First, Sean Carroll has a great post up on the "limits of science," that is almost exactly the subject I have been mulling over in the last few posts.
...what does science actually have the power to do?
I can think of one popular but very bad strategy for answering this question: first, attempt to distill the essence of “science” down to some punchy motto, and then ask what questions fall under the purview of that motto. At various points throughout history, popular mottos of choice might have been “the Baconian scientific method” or “logical positivism” or “Popperian falsificationism” or “methodological naturalism.” But this tactic always leads to trouble. Science is a messy human endeavor, notoriously hard to boil down to cut-and-dried procedures.
He's pointing out that impasses on "the limits of science" debates almost always hinge on the fact that the meaning of science gets slighted the more clear people try to be about it. That's not a rejection of clarity, it's rather a comment on an unfortunate state of affairs- it's hard to be precise about a dynamic process without also lopping off crucial parts of its dynamism.
Sean is exactly right- though "punchy motto" should be taken to mean a clear description of particular process. There is more than a motto: there is always, for any particular attempt to describe what science is, a large intellectual effort concerned with clearing the ground for a certain method: you identify a problem and criticize conventional approaches to it; you set up an internally consistent vocabulary; you make a set of arguments about why that vocabulary is the correct one to use. Eventually enough precision is reached to describe a process or set of processes that count as "science." But there always seem to be exceptions to each new account of what "science" is, which always seem to generate people newly convinced of the limits of science.
Science doesn’t do a bunch of experiments concerning colliding objects, and say “momentum was conserved in that collision, and in that one, and in that one,” and stop there. It does those experiments, and then it also proposes frameworks for understanding how the world works, and then it compares those theoretical frameworks to that experimental data, and — if the data and theories seem good enough — passes judgment. The judgments are necessarily tentative — one should always be open to the possibility of better theories or surprising new data — but are no less useful for that.
We tend to reduce science down to a method or a certain body of knowledge, and then leave it frozen up to those limits and suggest that whatever else is "beyond" science, which has doomed in their respective turns, verificationists, logical positivists, and falsificationists. Defenders of religious insight generally don't have a difficult time punching holes in "science" as outlined by these theories because you can (seemingly) always use language in a way that wasn't anticipated by people defining science, which always "proves" that there is some kind of truth that science isn't.
And if I read Chris correctly, he is using that strategy when he says:
Though these and other scientific models would later become refined and legitimated through measurement and observation, there is nothing evidentiary within hypothesizing itself. The creation of models in science [is] itself a poetic function.
and, a little downward:
Reason does not--cannot--produce the initial visionary flash that generates scientific ideas. It can only take these ideas as given and evaluate them methodologically. The ideas themselves come from elsewhere; from the imagination.
I'm curious what he thinks of the notion of an unanswered scientific question, which seems like an idea that straddles both sides of the "divide." For even to formulate a question is to make enough sense of your experience that you can recognize it as a problem, and fix its answer within certain limits. The P = NP problem for example, is obviously a scientific question, a question that is unanswered, and a question whose hypothesizing must occur within fixed limits. The "visionary flash" then, becomes quite corralled.
But more can be said. There are all kinds of things that count as visionary flashes (note the bit of conflation that crops up here- there are processes of thought other than the scientific method that nonetheless belong to a world describable by science- when we call something "not scientific" it should be taken to mean having less or a different kind of precision than our task requires, not that it involves a crossover into some other category of reality) but even in that initial casting about, it seems unfair to suggest that we get something different in kind from what we would obtain under the guidance of an explicitly rational process.
What exactly is going on, anyway? Well, one pulls apart concepts and recombines their pieces, shift contexts, adds or takes away assumptions, ratchets up or down the need for adherence to some principle, and every now and again looks back at the question, and at any available evidence to see how well such thoughts hold up. Where a "vision" fits in, I'm not sure.
If we didn't do this, whether via "vision" or rational process I'm not sure it would be possible to realize such a thing as a new model at all, and so to speak of it as though outside the limits of the scientific method deprives the method of a large part of its meaning, exactly the problem Sean Carrol was being careful to avoid.
A visionary flash seems to mean the sudden appearance of a new idea, together with an instant and firm grasp of its application. The closest analogy I can think of would be an explorer tasked with making a new map. She would already know beforehand the limits of existing maps, she would probably have already walked beyond those limits and perhaps have a vague idea of what the map is going to look like, and most importantly, she would already know how she was going to make the map of the new territory she encounters. What she can't control is what she sees. But this isn't quite right- what also needs to be said is, there is a chance of seeing landmasses wholly unlike anything that has been mapped before, so its not simply an issue of drawing up old images in new places.
But that's not a difficult problem- one can still draw. And a new discovery about reality can still be articulated in the same old mathematical language- its no problem that it describes wholly new rules.
So is it the fact that we can't anticipate what shape, or in what order, our new concepts might appear before us as we hypothesize (in that, an inability to anticipate tips us off to a process obviously outside our immediate control)? Or is it that we are constrained to speaking of them as though they were newly appearing in a "flash," when in truth there was an ongoing and finer struggle to "make sense" of something that precedes what we can deal with in empiricism or even language at all? If that's the case, it's not that things pop out of nowhere, rather there suddenly comes a moment when these ruminations produce something clear enough that we can "catch" it, deal with it in language, which makes the concept "new" in the sense of being for the first time as manageable as all the other ones we can comfortably deal with in language.
Even granting that the process was in some sense non-scientific, how could there be a point of crossover that consistently saw non-scientific concepts translating into ones with relevance and meaning essential to a scientific insight? What kind of mechanism could do that? We can refine and increase levels of precision- but that would mean there was something properly scientific about a concept from beginning to end. If we deny that there was, within a concept, something properly scientific from the beginning, we'd also have to deny that this was the case once it became clearly understood, and be left with the absurdity that one could manufacture "science" just by increasing the amount of clarity in the absence of content.
What is left, I think, is a body of experiences (in the most all-encompassing sense possible) that undergo a progressive escalation of confidence. If that is the case- this progressive escalation of confidence means that every bit of knowledge, however vague, is situated on the same track, even if only at the very beginning of it. If one wants to say not all segments on this track of escalation are equal and that only after a certain point do they become scientific, that's fair, but it totally leaves behind the question of why or how the "visionary" process was doing something fundamentally different from what science does.
It also affirms something that almost no one believes: that scientists and poets alike draw from a common reservoir of experience and that insights uncovered by poetry are always already promised to a corresponding scientific description.