[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Subject Index][Author Index]

Re: Fw: Popper and Palaeontology (was: Re: "running" elephants - locomotary analoges)




On Tuesday, April 22, 2003, at 08:23 PM, mjmurphy wrote:

I think one thing useful to note here is that Popper ideas concerning verification/falsifiability were developed largely with an eye on physics. Maybe because physics is highly mathematised, Popper's idea's tend to be very theory centric (as opposed to observation-centric). There is a technical sense in which what you choose as your "data" is arbitrary, and the whole notion of "gathering facts" in the absence of a guiding theory is derided as irrelevant.

I'll stick my neck out here and say that it is. And not just irrelevant, but impossible. Even naming something is offering an hypothesis about it. For example, say we want to measure the length of the femur of _Archaeopteryx_ - we obviously have to name the thing we are measuring - and by doing this we are hypothesising that the femur is different from other bones, and we also have to have theories about space, length, measurement, and so on. Now say we want to do something a little more complex - compare the length of the femur in _Archaeopteryx_ and _Microraptor_, not only do we need the theories above, but we also have to have a hypothesis of homology to be able to identify the bone. Already we are up to our eyeballs in deeply theoretical concerns, just comparing the length of two bones.


The idea, expressed by one of the Leakey's, that went something like "give me a good data set over a theory any day" is quite alien to Popper.

And for good reason, I'd say.

Further, the whole idea of inductive reasoning--reasoning from specific facts via some inductive method like "concomittant variation" (hope thats the right spelling) to generalities--is dismissed. It is invalid (by deductive logic) to conclude from the fact that the sun has risen every day for the past billion years that it will rise tomorow. In roughly the same way, it would be invalid (not just speculative) to conclude from the fact that many small theropods were feathered that all theropods were feathered at some point in their life cycle (or something like that). In a strict sense, for Popper a large portion of the argumentation that goes on on this list would have to be invalid.

Logically invalid yes. Popper was fighting against a _formalisation_ of inductive inference. However, to hypothesise on the basis of many feathered small theropods that feathers were present on all theropods (at some stage of their life) is perfectly popperian. Popper was not against making generalisations - on the contrary, he was for them - he was against the construction of a formal methodology for induction.


Phylogenetic bracketing can look like induction, but because it works just as well with two markers as with a million, it is more like an automated popperian generalisation machine :). We only needed _Sinosauropteryx_ and modern birds to believe that maniraptorans were feathered - the other feathered theropods just filled in the picture, and didn't falsify the hypothesis.

Now, Popper is somewhat ambiguous with regards to inductive reasoning. There are times when he literally sounds as though this kind of reasoning doesn't exist and if it does exist people should stop doing it immediately.

Yes, some times he says very clearly that there is no such thing, but if that is so it isn't clear what he is arguing against (actually the issue is a lot more involved than that - Popper didn't often make really obvious mistakes).


There are also times when he compares inductive reasoning to "rules of thumb"--I think he actually compares inductive reasoning to the way people reason about gardening (best time to plant petunias, and so on)--which, while they can be used to generate hypotheses, are irrelevant to the testing of hypotheses. However, I think in a discipline like paleontology its more likely that an argument will be taken as worthy of testing just because it takes a form that is recognizably inductive.

Popper was unashamedly normative - he thought his method represented the way science _should_ be done. So it could be argued that maybe palaeontology would be better off if it dropped any pretence at inductive method.


Cheers,

The shapes of things are dumb.
-L. Wittgenstein

(Wittgenstein once threatened Popper with a fire poker during a debate...)



John Conway, Palaeoartist & Protophilosopher

"Do I contradict myself? Very well, then I contradict myself, I am large, I contain multitudes." - Walt Whitman

Palaeoart: http://homepage.mac.com/john_conway/_palaeoart/
Palaeontology & philosophy discussion forum: http://clouds.proboards16.com/