Ronald A. Knox – ‘The Viaduct Murder’

Just a short entry, ’cause I didn’t much care for this, but I did finish it, and also I took some photos of a viaduct at the weekend and it’d be a shame not to use one here. If you’ll excuse the pun, there’s just no body to this book. Characters all speak in the same way (generally starting off, ‘I say’), and it’s too self aware. Knox’s rules for writing detective fiction intrude at every step, even breaking explicitly into the narrative on occasion. Mostly this happens near the beginning of the book, then on one occasion near the end, as though he’s consulted another set of rules and remembered he’d forgotten about them, but couldn’t be bothered to go back and amend anything in between. It reads as though it was written in one draft, all its ingenuity being reserved for the plot. And I’m afraid I’m not one to give two stuffs about the plot ingenuity of a murder mystery. I like a Sunday evening in with a cocoa and a Poirot as much as the next tired person, but I never follow the bloody things.

Having said that, this bit is quite good:

‘...a man with a documentary hypothesis can defy the rudest assaults of common sense.’

‘How does one do that, exactly?’

‘You have to start out saying, “This document consists of three parts. One part is genuine, one part is spurious, the third part is faked evidence put in to make the spurious stuff look as if it was genuine!” Then, you see, you are on velvet. You reject altogether the parts of the document which you don’t like. Then you take the remaining part, and find that it still contains a certain sort of dross – evidence which still conflicts with your theory. That dross you purge away by calling it a deliberate fake. The watch says 4.54 – that is proof positive that, in the first place, the murder took place at 3.54, and, in the second place, the murderer tried to pretend it didn’t. You see the idea?’ (pp. 241-2)