A post from my Prose blogchain.


This concept is articulated in the excellent series on LessWrong titled ‘Babble and Prune’, a useful framework for understanding how the author views their process for generating knowledge, and thus writing in general. The following definitions for the two modes are as follows:

  1. Babble. Use a weak and local filter to randomly generate a lot of possibilities. Is the word the right part of speech? Does it lie in the same region of thingspace? Does it fit the context? This is analogous to the ‘artist’ in other models of writing.
  2. Prune. Use a strong and global filter to test for the best, or at least a satisfactory, choice. With this word in the blank, do I actually believe this sentence? Does the word have the right connotations? Does the whole thought read smoothly? This is using your inner ‘critic’.

It follows, then, that to produce more writing we must reduce how strongly we prune our thoughts. Interestingly, the author notes that reading and conversation outsource Babble to others; that is, that we rely on others to generate the raw inputs of our thoughts instead of creating unique ideas ourselves—this is actually what I’m doing right now! We are reading pre-pruned babble and are (hopefully) refining it even further. The author goes on to note that for those who read and listen much more than they speak (guilty), an overly-strict Prune filter is applied to their writing; when these people go to write something of their own, their minds don’t produce thoughts nearly as “coherent, witty or wise as their hyper-developed Prune filter is used to processing”.

Hence, my dilemma and an opportunity to break out of this trap. I recognised that if I attempted to write at the quality I was used to reading at, first time every time, my brain would promptly grind to a halt—like trying to brainstorm with a group that laughs at your suggestions. Of course, we cannot expect to completely relax all notion of Prune; editing and re-writing certainly have their place in any worthwhile writing. But we must develop high-quality Babble, this is the first step. You cannot edit a trickle of creativity.

How, then, can we build our Babble to the point that we begin to produce sufficiently useful writing? The authors suggestion is a good one: artificially replace the component we do not wish to train (in this case, the Prune). Think Scrabble or poetry—a medium where we know what a success looks like and we just need to find the right combination of words to fit. If this is working well, try relaxing your constraints a little. Try writing as if you were having a chat to a friend, or writing in a personal notebook. You will not be able to use everything you writing but there will be some raw material that you can start to work with.