The general strategy is to tackle problems by adaptively translating the objective into intermediate goals and states— completing blocks— and basic actions— finding individual numbers. That strategy iterates, adapts, retreats when attempted solutions proves less promising than they appear. The methods of analysis that come naturally to us are oblique, and we do not use or enjoy direct, mechanical approaches. Anyone who buys a computer program to solve a Sudoku problem has missed the point of the game.

Because the process of achieving high-level objectives is necessarily iterative in this sense, the path to these objectives is bound to be oblique.

When you cannot measure something, said Lord Kelvin, ‘your knowledge is of a meager and unsatisfactory kind’.

In chapter 7 I described a spectrum of problems. At one end were those— like nought and crosses— best solved directly; at the other were those— the pursuit of happiness— best achieved obliquely. There is an analogous spectrum of decision-making styles, from direct to oblique.

The direct decision maker perceives a direct connection between intentions and outcomes; the oblique decision maker believes that the intention is neither necessary nor sufficient to secure the outcome. The direct problem solver reviews all possible outcomes; the oblique problem solver assembles all available information; the oblique decision maker recognises the limits of his or her knowledge. The direct decision maker maximises his or her objectives; the oblique decision maker is continuously adaptive. The direct problem solver can always find an explanation for his or her choices; the oblique problem solver sometimes just finds the right answer. The direct decision maker believes that order is the production of a directing mind; the oblique decision maker recognises that order often emerges spontaneously— no one fully grasps it. The direct problem solver insists on consistency, on always treating the same problem in the same way; the oblique problem solver never encounters exactly the same problem twice. The direct decision maker emphasises the importance of rationality of process; the oblique decision maker believes that decision making is inherently subjective and prefers to emphasise good judgment.

Good decision making is pragmatic and eclectic. Oblique approaches rely on a toolkit of models and narratives rather than any simple or single account. To fit the world into a single model or narrative fails to acknowledge the universality of uncertainty and complexity.

Good decision making is oblique because it is iterative and experimental: it constantly adapts as new information, of many kinds, becomes available. Much of that information comes from the process of decision making itself.