I came across the talk origins site on accident as I was doing some browsing about the subject of origins (particularly intermediate or transitional forms). As a complete neophyte to paleontology I had a couple of questions and your answers seemed to be some of the most cohesive that I found. How are partial fossil remains (I am supposing that most fossil finds are at least partially incomplete) classified and dated? Is a date deduced first then a classification or the other way around? And is most of the dating done by strata location or radioisotope dating of surrounding material? Sorry for my ignorance I am a professor of languages and this is new but interesting material for me.
I think that the example of “conodonts” may help illuminate all your questions.
First, check the Wikipedia page on conodonts.
And second, take a look here.
Conodonts were first described in 1856 on the basis of phosphatic tooth-like fossils. This sort of fossil was the only type of fossil remains known until 1983, when the same tooth-like fossils were found with soft-tissue preservation showing the organism that bore them. So for over a century, this group was solely known from highly fragmentary fossil evidence. There was plenty of conjecture, but the actual affinity of conodonts as members of Phylum Chordata was only worked out relatively recently.
The easily discovered conodont fossils are in the Cambrian or pre-Cambrian, and are commonly found in strata dating to the late Triassic period, so conodonts were around for somewhere near 350 MY. As fossils go, conodonts are relatively common. The group was diverse, too, and those two things meant that they were nearly ideal as index fossils. Index fossils are those fossils that are reliably found only in a limited range of strata. Because of this, they are used for assigning relative dates to those strata based on standard geological heuristics like superposition (a rock layer above another rock layer is generally a younger rock layer).
Conodont fossils are extensively used in petroleum geology for locating oil deposits. Besides their use there as index fossils, they also give evidence of paleoclimates due to temperature-dependent changes in their phosphatic composition.
Now, recall that all of this utility came well before anyone had any good idea of what a conodont might actually look like. In the absence of complete evidence, conodonts were assigned their own phylum. Classification within Conodonta proceeded entirely based upon morphology and chemistry of the tooth-like fossils. Despite knowing next to nothing about the organisms themselves, geologists and paleontologists found conodont fossils highly useful for assigning relative dates to other fossils found in association with them.
In this case, most of the dating associated with conodonts was relative, not absolute, and classification played a very minor role, as it was based on known-fragmentary information.
I can’t really say what the typical approach might be toward fragmentary fossil evidence (since I’m not a paleontologist), but the conodont story indicates that paleontology has successfully and fruitfully dealt with that situation before.
I think that you should have a look at the TalkOrgins “Fossil Hominids” sub-page.
There’s quite a bit of discussion there about the partial skeletons of various hominids and what we can learn from them.