Rumsfeld woz right?

well no, not about everything.... much of today's carnage in Iraq seems to fall into the eminently predictable. But having just read "The Black Swan" there might be rather more in the much mocked "unknown unknowns" - apparently a key part US DOD thinking rather than an example of another US politician misspeaking - than we normally give credit for.

The main message of the Black Swan is that turkeys would quite rationally vote for Christmas. After all, the average turkey day (subject to animal welfare standards) is pretty good - so a rationally forecasting turkey, basing predictions on past experience would quite reasonably think that December would be pretty much like November (assuming no turkey email with the US). But its the catastrophic last day which changes everything. So the message of the book is that unless we want to be suckers like the poor old turkey, we need to be much more aware that - to quote DR again - "stuff happens" - and that is all too often the stuff that renders all our forecasts meaningless and changes the course of history (and also may account for who you marry, and the very fact you exist at all - an occurrence with a very low probability.).

So apart from a lot of bile against social scientists and economists of all sorts, what are the useful lessons from assuming we know less about the future than we think (a case the book makes quite convincingly). The first message is to elevate preparedness over predictive capacity - predictive capacity is overdependent on normal distributions and assume that the future is like the past- and breaks down when it isn't. The second is to recognise that Black Swans can be good as well as bad -- so minimise exposure to bad Black Swans - and give yourself every opportunity to benefit from positive black swans - by for example taking chances to meet new people; hear new ideas, and put a small amount of money - money you are prepared to lose - invested in risky ventures with low downside but big upside potential.

The third recommendation is much more relevant for policy-making - that rather than overinvest in theoretical models trying to predict the world, adopt a more venture capital approach to policy-making. Don't assume you know what works in advance - better to try lots of things and then scale up the ideas that work and kill off those that don't. Substitute experience-based policy making for "evidence-based policy making". This accords with many of the prescriptions around for improving aid policy. Less Jeffrey Sachs and more Jeff Bezos.

The only problem is that a venture capital approach implies a number of failures - indeed that the majority of ideas might fail (and it also assumes we can understand and replicate the ideas that work) - and finding the Black Swan may not be enough to substitute for a bunch of dying white swans littering the street. But not clear that taxpayers or Ministers would be pleased to see that.

This is likely to be the last blog for a bit - finally back to work and to stay with a friend whose broadband doesn't work. Back to the dark ages. But latest news from the physio today is that am still on track - more, harder exercises and have just discovered I have been underutlizing the gym. I just need to make sure that Defra doesn't impede recovery - I can treat my personal "renewal" as a metaphor for the department.