Alasdair,
Oooh, my favourite!
"If I pulled a fossil of a chimp from 5000 years ago and a compared it to the skull of a man I can make the conclusion that one came from the other because of similarities, but I would have no proof of that only an assumption based on a preconceived idea."
No, because that would ignore the fact that any analysis would place the skull in the chimp lineage. In fact, I doubt whether 5000 years would be enough to even place the skull as being an "archaic" chimpanzee.
Were we to imagine a scenario in which evolution didn't occur, then the null hypothesis that your opponent postulates would be that fossils would appear "in-sequence" on occasion, but the rule is that generally you can't make lineages out of the older to younger fossils. In fact, the correlation between cladistics (ordering & ranking of organisms into groups based on (in this case) morphological characters) & where they are found in the fossil record (stratigraphy) correlates to a statistically meaningful degree. In other words, the null hypothesis is falsified & the evolutionary hypothesis is supported.
Assessing Congruence Between Cladistic and Stratigraphic Data
Stratigraphic Consistency Index
"The SCI metric may also be summarized either as a mean value for each taxonomic group or as a proportion of cladograms that score SCI values of 0.500 or more, an indication that half, or more, of the branches are consistent with stratigraphic evidence. By both measures, fishes and echinoderms score better than tetrapods. Mean SCI values are: echinoderms (0.773), fishes (0.757), and tetrapods (0.701). Proportions of cladograms with SCI values $0.500 are tetrapods (100%), echinoderms (94%), and fishes (93%). For both measures, values for all three groups are indistinguishable according to binomial error bars (Fig. 3).
Within the sample of echinoderm cladograms, nonechinoids show somewhat better results than echinoids but not significantly so (Fig. 3). The mean SCI value for echinoids is 0.724, and for nonechinoids 0.849; moreover, 90%of echinoid cladograms have SCI values $ 0.500,compared with 100% for nonechinoids.
SCI values for fish groups are variable but not significantly different (Fig. 3). For mean SCI values, the order is as follows: sarcopterygians (0.904), teleosts (0.744), placoderms(0.741), agnathans (0.733), and actinopterygians (0.722). In all cases, all sampled cladograms show SCI values > 0.500. The rankings of tetrapod groups by both aspects of the SCI metric are comparable. Mean SCI values give this sequence: mammals (0.837), “mammallike reptiles” (0.729), lepidosauromorphs (0.714), dinosaurs (0.698), archosauromorphs (0.660), and turtles (0.586). The low value for turtles is significantly lower than the high values for synapsids, mammals, and “mammallike reptiles”. Proportions of cladograms with SCI values $ 0.500 give this sequence: mammals (100%), “mammallike reptiles” (100%), lepidosauromorphs (100%), turtles (100%), dinosaurs (86%), and archosauromorphs (78%)."
In conclusion, Benton's study shows that there is a clear correlation between the evolutionary trees & the stratigraphic location that the fossils are found in. In other words, cladistics shows a sequence of fossil morphologies as per evolutionary expectations that mere chance cannot explain. In even easier language, you can make lineages out of older to younger fossils, & this is in overall terms the order of the day.
Mark
There are 10 kinds of people in this world; those that understand binary, & those that don't