I think it was Andy Tate who recommended I read Ian McEwan's Enduring Love while I worked on my undergraduate disseration on narrative physics. It was an excellent recommendation, as McEwan used his fiction to express the heart of what I wanted to say much more cogently than my more formal piece ever could.
His article in today's Guardian advocating the building of a sense of a 'scientific canon'—written on the 30th anniversary of Richard Dawkins' The Selfish Gene—continues to make that argument. Too much scientific education operates as though the latest discoveries exist devoid of their tradition, exempting itself from any sense that discoveries might be influenced by processes and losing along the way a rich understanding of how science has developed, and how valuable even false turns can be.
It's a shame that his final paragraphs diverge into a Dawkinsian attack on religious faith, an attack that he does not have space to fully articulate and which in its limited form seems an incongruous conclusion (if an apt tip of the hat to Dawkins). Nevertheless, it's worth a look.
I remember reading Dawkins and just getting mad at him for compromising reason by being so irrational about religion - strange that he should fall into his own trap.
Very much enjoy Alitster McGrath on Dawkins.