Returning to The Black Swan - one of the big ideas books that is worth mulling over more than once - I had a shocking thought. Is Nassim inconsistent? Might he be, err, wrong? My thought went something like this.

He opposes Platonicity, roughly making up theories about the world and taking them to be true. He also ridicules the use of the Gaussian and Gaussian tools to identify averages and ignore exceptions, with the qualification that the Gaussian works well in Mediocristan, those cases in which an exception does not matter.

But surely, I thought, Nassim relies on Platonicity to come up with his insights on human nature, both in assuming that there is such a thing as human nature and then in describing what it is. Further, the empirical psychology he relies on to describe human nature deploys Gaussian techniques. Examples include his use of Dan Gilbert's concept of 'affective forecasting', in which we fail to learn from the past and mispredict the future in so doing. What is empirical psychology in love with if not the very homme moyen moral that Nassim mocks.

Now, in response to this second point, he is aware of the accusation, and stresses that the insights on human nature derived from empirical psychology are 'robust to the mistake of using the bell curve' because 'these studies generally elicit a yes/no type of result. No single observation, by itself, can disrupt their overall findings.' (p245)

One should, I suspect, challenge the artificial nature of the scenarios that the empirical psychologist constructs in order to elicit these tidy yes/no results: they must be artificial in order to be so clear cut, and yet from them assumptions about a presumed human nature are drawn. But leave that to one side. For a tighter tension arises directly in relation to Nassim's argument, and his encouragement to us to deviate from the human average when it comes to assessing uncertainty by becoming more aware of black swans. If it's true we live in Mediocristan now, in relation to our handling of the unknowns of life, he effectively wants his readers to move to Extremistan, where we, like him, would not blindly follow 'human nature.'

If that is even a viable project, which presumably he believes it must be, else why write the book, then that challenges the first point, the Platonicity that assumes there is such a thing as human nature and, moreover, that it is bad at handling black swans.

To put it another way, he would presumably predict that most people will continue to act ignorantly when it comes to black swans; it seems very likely. Though if the small chance came about that people didn't and instead changed and wised up, that would make a huge difference to the world. So here's the rub: that would be to go against one of his central pieces of wisdom: 'Be fooled in small matters, not in the large... Do not listen to predictors in social science (they are mere entertainers).' (p203)

In short, does his desire to write the book rest on the assumption that the good black swan of us becoming smarter about black swans is a black swan that can be discounted? And if so, does that mean he's not following his own advice, on account of being fooled by the Platonicity of 'human nature'? Or have I got my black swans in a twist? Or has half the planet already raised this objection only to have it refuted?