few, poorly resolved groups of hypothetically related taxa; while cladistics usually presents a highly resolved nested series of related groups, but these groupings are often based on few characters compared to those possessed by real taxa. It is this ungeneralizable fine structure that is logically objectionable because of problems with convergence associated with high autapomorphy to synapomorphy ratios. A return to a dissatisfaction with "one-character taxonomy" (in groups for which one character differentiation is seldom taxonomically sufficient) seems warranted. Taxonomy is, however, seldom acknowledged by cladists as a worthwhile endeavor, and has, apparently, ended, just as science and history have ended for the postmodernist. Taxonomy is, for instance, totally eliminated from one recent definition our field: "...phylogenetics and biogeography, together referred to as systematics" (Erwin, 1991). A merely useful classification is itself apparently not a worthy end, though even Hennig (1966: 7) wrote that the task of systematics is to create a general reference system. O'Hara (1988, 1996) has demonstrated the apparent efficacy of cladistics in solving problems in stemmatics and has proposed a formal name for the "historical sciences" (presumably based on numerical analyses) using Whewell's term "palaetiology"; one might urge that the definitions of "ancestral" nodes of such analyses match those of the terminal elements to ensure generalization.
Mayr (1976: 471) wrote that "A sound classification of a group of organisms cannot be devised without a well-considered reconstruction of its phylogeny," but I feel that although "An eclectic classification which considers with equal care the branching points in phylogeny and all aspects of phylogenetic divergence would seem the best way to generate biologically meaningful classifications," this ideal is too lofty because it presupposes a wealth of data. If sufficient model ancestors cannot be found on which to base a phylogenetic classification satisfactorily as outlined above in a particular study, the next best way of developing a predictive hypothesis is to impose outside theory of special evolutionary scenarios, such as weighting for perceived convergence; if the cladogram remains poorly resolved, one can group the taxa phenetically, i.e. maximizing all shared characters; and if that proves poorly resolved, then a useful artificial classification must be developed since a classification is needed if only to pigeonhole information for later retrieval. But will any alternative to straight Hennigian parsimony, even in the form of the New Phyletics (Stuessy, 1990) melding cladistic, patristic and phenetic numerical study, be considered an option to the new generation of systematists?
Must cladistics, too, like the New Systematics and phenetics, be broken on the wheel of history by the next newly fashionable methodology, some parvenue nemesis complete with emblem and scurrilous, partisan web page? Or may cladistic methods and successes be added to those of New Systematics and phenetics to better illuminate different aspects of nature as part of the armamentarium of scientific study? Hedberg (1995) made a strong point that cladistics should be viewed as part of a systematics "toolkit." Kuhn (1970) had an elegant vision that progress in science means that whole concepts are destroyed and replaced in a series of revolutions, and only the list of explicable phenomena grows. Must this be so?
In sum, the revolutionary paradigm shift is in the phenomenal utility of personal computers in dealing with complex modern issues in communications, sociology, textual criticism and science, unfortunately involving over-simplification, reductio ad absurdum, deconstruction of past less-computerized methodologies, and insistence on monomethodical modes of analysis. According to O'Hara (Discussion Group PMC-Talk, 1992), among others, cladistics "constitutes a genuine conceptual revolution, one that has not only intellectual components, but all the characteristic socio-disciplinary turmoil that accompanies a scientific revolution as well." Stuessy (1990: 130) thought to be exaggerated such claims. According to L. Marx (1979), "It would be a serious mistake...for those concerned about the future of science to underestimate the appeal, or the force, of the neo-romantic critique of science as a mode of knowing built on an inadequate metaphysical foundation." Cladistics is not the revolution, just a part of it. The greater revolution is a vainglorious over-response in many intellectual fields to the real opportunities presented by the computer. Postmodernism has avoided nihilism, but the new sciences discussed above require the destruction of existing institutions before rebuilding them on a new basis. This multiplex revolution in science is the emergence of neo-romantic monomethodical systems of analysis or synthesis that are intended to be not just incremental or alternative but total replacements for the shoulders of giants.
Ashworth, W. B., Jr. 1990. Natural history and the emblematic world view. Pp. 303-332. In D. C. Lindberg & R. S. Westman, Reappraisals of the scientific revolution. Cambridge.
Barzun, J. 1958. Darwin, Marx, Wagner: critique of a heritage. Second Edition. Garden City, N.Y,
Brooks, D. R. 1981. Classifications as languages of empirical comparative biology. Pp. 61-70. In V. A. Funk & D. R. Brooks, Advances in cladistics. Proceedings of the first meeting of the Willi Hennig Society. Bronx, New York
Burtt, E. A. 1932. The metaphysical foundations of modern science. London.
Campbell, L. 1996. The postmodern method: science. Xenos Christian Fellowship. Crossroads Project: www.xenos.org
Chesterton, G. K. 1956. St. Thomas Aquinas. Garden City, New York.
Cohen, M. R. & E. Nagel. 1934. An introduction to logic and scientific method. New York.
Crichton, M. 1990. Jurassic Park. Ballantine Books Edition, 1991, New York.
Crowe, T. M. 1994. Morphometrics, phylogenetic models and cladistics: means to an end or much to do about nothing? Cladistics 10: 77-84.
de Queiroz, K. 1992. Phylogenetic definitions and taxonomic philosophy. Biology and Philosophy 7: 295-313.
Discussion Group PMC-Talk. 1992. Two threads: cladistics and cut-ups. (Excerpted from the discussion group pmc-talk@ncsuvm, 7/92-8/92). Postmodern Culture 3(1) September, 1992: firstname.lastname@example.org
Eldredge, N. 1982. Introduction. Pp. xv-xxvii. In E. Mayr, Systematics and the origin of species. New York. Reprint of 1972 Edition.
Eldredge, N. & J. Cracraft. 1980. Phylogenetic patterns and the evolutionary process: method and theory in comparative biology. New York.
Eliot, K. & M. Cantsin. 1996. A note from the editors of "Smile." Smile Magazine No. 5 of 1-17, 20 Aug. 1992: www.neoism.org
Erwin, L. T. 1991. An evolutionary basis for conservation strategies. Science 253: 750-752.
Felsenstein, J. 1978. Cases in which parsimony or compatibility methods will be positively misleading. Syst. Zool. 27: 401-410.
Felsenstein, J. 1983. Parsimony in systematics: biological and statistical issues. Ann. Rev. Ecol. Syst. 14: 313-333. h
Fischer, D. H. 1970. Historians' fallacies: toward a logic of historical thought. New York.
Gell-Mann, M. 1995. The quark and the jaguar. San Francisco.
Gleick, J. 1987. Chaos: making a new science. New York.
Golinski, J. V. 1990. Chemistry in the Scientific Revolution: problems of language and communication. Pp. 367-396. In D. C. Lindberg & R. S. Westman, Reappraisals of the Scientific Revolution. Cambridge.
Greuter, W. 1996. On a new BioCode, harmony, and expediency. Taxon 45: 291-294.
Greuter, W., D. L. Hawksworth, J. McNeill, M. A. Mayo, A. Minelli, P. H. A. Sneath, B. J. Tindall, P. Trehane & P. Tubbs. 1996. Draft BioCode: the prospective international rulles for the scientific names of organisms. Taxon 45: 349-372.
Hatfield, G. 1990. Metaphysics and the new science. Pp. 93-166. In D. C. Lindberg & R. S. Westman, Reappraisals of the Scientific Revolution. Cambridge.
Hedberg, O. 1995. Cladistics in taxonomic botany--master or servant? Taxon 44: 3-11.
Hedberg, O. 1996. Reply to Neil Snow's 'The phylogenetic paradigm of comparative biology.' Taxon 45: 91-92.
Hennig, W. 1966. Phylogenetic systematics. Urbana.
Hoffer, Eric. 1951. The true believer: thoughts on the nature of mass movements. New York,
Holton, G. 1995. Einstein, history, and other passions. Woodbury, New York.
Horgan, J. 1996. The end of science: facing the limits of knowledge in the twilight of the Scientific Age. New York.
Hubbs, C. L. 1934. Racial and individual variation in animals, especially fishes. Amer. Nat. 68: 115-128.
Hull, D. L. 1983. Karl Popper and Plato's metaphor. In N. I. Platnick & V. A. Funk (eds.), Advances in cladistics. 2: 177-189.
Hull, D. L. 1988. Science as a process. Chicago.
Huxley, J. 1940, ed. The new systematics. Oxford.
Jary, D. & J. Jary. 1991. The Harper Collins dictionary of sociology. New York.
Kipling, R. 1966. Just so stories. Illustrated by the author. Airmont Edition, New York.
Kosko, B. 1993. Fuzzy thinking: the new science of fuzzy logic. New York.
Kuhn, T. S. 1970. The structure of scientific revolutions. Second edition, enlarged. Chicago.
Lamboy, W. F. 1994. The accuracy of the maximum parsimony method for phylogeny reconstruction with morphological characters. Syst. Bot. 19: 489-505.
Lindberg, D. C. 1990. Conceptions of the Scientific Revolution from Bacon to Butterfield: a preliminary sketch. Pp. 1-26. In D. C. Lindberg & R. S. Westman, Reappraisals of the Scientific Revolution. Cambridge.
Lyotard, J.-F. 1986. The postmodern condition. Transl. G. Bennington & Brian Massumi. Manchester.
Marx, L. 1979. Reflection on the Neo-Romantic critique of science. Pp. 61-74. In G. Holton & R. S. Morison. Limits of scientific inquiry. New York.
Mayr, E. 1976. Evolution and the diversity of life: selected essays. Cambridge, Massachusetts.
Mayr, E. 1980. Prologue: some thoughts on the history of the evolutionary synthesis. Pp. 1-48. In E. Mayr and W. B. Provine. 1980. The evolutionary synthesis. Cambridge, Massachusetts.
McMullin, E. 1990. Conceptions of science in the Scientific Revolution. Pp. 27-92. In D. C. Lindberg & R. S. Westman, Reappraisals of the Scientific Revolution. Cambridge.
Medawar, P. 1984. Pluto's republic. Oxford.
Merrell, F. 1985. Deconstruction Reframed. West Lafayette, Indiana.
Mickevich, M. F. 1983. Introduction. Pp. 3-5. In N. I. Platnick & V. A. Funk, Advances in cladistics. Volume 2. Proceedings of the second meeting of the Willi Hennig Society. New York.
Morison, R. S. 1979. Introduction. Pp. ix-xviii. In G. Holton and R. S. Morison, Limits of scientific inquiry. New York.
National Research Council. 1996. National science education standards. Washington, D.C.
Neale, J. M. & R. M. Liebert. 1973. Science and behavior: an introduction to methods of research. Englewood Cliffs, New Jersey.
Neville, R. C. 1992. The highroad around modernism. Albany, New York.
Norris, C. 1990. What's wrong with postmodernism? Baltimore.
O'Hara, R. J. 1988. Homage to Clio, or, toward an historical philosophy for evolutionary biology. Syst. Zool. 37: 142-155.
O'Hara, R. J. 1996. Palaetiology. http://rjohara.uncg.edu/palaetiology.html
O'Sullivan, J. Hartley, D. Saunders, M. Montgomery & J. Fisk, eds. 1994. Key concepts in communication and cultural studies. 2nd ed. New York.
Piercy, V. 1995. FAQ alt.postmodern newsgroup ver. 1. Ftp://rtfm.mit.edu/pub/usenet.
Planck, M. 1949. Scientific autobiography and other papers. New York.
Platnick, N. I. & H. D. Cameron. 1977. Cladistic methods in textual, linguistic, and phylogenetic analysis. Syst. Zool. 26: 380-385.
Polly, P. D. 1993. Hyaenodontidae (Credonta, Mammalia) and the position of systematics in evolutionary biology. Ph.D. dissertation, University of California at Berkeley.
Popper, K. 1965. The logic of scientific discovery. Revised Torchbook Edition (orig. publ. 1959), New York.
Raloff, J. 1996. When science and beliefs collide. Science News 149: 360-361.
Rawls, J. 1971. A theory of justice. Cambridge, Massachusetts.
Rawls, J. 1985. Justice as fairness. political--not metaphysical. Philosophy and Public Affairs 14: 225-251.
Rorty, R. 1991. Objectivity, relativism and truth. New York.
Saether, O. A. 1986. The myth of objectivity--post-Hennigian deviations. Cladistics 2: 1-13.
Sagan, C. 1995. The demon-haunted world: science as a candle in the dark. New York.
Scheffler, I. 1982. Science and Subjectivity. Second Edition. Indianapolis.
Schick, T., Jr. & L. Vaughn. 1995. How to think about weird things: critical thinking for a New Age. Mountain View, California.
Scott-Ram, N. R. 1990. Transformed cladistics, taxonomy and evolution. Cambridge.
Simpson, G. G. 1953. The major features of evolution. New York.
Smith, C. 1975. The best of Cordwainer Smith. J. J. Pierce, ed. Garden City, New York.
Smith, K. C. 1992. Neo-Rationalism versus neo-Darwinism: integrating development and evolution. Biology and Philosophy 7: 431-451.
Snow, N. 1996. The phylogenetic pradigm of comparative biology--a response to Hedberg. Taxon 45 85-90.
Sokal, R. R. & P. H. A. Sneath. 1973. Numerical taxonomy, the principles and practice of numerical classification. San Francisco.
Stebbins, G. L., Jr. 1950. Variation and evolution in plants. New York.
Stuessy, T. F. 1990. Plant taxonomy: the systematic evaluation of comparative data. New York.
Terdeman, S. 1996. Web authoring. PC Magazine 15(15): 115-117.
Whewell, W. 1857. History of the inductive sciences. 3rd Ed. London.
Wiley, E. O. 1981. Phylogenetics: the theory and practice of phylogenetic systematics. New York.
Woolley, B. 1992. Virtual worlds: a journey in hype and hyperreality. Oxford.
Wright, S. 1932. The roles of mutation, inbreeding, crossbreeding, and selection in evolution. Proc. VI Int. Congr. Genetics 1: 365-366.
Wynne-Davies, M. 1990. The Prentice-Hall guide to English literature. Englewood Cliffs, New Jersey.
Zander, R. H. 1993. Genera of the Pottiaceae: mosses of harsh environments. Bull. Buffalo Soc. Nat. Sci. 32: vi + 378 pp., 113 pls.
Zander, R. H. 1995. Phylogenetic relationships of Hyophiladelphus gen. nov. (Pottiaceae, Musci) and a perspective on the cladistic method. Bryologist 98: 363-374.
Zander, R. H. 2001. A conditional probability of reconstruction measure for internal cladogram branches. Syst. Biol. 50: 425-437.
This paper is not much dated, even with the great advances recently in statistical phylogenetics. At the present time, this same postmodernist attitude accounts for the push to expend vast resources, including National Science Foundation support, for a grand Tree of Life, one that seldom provides a measure of reliability at any one node though (naturally) generally reflective of our ideas of evolution. The Tree of Life's main fault that there is no reliability measure presented for problematic branch arrangements. Every instance of support for obvious (uncontested, accepted) branch arrangements is a misdirection of attention away from problematic relationships. It is the well supported elucidation of problematic relationships that constitutes an advance in evolutionary understanding.
In addition, there a major push on the Phylocode as a substitute or alternative to the International Codes for plant and animal nomenclature is in train. The Phylocode, again, is an attempt by partisans of a new science to set their own standards (of reliability in this case) because other (or previous) sciences are incommesurate in the Kuhnian sense. The Phylocode is based on one simplistic method of organizing taxa, and problematic areas are dealt with arbitrarily under the assumption that resolution is sufficient for reliability. This is incorrect because shortest trees generated from random data sets are (almost) all well resolved. See also my "The Number of Gene Trees Necessary for a Probabilistic Reconstruction of the Species Tree."
I now feel that I am sufficiently right or at least insufficiently wrong in this paper to at least squeeze it onto the Web. Yes, it does appear as though I scanned the literature and harvested any quotations that seemed to support my philosophical stance, but if one is right, this is okay, isn't it? -- R. Zander, January 18, 2003.
[Note: Kosko (1993) stated on p. 14 that "The yin-yang symbol is the emblem of fuzziness." (Think your own comment now.)]
Another note, September 30, 2004: For a delightful, erudite, compelling rendering of the major historical aspects of all of human culture and science into postmodern drivel, read:
James Burke. 1985. The Day the Universe Changed. Little, Brown and Co.; Boston.
He points out that Linnaeus sought to know the Mind of God by naming all the plants and animals, this in his context that paradigms have changed since then and so much for taxonomy.
Burke's book is interesting in no small part because of the challenges to the reader in detecting the logical fallacies: huge leaps assuming causal connections, arguing from the small to the large and vice versa, concepts meant to be restricted to one situation applied to other situations, and so on. The text is attractive for its pyrotechnics and interesting connections, yet the last chapter is horrific in its party-line postmodern condemnation of science and rationality.