The Bucket or the Searchlight?

“The Bucket and the Searchlight: Two Theories of Knowledge” from Karl Popper presents two views of knowledge. We advance knowledge in research and disseminate knowledge teaching. How we undertake these tasks depends on which view we take.

The bucket fallacy underlies many mistakes. Notably in universities.

I raised Popper’s question in two papers published in 2001, illustrated nicely with cartoons by my daughter, Sarah. The cartoon data to be scooped up or interrogated, according to one’s viewpoint, was from DNA microarrays. Today I’d think more of genomics, perhaps GWAS, while the microarray example is not entirely outdated. Think, perhaps, transcriptomics and RNA-seq.

Bucket
The Bucket. Data are scooped up, at random.
The race continues – the race to acquire a bigger bucket than anyone else, one large enough for “big data”. The bigger the bucket, the more expensive, and the more attractive the bucket theory becomes to the clowns and crooks who hold that research output is not knowledge, but grant income. Then there is factory science, as described by Sydney Brenner.

“So we now have a culture which is based on everything must be high-throughput,” Brenner continued. “I like to call it low-input, high-throughput, no-output biology.

I suppose high-throughput biology is roughly equivalent to equipping the bucket-brigade with a hosepipe. Or water-cannon.

And the race continues, in teaching, to fill students’ empty buckets as quickly and completely as possible, while obsessively trying to gauge how much they’ve retained. The examination as dipstick.

What a waste of time. And energy. And money. And human potential.

While all the while the searchlight is there for us to use, to share, and to pass on.

Searchlight
The Searchlight. Data are examined to see how they compare with the prediction of an hypothesis. The prediction is on the clipboard.

Popper, K.R. The Bucket and the Searchlight: Two Theories of Knowledge. Appendix to “Objective Knowledge. An Evolutionary Approach”. Oxford University Press, Oxford. 1972.

Allen, J.F. (2001) Bioinformatics and discovery: induction beckons again. Bioessays 23: 104-107.

Allen, J.F. (2001) In silico veritas – Data-mining and automated discovery: the truth is in there. EMBO Reports 2: 542-544.

Leave a Reply