Governing by speech acts and magical thinking
Policymakers can trick themselves into thinking words are a substitute for deeds
Senior leaders in any organisation, but especially in government, operate on the world not by physically doing things, but by ordering that they should happen. This can lead to a sort of magical thinking, which is the belief that just saying something means that changes will occur in the real world. I want to expand on this point and explore its implications for managing policy.
Speech Acts
In philosophy and linguistics, a ‘speech act’ is speech that not only conveys information, but performs an action. Classic examples might be something like “I now pronounce you husband and wife.” In saying this, the officiant actually performs the act of marrying the couple. It is not merely a description of the world, it changes it. I can say that “Adam and Eve are married”, but I haven’t caused them to be married by saying that, unless I’m ordained and doing so in the context of the appropriate ceremony.1
Managers mainly get things done via speech acts, especially senior managers because they are the ones who make decisions. Frontline staff operate directly on the physical world by doing things like laying bricks or performing surgery. Middle managers mainly operate by coordinating information and advice that then goes to senior managers to make decisions.2 A senior manager’s day mainly consists of meetings and one-to-ones with their direct reports, making decisions and giving steers or orders.
I suspect policymakers have this even worse than most senior decision-makers, for two main reasons. The first is that government is run by legislators. It is the primary job of MPs to debate and update legislation. Ministers have all sorts of other roles on top of this, but they are all also MPs or Lords and based in a legislative culture. Legislation is the ultimate speech act, with Parliament agreeing that someone else can or must do something.
Secondly, most organisations are more vertically integrated than government - that is, the same organisation that delivers things on the ground is the one making the decisions about how that happens. On top of Parliament already being removed from delivery, even within government departments there are often many layers of abstraction. A lot of government delivery is outsourced to separate organisations, sometimes private contractors but mainly public bodies like schools or the NHS. Even for work delivered directly by departments, decisions about that will be made in collaboration with the Treasury, No.10 and Cabinet Office who are not directly responsible for delivery. The main way policymakers influence the work done by these frontline organisations is ust by setting out legislation, guidance or strategy publications - speech acts.
Government also has a lot of soft power. If the government comes out and says that something should be so, people outside of government may start moving to make that so. This signalling power shouldn’t be overstated but it does tend to be stronger than even the most powerful of private organisations. It is also most effective when combined with supporting action or clear intention to act.
Magical Thinking, or Wizards of Westminster
A related concept is magical thinking. There are a few definitions of magical thinking, but I am using it to refer to a belief that if you say something should happen, it will. Exactly like a magic spell or incantation, speaking the correct series of words will mold the world into the shape desired by the conjurer minister.
I don’t particularly intend this as a criticism of senior policymakers. Of course it’s poor thinking and can lead to bad policy. But on an individual basis, it makes perfect sense for them to think like this. As described above, speech acts are how they make things happen. Magical thinking is simply the fallacious over-extension of the appropriate mechanism for senior policymakers to affect the changes they want. It must require a lot of intellectual discipline and humility to resist the urge to start acting on the hidden assumption that your words can directly impact the world. A lot of the time their words do impact the world.
The tendency to magical thinking can be exacerbated by the necessities of abstraction involved in knowledge work. Managers cannot directly observe every physical occurrence in whatever system, process or team they manage. Instead, they rely on data that abstracts from this, representing the real world as numbers on a graph or concepts distilled onto a slidedeck. Eventually, the information can become so abstracted that the signifiers begin to no longer clearly reflect their intended meaning - people and things can just become numbers. As you get more senior, the information you receive tends to be increasingly abstracted, by necessity to allow you to process the quantity of information. There are also more intermediaries between you and the information. A frontline manager may collect data directly from their staff. An executive receives data aggregated by perhaps hundreds of managers across their organisation. This problem can be even worse for policymakers, who are trying to understand and analyse complex social systems, perhaps aggregated from population-size datasets. Policymakers also tend to have weaker feedback loops than someone selling a product, who quickly figures out if people aren’t interested in buying it.
Despite being more senior, ministers may actually have better protection from this than senior civil servants. Civil servants can get ahead just fine wrapping themselves in an elite abstract bubble, although good ones will try to avoid this. But ministers are rooted in society through regular visits, trips and especially constituency days. However, there’s only so much that ministers making these connections can do. The minister has limited time and many people they could meet, plus this method is better suited to certain types of knowledge than others. Ministers may understand some of the struggles of their constituents or stakeholders, but a series of anecdotes can obscure broader dynamics, and for complex systems it may require a more analytical approach to understand the root cause of problems. So ministers still suffer from the same core problem of the senior executive.
Solving the problem
A lot of standard good management practice is focused on avoiding this tendency to vague abstraction and reorienting decision-makers to the real world. They help managers make gears-level models.
Michael Barber’s ‘Deliverology’ and the work of implementation & delivery units is partially built around combating this (for what it’s worth, I used to work in BEIS’ unit). The general idea is to strip out distractions and closely focus on a small number of key priorities. This is accompanied with clear metrics to monitor all the key parts of the system, and a small team of top people to dig into the system and understand how things are working on the ground, so they can clearly report this to senior decision-makers. There is slightly more to it than this, but a central reason why it works (when it does) is that it brings in a small team to directly link the decision-maker to the frontline, without the various layers of hierarchy and intermediaries.
Benefits management, logic models and theories of change do this too. All three of these are essentially the same thing, they just originate from different professions (project delivery, analysis and strategy respectively). These techniques produce a map that encourages the manager/analyst to explicitly draw out the logical connections between the activities undertaken by the organisation, the outputs delivered and therefore the desirable outcomes produced. These connections can then be examined, evidenced and monitored to ensure the effective delivery of the intended outcomes. It is typically done poorly and as a tick-box exercise, but when done properly policymakers can use it as a way of getting out of their abstractions and assumptions and thinking more concretely about what needs to happen in the world to enact the desired changes.
I don’t have a particularly smart way to conclude this post. I’ve tried to explain a problem I see happening a lot in policymaking, and explained that there are existing methods that can address it. My general conclusion is that all we need to do is effectively implement basic management techniques, which I suppose could be considered the core theme of this blog. Underwhelming, but maybe that’s the point. Management was never supposed to be sexy.
You could argue signing documents is what officially marries someone, but hopefully the distinction still makes sense.
I think this is a good one-sentence explanation for why middle-management is often demotivating.