« What to do with a revise and resubmit | Steven French | How To, Part 4 | Main | How to anonymise your paper | Beth Hannon | How To, Part 5 »



Feed You can follow this conversation by subscribing to the comment feed for this post.

Greg Frost-Arnold

Very interesting! I did not know the details of this history.

I just had one comment. You write:
"The realist strategy here is to argue that given the purely empirical/observational knowledge of the day, one could—with a bit of imagination—come to the gill slit prediction without any substantial theoretical ideas."

I wonder if this move (at least, if applied widely) could potentially hoist realists by their own petard: if lots of novel, interesting predictions can be created via extant purely observational knowledge plus a little imagination, then we often don't need theories (of unobservable processes) to generate novel predictions. In other words, if this type of explanation of novel predictions can be applied widely, then the more theoretical parts of theories potentially become 'idle wheels'/ 'inessential posits'/ whatever the selective realist likes to call the parts of the theory that are NOT responsible for the novel prediction.

So then the key question becomes: can the realist who uses this strategy for explaining Meckel's prediction of gill slits give a well-motivated, principled distinction between predictions made on the basis of phenomenological laws + imagination, vs. predictions which make indispensible use of 'pure'(?) theory? I don't ask this in the tone of thinking it's impossible, but I'm not 100% sure I see how it would go.

Here's why I say that: if the realist says "It's the reasoning the historical scientist ACTUALLY used," then the original claim is in trouble, since the historical scientists often actually used their false theories in one way or another (or often, we just can't really tell what the historical actors reasoning process was as precisely as needed to answer this question). On the other hand, if the realist says "It's the reasoning the historical scientist COULD'VE used," then the threat of lots of theoretical components becoming idle wheels looms (at least, if the historiographer is clever/ inventive enough).

(Final question: are any realists on record endorsing this move of 'phenomenological regularities + imagination --> successful predictions from people who hold false theories'?)

The comments to this entry are closed.