Cladistics and Eclecticism
skala at INCOMA.CZ
Mon Feb 11 09:20:11 CST 2002
>This sounds like a description of character compatibility analysis
>(with its emphasis on maximal "cliques" ...
>...not of parsimony methods in general.
>What we really do in cladogram construction is summarizing the "amount of >character change" (i.e. the largest clusters of mutually compatible
>synapomorphies) to construct/recognize "real" clades.
It was not my intent to restrict on the clique analysis. This description covers at least the usual congruency analysis as well - the difference is if you optimize "by rows" (= by species/nodes optimizing at each clade as a separate unit; then you have congruency analysis) or "by columns" (= by characters optimizing among characters through all cladogram; then you have clique analysis).
But, you are right that my description does not apply to all parsimony methods - e.g. the maximum likelihood estimation is indeed based on parsimony but is not covered by the above definition. My intention, however, was to discuss the "standard model" of cladistics.
More information about the Taxacom