Greg left a very interesting comment on my last entry "The secret is to hide your sources".
Greg: It sounds reasonable that new discoveries emerge from sources that we may not be able to trace directly. However, there appears to be some tension with this idea and the underlying philosophy for academic writing where authors are expected to move from known scholarship to new inferences and discoveries. There is the impression that scholarship moves incrementally from the known to unmask the unknown through sound reasoning, observation, and experimentation. Perhaps, this is an ideal as well and that what happens is something between the two ideas.
Yes, indeed! Thank you for making this good point, Greg!
Academic paper writing is all about continuity and incremental knowledge building. Therefore, when you write the paper, it is necessary to cite all the relevant references, and state clearly what new things your present work is bringing into the field.
(Having said that, it is a fact that the Einstein paper in 1905 did not cite a single paper. I wonder if that was an acceptable mode of writing in that era.)
I remember a chat that I had with Sir Roger Penrose some years ago in Oxford. When I asked where his ideas came from, Penrose answered that it came from nowhere. Penrose does not usually look up what people have already done. He just does whatever he likes, and checks only afterwards whether any others have done similar things before. The order of things is actually opposite to what is normal in academic paper writing. I suspect that it is the case with quite a few people, especially those with originality.