Influence

Find me someone who wishes to have no influence in the world. Whether it be influence over their children’s development, influence over who is elected to hold office, or influence over preventing environmental disaster, we long for influence; to be able to bend the course of events towards our will.

And what is this wish but the desire for power? Impotence affords us no influence and our wishful thinking will amount to nothing.

Much has been said about how this influence is acquired. It may be bestowed upon us through appointment or privilege. Or it may be hard won as we work to assert ourselves and win trust. Some believe that our influence lies in charisma which we either have or do not. Our popular conceptions of influence are that it is a property of an individual or the position they hold. I find the notion of influence as simply a property of an individual problematic.

Discourse about leadership fetishises influence. If influence is a possession then leaders must possess more of it than their followers, indeed they may have secured their position because they already do. This assumption underpins many leadership theories. Trait theory describes the characteristics which ensure that leaders exert their influence. Leadership styles describe the ways influence may be employed. Distributed leadership advocates gifting others with influence. Influence as gift, bestowed on the leader or bestowed to others by the leader, is a common metaphor that is disguised by pseudo-scientific theory.

Of course, we do want our leaders to influence things. Thinking about what they can do to be influential is desirable to an extent. The problem is that if influence is imagined to be merely a property of the leader then leadership development becomes inward looking. We try to become more charismatic, more persuasive, more trustworthy, more just, and more respected. We falsely believe that influence is a skill we can cultivate regardless of context or situation.

Some have tried to shift the discourse away from the leader and towards the social fabric of the organisation. Such conceptions position influence as residing in the relationships between people; influence as social contract. This socio-psychological perspective may lead to less egotistical attempts by leaders to extend their influence, for example working towards building trust and belonging through healthy interactions between members of the group. Rather than influence as an exercise of power, the focus is on the social transmission of behaviours as people are tightly bound to each other throughout the organisation.

I am more comfortable with the relational view of influence than with the leader-as-influencer narrative, but it only takes us so far. Despite the work being done to model how we build stronger and more productive social contracts (an example of which I wrote about here), it still feels a bit… well, woolly.

If we zoom out for a moment, let’s consider what we are trying to achieve. Our goal in education is to secure better outcomes for children – a better school system. What role does influence play in this?

We could argue that both of the perspectives discussed above have a role to play. We could no doubt find examples of schools which have been turned around through the force of personality of the headteacher. Similarly, we could identify schools which have developed a collegial culture where innovations in teaching practice spread rapidly and embed. But what can we conclude from these examples other than ‘that worked here’ but ‘something else worked there’? That there was influence at work is without doubt, but what general form that influence took is less important than what specific mechanisms were triggered.

A mechanistic view of influence seeks to determine how exactly A leads to B, and so on to X, Y and Z. What is influence if not a form of cause and effect? If we better understand the mechanisms which lead to improved outcomes for children, we will come to exercise our influence more productively.

My thinking on this topic has been taken forward considerably by a recent post by Dr Sam Sims (here). In it, Sims repeats the claim made by Rob Coe that the school system in England has not improved for decades. He borrows a framework proposed by Joel Mokyr which explains why economic growth suddenly took off after 1800 when it had been effectively zero for centuries before. In applying it to education, Sims starts by setting out the following theoretical claims:

• We can distinguish between two types of systemic knowledge: knowing that and knowing how.
• Knowing how is constrained by knowing that. For example, knowing that the condensation of steam creates a vacuum is necessary (but not sufficient) to know how to build a steam engine.
• Items of knowing that often open up a multitude of knowing how.
• If knowing that is broad knowledge then it leads to more reliable knowing how. For example, knowing the condensing point of steam varies at different altitudes leads to more reliable steam engine technology.
• The least broad form of knowing that is ‘x works’ (no understanding of why). This knowledge takes longer for people to trust and adopt.

Sims then uses these theoretical assertions to explain why educational improvement flatlines. He argues that individuals in schools tend to learn how through personal experience, therefore their knowledge that is limited to context and does not spread. Knowledge passes on through trust (the point above about social contracts is relevant here), but both successful and unsuccessful learning spreads. The problem is that the system leaks knowledge (as people leave) at the same rate it develops it. This argument is similar to the weak knowledge-building described in our book on school improvement (here).

To improve the education system, this analysis suggests that there must be a better accumulation of knowing that across the system. Sims suggests this has started to happen in relation to the science of reading, some pedagogical devices, and professional development methods. Robust knowing that leads to more reliable knowing how. In other words, we have insight into the mechanisms that work and why they work, and these can be spread more rapidly and with greater effect than experiential knowledge.

What insights might we glean from Sims analysis to enlighten our discussion about influence?

First, we see the role that knowledge has in influence. School leaders have personal knowledge which is built through experience or shared (weakly) by others. But to improve the positive influence of leaders across the system we should concern ourselves with the domain of public knowledge. This perspective does not deny that improvement is secured by developing the leader (in this case their knowledge rather than traits), but it situates that improvement within a system that is also knowledge-building. In other words, to improve school leaders’ collective influence our best bet may be to improve systemic knowledge-building rather than the personal characteristics of the leader.

Second, we should ensure that our knowing that is robust and broad. If we are limited to knowing ‘x works’ without knowing why or under what conditions then our application in schools will be less reliable. School leaders will exert greater positive influence if we equip them with mechanistic explanations and an understanding of the contextual preconditions for mechanisms to have an effect. For example, a deep understanding of how children learn to read and what instructional approaches work, when, and how.

Third, we should learn to view theories of influence with caution if they address only the generic and not the specific. Building a culture of trust and strong social ties may be desirable but in itself will not lead to significant improvements in educational outcomes. Developing our powers of persuasion may help bend others to our will but what is the justification for what we are persuading them to do? The leader who understands the mechanisms for school improvement will secure impact by influencing the things that matter. The choice of mechanism is itself an act of influence, and the wisdom of this choice depends on our individual and collective knowledge.

We’ve moved from an individually-situated view of influence (influence as gift), via a socially-situated perspective (influence as social contract), to influence as a systemic quality. Which of these takes us forward?

If our goal is to improve the school system so that it delivers better outcomes for children then systemic knowledge-building is likely to yield better returns than developing the personal qualities of school leaders in my view. That is not to say we shouldn’t develop individuals or build better cultures, but without greater collective intelligence we will merely spread anecdotal knowledge through weak transmission mechanisms. We cannot rely on either charisma or trust to convince others that we know a better way to do things – we have to actually know a better way to do things!

It has been suggested that leadership may simply be defined as the exercise of influence. If true, then leadership is no better than the collective intelligence of the system. Perhaps we should concern ourselves with increasing our knowledge of the mechanisms by which educational outcomes improve more than with the qualities possessed by the lone leader.