Thursday, 5 March 2009

Causality, or consistency of boundary conditions?

On Radio 4 this morning is an amazing programme exploring some really deep and technical issues about quantum theory and the nature of reality in physics.

Amazingly in that they are brave enough to broadcast it at a time more associated with The Archers. Amazing in that the level of discussion is one that would challenge most physics and philosophy graduates, and they let it loose on a mainstream audience. I'm impressed.

But then I read Robert Peston's latest post about China, and it became even more resonant. That old question is asked: who or what is to blame?

In particular, was the financial crisis caused by excessive borrowing by Westerners or excessive lending by China?

The thing is, of course, that neither of those things could happen without the other. That makes it impossible to really debate whether saving caused lending or vice versa, let alone which one caused the financial crisis.

And that's where I am reminded of the methods of quantum physics and how we solve the Schrödinger wavefunction. In the mathematics of Schrödinger (at a first level), we don't think about causality; indeed we assume time is reversible. We just set certain conditions and we ask "what wavefunction would be consistent with these conditions"? In a sense, the universe simply obeys the conditions and never mind how it got there.

In this case, the conditions are: borrowing is equal to lending; a credit crisis emerged in 2007 and developed through 2008; China is very competitive in exporting consumer goods. What set of capital flows are consistent with these conditions?

The answer falls out quite easily. It's much easier to find a set of data consistent with the conditions, than to assign causality and blame. The interpretation, you might say, is overdetermined by the boundary conditions.

Now I've mentioned before the dangers of making weak analogies, and I don't like to leap from one science to another without rigorous demonstration that the results apply. So I don't make any assertions about this. It just makes me think, that's all.

No comments: