Goal Control
What links Trump's Iran war, the contradictions of energy policy and a donation to the University of Cambridge?
Donald Trumps’s stratagem of refusing to set out any clearly defined aims for his war on Iran and then preparing to declare victory on the basis of aims he invented later is comically obvious. His claim that the assassination of a layer of Iran’s leadership amounts to ‘regime change’ when all that has happened is a change in personnel fools no one, not even Trump’s acolytes who profess to believe it. His claim that landing some very large bombs and missiles on Iran’s nuclear facilities amounts to ending Iran’s capability to develop nuclear weapons is also nonsense. He has failed to destroy either the Iranian government’s desire to develop nuclear weapons or the expertise at its disposal, domestically and internationally, necessary to achieve that end. And his assertion that Iran is defeated because he has destroyed a large proportion of its weaponry ignores the fact that it retains enough drones and missiles, and enough capacity to make or import more of them, to close the Strait of Hormuz and so to raise petrol, food and medicine prices in the USA and around the world.
Trump’s technique is not new. I first became aware of it nearly twenty years ago when I served, briefly, as the Liberal Democrats’ shadow energy minister. Energy policy notoriously involves competing goals, including promoting GDP growth, reducing poverty, eliminating greenhouse gas emissions, keeping down government expenditure and taxation, reducing national security vulnerabilities and restraining inflation, several of which are incompatible with some or all of the others. But no government says precisely what it intends to achieve on each goal or ever admits that trade-offs exist between the different goals. They simply claim credit for whichever metric seems to be improving and disregard any that are deteriorating. Opposition parties make the situation worse by complaining about one goal at a time, usually the one most currently pressing for the public according to opinion polls, and ignoring all the others.
I call this technique ‘goal control’. Its essence is to refuse to specify the goals of a policy so that, whatever happens, the policymaker will be able to identify something that counts as success. Refusal to specify is sometimes done retrospectively, which is what Trump is doing about Iran, insisting, for example, that he never implied that success in Iran necessarily included Iran making a deal with the USA.
The most important example of retrospective goal control is Brexit, whose advocates now deny that they had any goals other than the mere fact of constitutional separation. No sign exists of lower prices for “food, clothing and footwear” as promised by Jacob Rees Mogg, for example, or the NHS improvements promised on the side of the Brexit bus. So Brexiters have quietly dropped those goals and now claim that Brexit is a success simply because ‘Britain has successfully left the EU’.
Goal control has a wider significance. It illustrates why politics matters for public policy and why public policy is not simply a technical matter for experts. All engineering designers know that designing a successful new product is very difficult, or even impossible, if the client refuses to specify precisely what they want. Even more importantly, clients also duck setting out precisely what they want to avoid, for example the extent to which, if at all, they want to take into account ethical and environmental considerations. But this is a known problem, and part of the training of engineering designers is how to elicit from clients enough precision about goals and constraints to enable a project to go forward and also to know when a project is so lacking in specification that it is fundamentally a bad idea to take it any further. The same applies to other design professions, such as architecture and, as I tried to show in my book Law as Engineering, law, especially for commercial and private client lawyers who design contracts, companies, trusts and wills.
In the design of public policy, resistance to specifying goals and constraints is even stronger than in the private sector. And it is far more difficult for civil servants and policy advisers to tell their reluctant ministerial clients that designing anything whose objectives are completely unclear is impossible and that they are therefore unable to propose anything. Opting out is severely career-limiting.
The main cause of public policy being different is that , as the name implies, it is public. Behind politicians’ reluctance to specify their goals lies not only their inherent desire to avoid any blame for disasters they instigate, which seems to be the main motivation in Trump’s case, but also their knowledge that their electorate is even more reluctant to confront difficult trade-offs than they are. Voters notoriously want higher spending, lower tax and lower deficits all at the same time, in which self-contradiction they are encouraged by irresponsible and mindless media, both mainstream and social.
Overcoming the voters’ desire only for good news and their tendency to blame the messenger for any bad news is not easy. Educating the entire population in the principles of good engineering design might work, but only after decades and vast expense. Educating a few journalists in those principles might help a little, but the problem is that editors want pieces that will excite the public’s emotions, not ones that tell people things they do not want to know. That is also why news outlets carry far more items about new policy proposals, which can be used to claim that someone is ‘furious’ or ‘fearful’ of something, than about studies of whether existing policies have fulfilled their goals or not.
We are, I fear, stuck with goal control in public policy for the foreseeable future. And that does raise another point. If politicians refuse to say what the goals of their policies are, one of two things will happen: either effective policy becomes impossible or someone else in the policy process will fill in the gaps with their own goals and constraints. The former leads to inaction or fake action. The latter leads to policy being set by people who have little or no authority to set it.
No purely technical ways exist of deciding on goals. The only people who believe that such a thing is possible are a small group of deluded economists who think that we can put a price on everything, even values such as ‘justice’ and ‘fairness’. Values are not desires but goals and constraints about which we can think and talk and change our minds. And so, if we are to have any public policy at all other than fake policy, it matters who the people are who are filling in the gaps when political leaders refuse to specify their goals and constraints, and it matters how they have obtained their power to set policy goals without being elected.
This brings me to a final point, perhaps parochial but I think of broader significance. The University of Cambridge has just accepted a gift of £130 million, which might rise to £190 million, to create a new School of Government. I have been campaigning for a Cambridge School of Government since 2009, and so I am basically very happy. But one aspect of the plan is jarring, even disconcerting. Cambridge’s School of Government will sit not with the University’s departments of politics, economics, philosophy, public policy or law, or even with its rightly celebrated engineers or its business school, but with physics and chemistry. The University’s report on the new School talks about “applying analytical, computational, and scientific approaches to international relations, economics, business, society, and culture” and producing “systems-focused graduates who are fluent in integrating STEM-style analysis with the breadth of humanities and social sciences”. I am all for systems thinking, although I would gently point out that it developed as much in engineering and in sociology as in the physical sciences, but the problem is that the University seems to be assuming that goals can be set purely by “analytical, computational, and scientific approaches”. They cannot be. If we are going to try to create philosopher-kings for the age of AI, at least let them learn some philosophy.