Thomas Kuhn defines research as “a strenuous and devoted attempt to force nature into the conceptual boxes supplied by professional education.” To me, it seems a fairly reasonable, if somewhat cynical, way of looking at it.
There are two implications within the statement that strike me as interesting. The first is the assertion of nature. The idea that it is something out there, to which research is applied, is more empiricist than I expected from a book that Tristan recommended to me (that said, I am only halfway done, and all manner of complexities could yet emerge). All that said, here are a number of different ways in which you could interpret ‘nature’ in the above sentence.
Most obviously, you could take it to mean the external world of atoms and galaxies and ocelots. Naturally, ‘atom,’ ‘galaxy,’ or ‘ocelot’ is just a description, but it is not unreasonable to assert that there is something out there that can be reasonably assigned a term. There is a problem akin to the naming of constellations – it is arbitrary which stars you include in which grouping – but any possible set of constellations is at least a valid description of the orientation of stars in the sky. You might group them by proximity and geometric patterns, or you might group them according to their spectral profiles or any of myriad characteristics, but it should be possible to go back from whatever model is created to either re-create or at least recognize the phenomenon being described.
Another possible meaning for ‘nature’ is just experience. When we look at the stars (or anything else), our brains are performing a massive amount of signal processing. What you see is not, in many important ways, an accurate reflection of what is actually there. Details that evolution has determined to be unimportant are given little or no attention, whereas ones that natural selection has marked out as important are highlighted. This is the inevitable product of how genes that do a good job of sorting important data out from trivial data will tend to find themselves copied more often than those that do the same task badly. Very bright things are dimmed beside darker ones, and vice versa. Learning to undo a lot of this trickery is an important step towards becoming a good photographer. If we take ‘nature’ in this way, the object of our research is our own experiences of a natural world, rather than that world itself.
The second is the implication that we could somehow deal with nature in a more meaningful or comprehensive way. This is an assertion that comes into conflict with limitations in human cognitive power, and the time that can be applied to problems. We can, for instance, only really think in three spatial dimensions. We can only remember so much, and only grasp connections of certain types and complexities between phenomena and ideas. As such, the choice is not between modelling through categorization and some some of ideal holistic understanding of the universe; it is between modelling through categorization and some alternative form of modelling that is still bounded by human cognitive limits.
To me, the evident success of category-based modelling (as manifest most obviously in technology) demonstrates that it is clearly the world comprehension system to beat. Believing that light is a quantum phenomenon as described by certain equations is demonstrably better than believing it is the result of some kind of active broadcasting from the eyes. The most obvious way to show that is that you can build fibre optic cables and fancy lenses and optical disc drives on the basis of the former conception, but not the latter. One day, we will probably have an even better understanding of light, as demonstrated by a greater ability to do things that we want to do using it. Research, as Kuhn defines it above, is an essential activity and a worthwhile application of time and effort. While there is every reason to question and refine our methods, they are not worthy of outright denigration, as I am sure he would agree.