The drunkard's walk and the red wheelbarrow

so much depends

upon

a red wheel

barrow

glazed with rain 

water

beside the white

chickens

...that's if you take William Carlos Williams' word for it. And, on the whole, we do. So much depends... but why? The spare elegance of this now classic poem relies not just on its deft simplicity and nicely audacious line breaks, but on our capacity for over-signification. In truth, so much depends on you, dear reader.

In my last blog post, I was looking at the notion of an optimism bias and how it might influence reading behaviour as well as a general outlook on life. But we aren't just overly optimistic by nature. In his recent book 'The Believing Brain' (2011), Michael Shermer takes a fascinating look at a more fundamental cognitive bias, a tendency he calls 'patternicity'. Patternicity is 'the tendency to find meaningful patterns in both meaningful and meaningless data'. Shermer describes the brain as a 'belief engine' and argues that patternicity is also accompanied by agenticity, 'the tendency to infuse patterns with meanings, intentions and agency': in other words, we see patterns everywhere and we assume that they aren't random.

I first came across the notion of patternicity in a Cambridge pub (where else) late at night when someone far more intelligent and eloquent about these matters lent me a copy of Leonard Mlodinow's 'The Drunkard's Walk'. It's a fascinating, statistical and mathematical examination of how we miscalculate probabilities in daily life and often underestimate the role of chance. Amongst other things, such miscalculations can influence gambling behaviour, social and political decision making and economic decisions. To quote Tim Radford in The Guardian, Mlodinow engagingly demonstrates that 'almost everything that happens in life is contingent upon a series of unconscious gambles: of turnings taken, of chance encounters and unconsidered choices - in short, the drunkard's walk of the title.'

Why do such tendencies persist? In 'The Believing Brain', Shermer makes a compelling if basic case for the role of natural selection. Imagine a hominid, three million years ago, walking along the savanna and hearing a rustle in the grass. Is it just the wind or is it a dangerous predator? Assuming the latter when it's really just the wind would be a Type 1 statistical error or false positive - a non-existent pattern. But in this case, the Type 1 error has no negative consequences. Assume the noise is nothing to worry about, however, when a predator is lurking in the bushes and you're prehistoric mincemeat, no longer a member of the homonid gene pool. False positives are less harmful than false negatives.  Thus, as Shermer suggests 'there was a natural selection for the cognitive process of assuming that all patterns are real and that all patternicities represent real and important phenomena.'

Amongst other things, we use these patternistic tendencies for facial recognition and for mimicry, an essential aspect of learning. Thus, it's not surprising that Shermer goes on to posit a connection between mirror neurons and agenticity. Our capacity for Theory of Mind makes us more likely to assume patterns (particularly with regard to human behaviour) are meaningful. Shermer also believes that dopamine - a chemical transmitter substance - is most closely related to neural correlates of belief. Put simply, dopamine assists learning behaviour on a neural level, enhancing the transmitting ability of neurons at a given time and thus increasing synaptic connections in response to a perceived pattern. Interestingly, experimental research by Brugger and Mohr showed that people with high levels of dopamine were more likely to find significance in coincidences and patterns where no real patterns existed in an experimental context.

Creativity, of course, is a kind of discriminate patternicity (as opposed to some forms of psychosis, which can be characterised by indisriminate patternicity). But reading poetry can be an unconscious exercise in patternicity too. Don Paterson has written about the 'contract' the reader enters into when they know they are reading a poem: in simple terms, we assume that the words in the poem have connotative as well as denotative meaning. We assume that no image is arbitrary. We assume that William Carlos Williams couldn't just as easily have chosen a blue wheelbarrow and some slightly off-colour ducks (to your left, a group of said ducks in North Yorkshire this Christmas). Reading a poem is an over-signifying enterprise. As Paterson puts it in 'The Lyric Principle':

'Humans – no doubt in an act of vital compensation for their habit of hypercategorization, and the fragmented perception it brings - will connect any two unrelated things you care to throw at them...Poets take advantage of this by prompting or initiating just such a game of connection, presenting the reader with elements that, on a casual glance, seem only indirectly related - or not related at all.'
Of course, this means readers may see patterns or connections the poet did not intend - most writers will have read an over-analytic response to one of their poems at some point, or been accosted after a reading and commended on a meaning they never dreamt of themselves. Billy Collins pokes gentle fun at this in his poem 'Litany', a piece that suggests the arbitrary nature of some imagery:

You are the bread and the knife,
the crystal goblet and the wine.
You are the dew on the morning grass
and the burning wheel of the sun.
You are the white apron of the baker,
and the marsh birds suddenly in flight.

However, you are not the wind in the orchard,
the plums on the counter,
or the house of cards.
And you are certainly not the pine-scented air.
There is just no way that you are the pine-scented air...


(I strongly recommend listening to the whole poem).

Poets can exploit the patternistic tendencies of readers too. To quote Don Paterson again: 'when the poem is too discontinuous, and insufficient context has been provided to link its elements, the reader compensates by sending their connecting faculty into overdrive, and starts finding connections and significances for which they were given absolutely no cue.' (I did an experiment with small focus groups in 2007, which involved a blind reading of transcribed schizophrenic speech alongside certain poems with their lineation taken out to see if participants could tell the difference - the results were interesting, but would warrant a blog post all of their own!).

To suggest that reading poetry involves our innate patternistic bias isn't to diminish poetry's power. Our tendency towards patternicity is part of what gives us continuing pleasure in reading poems, full of rich connections in sound, sense and structure for us to enjoy. I'll leave you with a poem by Paul Muldoon, that master of hyperconnection. William Logan has described his recent poems as driven by 'gusts of rhymes' and 'whirlpools of puns': 'a phrase in Muldoon may scuttle in and out of a poem, meaning something different each time'. The effect in his later collections can be bewildering on a first read; patternicity accelerated, perhaps. This poem is taken from his 1987 collection 'Meeting the British' and I've always thought it has something interesting to say about the nature of connection-making itself.

Something Else

When your lobster was lifted out of the tank
to be weighed
I thought of woad,
of madders, of fugitive, indigo inks,

of how Nerval
was given to promenade
a lobster on a gossamer thread,
how, when a decent interval 

had passed
(son front rouge encor du baiser de la reine)
and his hopes of Adrienne

proved false,
he hanged himself from a lamp-post
with a length of chain, which made me think

of something else, then something else again.