Saturday, 27 June 2009

Madame Bovary on a beam of light

I’ve been rereading ‘Madame Bovary’ and am struck, all over again, at how Flaubert lets us draw our own conclusions about Emma Bovary’s actions, without forcing any ‘morals’ down our throats.
Flaubert was one of a set of nineteenth century writers, along with others such as Zola, who were very influenced by the rise in science and who saw himself as a sort of human and/or social experimenter. Many other writers have also consciously seen themselves in this way. Brecht commented that he conducted experiments on audiences with his dramas and spoke of turning the theatre into a laboratory.

‘Copenhagen’ by Michael Frayn can be seen as an experiment in the way it repeats variations of the real-life conversation between Heisenberg and Bohr in 1941 in Nazi-occupied Copenhagen, in a repeated attempt to understand what really happened at this meeting. But, literature isn’t really the same as an experiment? In ‘Copenhagen’ the audience is presented with different versions of the events but we never get to understand what really happened, and why. But that’s not the aim of this particular experiment. What we’re left with is an understanding that it is fundamentally impossible to know what happened that night in Copenhagen, despite all the written records.
That understanding is a satisfactory result of this dramatic experiment. It doesn’t matter that this experiment cannot really happen, that it’s only acted out for us. It’s actually a perfect example of a thought experiment.

Thought experiments have a long pedigree in science. They’re best characterised as hypothetical experiments which we can imagine, but which we’re not able to perform. They allow us to set up a scenario and think through the repercussions. Einstein, influenced by the philosopher and physicist Ernst Mach, was a whiz at developing thought experiments. Some of the best ones (such as imagining a person riding on a beam of light) show how his theories developed from considering everyday objects such as clocks and rulers. They also provide the most accessible illustrations of the ramifications of his theories, without having to wade through all the maths.

As twentieth century physics became more esoteric and abstruse, physicists became more and more reliant on thought experiments to illustrate the implications of their work.
The most famous of these thought experiments; Schrodinger’s cat, was developed as a reductio ad absurdam by Schrodinger to criticise the bizarre, and clearly to him wrong headed, ‘standard’ interpretation of quantum physics. In this experiment a cat is locked in a box with a vial of poison. After a fixed amount of time, depending on a random process, the cat will either be killed by the poison, or not. But according to quantum physics, until we open the box, observe the cat, and therefore measure the outcome of the experiment; the cat is in a superposition of quantum states and is both dead and alive.

This experiment is not about to be performed in any laboratory soon. It doesn’t need to be. Thought experiments are a beautiful use of imagination in science. They allow the scientist to imagine ‘what if?’
This is precisely the question that fiction writers ask. Just because the results aren't scientifically provable doesn't mean they're not true.


  1. Does it matter that thought experiments are actually unimaginable? By the time my light-speed tram has travelled far enough for me to notice that the clock is running slow/standing still, we are so far apart that for the clock to be still visible it would have to be impossibly gigantic. I know that this does not obscure the point that is being made, but I worry that we arrive at it by what seems a sleight of hand - it is a bit like Platonic forms, which we approach by thinking of them as like real objects only apprehended directly by the intellect, so without all the deceptive trappings that the senses import - but without them, what do we have? And if we do not see the clock stand still, hasn't the wonder that beguiled us gone out of it?

  2. yes.. they are a bit like Platonic ideals.
    some of them were originally conceived as pretty off-the-wall stuff, like tracking single photons. But this has actually been done (in Paris in the 80s) so perhaps this sort of experiment can stretch our idea of what is reality>

  3. In Einstein's Clocks, Poincare's Maps, Peter Galison argues that Einstein's thought experiments on time grew directly from his work at the patent office, where he evaluated patents for techniques to synchronise railway clocks in remote cities using light or electrical signals. It was the very real, concrete problems posed by the inevitable time lag that led him to posit a relationship between time and signal velocity. His thought experiments were not so much a kind of platonic abstraction, as a convenient way of representing the problem to his readers. His theories are now re-realised in the relativistic time adjustments that have to be made when triangulating position using GPS satellites. Arguably, the issue in this instance is not about how impossible thought experiments relate to observable reality. Rather, it's about how what is observable is determined and perhaps even shaped by the technologies by which we observe.