On serendipity and knowledge
A great debate in the philosophy of knowledge (where knowledge is defined as “justified true beliefs”) is known as the “Gettier problems.” The debate is this: if you think you know something, and that something turns out to be true, but not for the reasons you thought … does it count as knowledge?
I tend to agree with the pragmatic view of Gettier problems. Basically, the only thing that matters is whether used knowledge is fruitful for the reasons that the knowledge was justified, true, and believed.
This has implications for serendipity. In serendipitious observations, the knowledge we generate was not necessarily justified or believed a priori. Only in retrospect does the “knowledge” become useful.