More on getting it less wrong

From a graduate school admission essay by Bethany Keffala, BMC '07, posted with her permission ...

Physicist Enrico Fermi allegedly said that, “There are two possible outcomes: if the result confirms the hypothesis, then you’ve made a measurement. If the result is contrary to the hypothesis, then you’ve made a discovery.” It is this sentiment, this promise of discovery that draws me to academics in general, and theoretical linguistics and phonology in particular. I entered university, as I believe most do, with the naïve assumption that scholars, and especially scientists, search for ‘truth’. Through experience, trial, error, and reflection I have come to a revised understanding. We do not so much search for truth as for solutions that are “less wrong”.

“Getting it less wrong”, a concept introduced to me by former major adviser Paul Grobstein, is itself a less wrong approach to thinking about the process of scientific inquiry and discovery. One cannot prove a hypothesis to be correct, only that it is supported by data. Scientific inquiry is not a search for an answer to end all questions, but a process of asking questions that generate answers that themselves ideally create more questions, pushing our understanding to constantly, continuously, progressively evolve.

I found this vein of thought to be particularly useful in linguistics. There are arguably an infinite number of solutions to any given problem, however, some are less wrong than others. We know this as the ideal of the elegant solution – one that is simple, but contains powerful potential for explanation and prediction. There is always room for revision or refinement, for new solutions or theories to provide a less wrong understanding of data or phenomena.

Nice Fermi quote. And nice, of course, to have an idea developed in a biology/neurobiology context prove useful elsewhere. Interesting question, whether the "elegant solution" always turns out to be "less wrong than others".