Benefits of randomness
Tue, Jan 9, 2018
The Greeks at Delphi used the ravings of a priestess possibly high on methane and ethanol to make famously ambiguous predictions. Ancient Rome used sacred chickens and the innards of slaughtered animals. Other historical methods of divination include:
- pyromancy (by fire)
- ophiomancy (by snakes)
- pegomancy (by spring water)
- anemoscopy (by wind)
- daphnomancy (by burning laurel wreaths)
- hippomancy (by horses)
- tiromancy (by cheese)
Way out East, Ancient China used the patterns made by throwing yarrow stalks to the ground as a starting point for cleromancy using the I Ching.
On the reasonable assumption none of these work in the sense of actually predicting the future, why did people use them, in some cases for many centuries? I think it’s because all these practices rely on seeding thought processes with some random element – and that inclusion of randomness can be beneficial.
More recently, Brian Eno and Peter Schmidt have created decks of cards printed with gnomic advice as a means of overcoming creative blocks. Their Oblique Strategies include suggestions such as:
- Use an old idea
- State the problem in words as clearly as possible
- Only one element of each kind
- What would your closest friend do?
- What to increase? What to reduce?
- Are there sections? Consider transitions
- Try faking it!
- Honour thy error as a hidden intention
- Ask your body
- Work at a different speed
Eno and Schmidt are clear that their cards are intended to jerk people out of thought tramlines. And that’s what ancient divination methods probably did too: they gave licence to express thoughts, and they prompted thinking and action in new and unexepected directions.
The usefulness of random pertubations
Evolutionary biology makes use of the notion of an Evolutionary Stable Strategy, borrowing from economics and the Nash Equilibrium (that’s John Nash of ‘Beautiful Mind’ fame).
An ESS is a fitness solution that has arrived at a (local) fitness maximum. It may be useful to imagine a landscape of hills dotted about a plain, where any ESS would be represented by the summit of one of those hilla. A relatively small departure from the ESS would be the equivalent of heading downhill; that is, the solution will result in decreased fitness – ie, decreased reproductive success – and so won’t spread through the population.
But in the whole landscape of solutions there may be higher summits - more effective fitness solutions. It’s just that any mutation pushing an organism in the direction of a higher hilltop will be instantly penalised by first having to head downwards.
Now if the organism were forced downhill by something it would have to start again, in a random direction, trying to find higher ground (it’s very difficult to escape teleological language but it could be done). In doing that, it might head off in a direction that took it towards a better solution.
In fact, some neural networks have a little randomness thrown in exactly to get round this problem of settling on a local but non-optimal solution. And that’s what Eno was doing in his Oblique Strategies.
The I Ching, Again
I first came across the I Ching when I was at school and read in The Glass Bead Game about Joseph Knecht, Magister Ludi, casting yarrow stalks . The text of the I Ching is of an age with Homer, probably older. The part used for divination lists 64 patterns and adds commetary for each. The commentary has been expanded by scholars and Jung said of it: ‘Even to the most biased eye, it is obvious that this book represents one long admonition to careful scrutiny of one’s own character, attitude, and motive’.
Jung was a bit of a mystic himself but I’m appealing for the use of constructive randomness in problem-solving and I’m quite intrigued by the use of a handy divination tradition, with supporting mobile phone apps and a long history of commentary to help tease out thoughts and directions you were half-suspecting might work but weren’t quite brave enough to implement.
So I think I’m going to incorporate the I Ching into my decision-making, to throw in that little bit of left field thinking that might make all the difference.