Monday, 30 March 2009

Imagining accuracy

How important is accuracy in fiction? If you’re writing about something ‘real’ based on real information, experiences, or events do you have to stick to the facts?

If I spot a mistake in the use of science in fiction, it can throw me off course. I feel that the universe set up by the writer is flawed. If the writer can make one mistake, then perhaps others have been made too. Should I continue to believe in this universe?
And I’m more likely to be on the hunt for mistakes if I suspect that the author is using science for reasons other than telling a story.
For example, some authors appear to use science to bolster their authority. Ian McEwan does this in ‘Enduring Love’ with his use of quasi-medical papers to give a scientific ‘explanation’ for the way that one of the characters behaves. Others use science to provide pretty-sounding metaphors. Quantum physics and relativity seem to be particularly popular. The first line of ‘Cat’s Eye’ by Margaret Atwood is
‘Time is not a line but a dimension…’
After I read this oxymoron (a line does have a dimension), I very nearly didn’t read on.

And yet. A desire for accuracy can shade into pedantry. The narrator of ‘Cat’s Eye’ is an artist. She’s not likely to understand the finer points of general relativity, and more importantly, she doesn’t need to for the story to work. All she, and therefore the reader, needs to know is that her brother has become a physicist and is removed from the hum-drumness of daily life (This depiction of an egg-head scientist seems somewhat clichéd but that’s another matter).

Too close a reading of the text in an effort to check its accuracy can stop the reader from appreciating the multiple interpretations that are always possible. When I first read the following lines from the poem ‘Carnal Knowledge’ by Rebecca Elson;
‘Performed the calculus
Of the imaginary i…’
I took the ‘imaginary i’ to refer to the square root of minus 1, which is depicted as i in maths and is the foundation of all so-called imaginary numbers. It took several re-readings of the poem for me to realise that this imaginary i could also be a person, a body. (I don’t know why it took me so long, the whole poem is about bodies…)
My knowledge of maths perhaps led me to assume that there was only one meaning of this phrase, and this actually prevented me from getting a wider appreciation of what the poem could offer. I might also have made this assumption because I knew that Elson herself was an astronomer and much of her writing is about astronomy, and science.

So I think there is a danger of being too proprietorial about knowledge. It shouldn’t be off-limits. If writers make mistakes which the vast majority of their readers won’t spot, then what does it matter? They have at least stretched their language to encompass new ideas.

Monday, 23 March 2009

Petals and particles

Popular (and unpopular) science frequently relies on the use of metaphor in explanations. Metaphors have occasionally even been responsible for scientific discovery; in 1865 August Kekulé dreamt of a snake biting its own tale. He said this was the inspiration to his figuring out the structure of benzene.

The description of the expanding universe as a balloon being pumped up is ubiquitous in cosmology. But this ubiquity can be a problem; too often the metaphor ‘becomes’ the thing you are describing, and nothing is ever exactly the same as anything else. Any description of reality is limited in its accuracy by its reliance on words.

In quantum physics, light can either be thought of as particles or as waves, depending on how you observe it (the same is true of sub-atomic particles, i.e. they can equally well be thought of as sub-atomic waves). Thomas Young’s famous experiment at the beginning of the nineteenth century showed that light makes diffraction patterns when travelling through parallel slits. Diffraction is a property of waves. Conversely, Einstein’s early work showed that the photo-electric effect, in which light strikes a metal surface and liberates electrons, can only be explained if you treat light as a particle. So, clearly, our everyday concepts of ‘particles’ or ‘waves’, which are complementary, are inadequate to explain the true nature of light.

But physicists never let mere paradoxes stop them and this drawback was elevated to ‘the complementarity principle’ by Niels Bohr. He stated that something can be both one thing and its opposite, and that it didn’t matter, because physics can only be concerned with what you observe and not with the true underlying nature of reality. ‘There is nothing outside the experiment.’ (A nice counterpoint to Derrida’s ‘there is nothing outside the text’.) Different experiments show different aspects of reality, but there is no reason to suppose that you can have an experiment which shows all aspects.

The complementarity principle is an interesting riposte to those people who accuse scientists of having one-track minds, unable to see the subtleties inherent in reality. Keats claimed that Newton ‘unweaved the rainbow’ by explaining the physics behind this phenomenon. On the contrary, Newton deepens our perception of the rainbow through his description of light being diffracted by water droplets in the atmosphere.

Ezra Pound’s famous poem ‘In a Station of the Metro’ runs (in its entirety)
The apparition of these faces in the crowd;
Petals on a wet, black bough

The two images in this poem are so finely balanced that they mirror each other and it is never clear which is the metaphor and which is the reality. Dangerous for science, but prescient in its complementarity. Pound wrote this in 1913, when Bohr was developing his model of the atom.

Wednesday, 11 March 2009

Which comes first, style or content?

When I write, I worry about both style and content. I want my sentences to be pleasing aesthetically, but also meaningful.

Perhaps I am displaying my scientific roots by always making aesthetics play handmaiden to the characters and the story - the actual 'facts'. And yet – there is a powerful appreciation of aesthetics running through science, as well as maths. There is always the search for ‘an elegant solution’ to the problem in hand. What is meant by elegance here? I think it’s something to do with simplicity and conciseness, and perhaps novelty.

Aesthetically pleasing science makes apparently complex phenomena simple. Why are there so many different species of finches on the Galapagos – all with different habits? Darwin said that they’ve each evolved to fit a precise environmental niche.

It can relate disparate phenomena by revealing the underlying laws. Newton showed the movements of orbiting planets and falling apples can be explained by a universal force called gravity. Maxwell’s equations brought together all the different observations of changing electric and magnetic fields to show that each is a transformation of the other.

It leads to new areas and gives big bang for your bucks. (It’s no good having a good-looking theory if you can’t do much with it.) Einstein’s concept of light as a particle led to a whole new understanding of the structure of the atom.

It can decide between different ideas. Penzias and Wilson’s discovery that the thermal noise they detected at Bell laboratory was the remnant of an early stage in the evolution of the universe (and not pigeon droppings which was their initial assumption), ruled out the steady state model in favour of the big bang one.

It may even be visually pleasing in some way. Crick and Watson’s analysis of DNA revealing its double helix structure, has created an image now embedded in our collective consciousness.

Behind many of these aspects of aesthetics lies simplicity. Simplicity is a powerful driver in creating science. Occam’s razor says that we should not ‘multiply entities’ unnecessarily; so if you’re fitting a mathematical model to your data, you choose the one with the fewest parameters. And a new scientific theory should have as few arbitrary factors as possible. But in judging competing scientific theories, it’s not always obvious which one best obeys Occam’s razor.

For example, the Copenhagen interpretation of quantum mechanics says that a wave function is attached to each possible outcome of an event. Once a particular outcome is measured, the wave functions corresponding to all the other outcomes collapse. Until that happens everything related to the event is in a sort of probability ‘fuzz’ (for example, the cat in the box is both dead and alive, until you open the box and discover its fate). But what exactly is a measurement? By definition, it has to be a non-quantum event, otherwise you just get more fuzz. So, the interpretation fundamentally limits what quantum mechanics can describe, by saying there always needs to be something outside its description!

The Many Worlds Interpretation avoids this in-built limitation, but only at the expense of having to start up a new universe each time an event happens. This seems to be multiplying entities to an extreme degree; and quite wasteful.

So which is the ‘simpler’ explanation of quantum reality? They clearly both have their problems, and not ones which can easily be resolved by examining their aesthetics.

Thursday, 5 March 2009

The ghosts in the human genome machine

The purpose of the Human Genome Project was to map all the individual bases that make up our DNA. There are four types of these bases, adenine, thymine, cytosine, and guanine, and they combine to form genes.

Now that the project has achieved its main goal and finished the sequencing, the next step is to identify the actual genes as distinct from the majority of the material, the so-called ‘junk DNA’. (Only a tiny proportion of the overall DNA actually consists of genes.)

Humans share 99.9% of their genes, and so the information discovered through the HGP is relevant to all of us. As we have about 20,000-25,000 genes, only a handful are unique to any one of us. But whose DNA was actually sequenced? The project used samples from anonymous donors. Neither the scientists nor the donors know whose samples actually ended up being sequenced, but it is clear that more than one person’s was used, i.e. the information we have is an amalgamation from different donors. Because the ‘map’ created by the HGP is actually linear – a sequence of letters corresponding to the order of bases, at any one point in the sequence, the information we have corresponds to just one unknown donor, but we don’t know who.

This deliberate uncertainty interests me. Scientists (whatever their discipline) spend so much time battling uncertainty, trying to quantify or eliminate it from their work. Much of this uncertainty is caused by random fluctuations or systematic biases in what they are trying to measure. Both need to be understood and accounted for, if you’re trying to make sense of your external world. And more fundamental uncertainties exist in quantum physics, which are not simply due to errors or limitations in the way that we measure things.

So it seems counter-intuitive to increase the amount of uncertainty in this major experiment. But clearly it has several purposes. Uncertainty about knowledge of the donors can protect them from the consequences of having their genome sequenced. For example, subsequent identification of genes relating to disease can’t be attributed back to a particular donor. Also, by banishing information on the particular donors, the experiment is able to interest everyone. It encourages us all to feel that that map has a direct relevance to each and every one of us. As a result, the project is allowed to gain a certain amount of authority.

But you could just as easily say that the project is relevant to no one. What does it mean to sequence a genome of a person that doesn’t actually exist?

This partial information is an excellent example of synecdoche, a type of metaphor in which part of a thing is used to stand in for its entirety. We don’t yet have DNA sequences for all humans, and we don’t yet know which parts of the sequence we do have are shared by all humans. So ‘The Human Genome Project’ is a misnomer.

Perhaps I shouldn’t be too harsh. In practice, it’s actually impossible to communicate without using synecdoche. Fiction writers know that they can’t get across the entirety of their fictional characters. They write about aspects of these characters, and the reader uses these a bit like dried milk, to reconstitute them and make up their own pictures.

Or you can say ‘there is nothing outside the text’, and therefore the human genome is just a sequence of bases, and we are free to give it as much or as little significance as we wish.