7
Steps Toward a Theory of Systemic Action

There is nothing so practical as a good theory.

—Kurt Lewin

When the first astronauts went through their paces in the high deserts around Edwards Air Force Base, they were reluctant instruments of what Heinrich Heine calls “men of thought.” They were asked by the “white-coats” to do things that made little sense to them. From time to time they protested against being treated like experimental chimpanzees and pushed back. Ultimately, however, they made their grudging peace with the fact that they were subjects in a vast experiment based on theories they did not know.

Much of the BAU behavior is the product of theory, most of it long forgotten or remembered only by specialists. If we are to come up with effective responses to complex actions, we need to align our actions with our best thinking about complex social challenges. Theory must inform our actions. When we think that’s not the case, it’s usually because invisible theory is informing our actions. We therefore need to be aware of these underlying theories. At the moment, much is guided either by theory ill suited to complexity or what are sometimes labeled “theories of change,” which are little more than elaborate hypotheses labeled as theory.

While what is being outlined here is being called theory, this needs a little clarification. As previously explained, practical wisdom is best expressed through heuristics (rules of thumb, checklists) that are highly context-dependent, as opposed to universal principles, which are independent of context. A theory rooted in phronesis is therefore more akin to a recipe or a checklist than to a normal science theory or even a social science theory.

A recipe or a checklist is predictive in the sense that if one combines certain ingredients under particular conditions and uses the prescribed techniques, one gets the expected outcome.¹ However, it does not make sense to think of either being falsifiable in the way we would think of theory in the natural sciences. A recipe that fails in the hands of a bad cook does not mean the recipe does not work. But then how does one tell if a recipe is bad? Well, one simple answer is experience. When you cook it, it tastes bad. If a recipe corresponds to a set of instructions for producing a particular outcome, then the point of phronesis is that the only real way of learning how to cook is to cook.

Just as handing a novice a recipe will not produce a world-class dish (other than in the same sense that enough monkeys bashing enough typewriters will produce a Shakespearean play), no amount of simply handing people maps or tools will produce systemic action. A single-minded focus on either episteme or techne is like buying the best cookbooks and top-of-the-line cookware without bothering to learn how to cook. It doesn’t make much sense.

1ST REQUIREMENT: CONSTITUTE A DIVERSE TEAM

The recent focus on the imperative for business to innovate has resulted in a new slew of insights into how best to build teams that are good at innovation. In their book X-Teams, Deborah Ancona and Henrik Bresman make the case that building teams “that lead, innovate, and succeed,” requires what they call X-teams, where X stands for external.²

They elaborate:

The world is plagued by complex problems like poverty, global warming, and political violence. These problems can only be addressed when people from diverse sectors like business, government and nongovernmental organizations (NGOs) work together. It is teams that will ultimately be the major actors in carrying out this important work.³

X-teams differ from traditional teams in that they have “high levels of external activity,” and they must “combine all that productive external activity with extreme execution inside the team.”⁴ Extreme execution is simply another way of saying that the team must focus not only on the external, but must be able to effectively combine external activity with the development of effective internal processes. Finally, X-teams “incorporate flexible phases, shifting their activities over the team’s lifetime.”⁵ In other words, X-teams need to be agile in terms of their processes. Focusing on the external, as opposed to inside an organization, clues us into how to convene teams that are good at innovating.

Given the opportunity to bring together a dream team to work on a systemic challenge, who would you pick, and how would you pick them?

On the first-generation social labs we ran, the rule of thumb for answering this question was “diverse and influential.” There are two components to this rule of thumb that bear examination and explanation. Diverse was interpreted to mean multi-sector, which in turn was interpreted to mean that we constituted teams with representatives from government, civil society, and the business sectors.

The second half of the rule of thumb—influential—raises the question: who in the first half of the twenty-first century is influential?

The Half-Life of Power

Power is decaying.

— Moises Naim

In 1956 the sociologist C. Wright Mills wrote a book called The Power Elite.⁶ The premise of the book was that the leaders of the military, corporate, and political spheres of society held power and that ordinary citizens were comparatively powerless. Mills’s conception of power was relatively traditional—a small elite leadership has the most influence on society.

The convening strategy we deployed in order to recruit “diverse but influential” participants for first-generation social labs was a classic Millsian approach. Captains of industry, government ministers, and the heads of large NGOs were deemed to be influential. Grassroots activists, artists, care workers, farmers, and members of rural communities, for example, were not deemed influential enough.

The French philosopher Michel Foucault revolutionized our understanding of power in the decades following Mills. Foucault showed through detailed historical analysis that the nature of power was very different from traditional conceptions of hard power, that is, as something that could be owned and used like an instrument. Power was instead symbolic and relational. He argued, “There is no power relation without the correlative constitution of a field of knowledge, nor any knowledge that does not presuppose and constitute at the same time power relations.”⁷

In 1991 Joseph Nye from Harvard University coined the phrase “soft power,” describing it as “the ability to get what you want through attraction rather than coercion or payments. It arises from the attractiveness of a country’s culture, political ideals, and policies. When our policies are seen as legitimate in the eyes of others, our soft power is enhanced.”⁸

Then, in the last ten years, the notion of who is influential and who is not has morphed several times. Bruce Sterling commented, “Wikileaks and Facebook—which weren’t even around five years ago—have more political clout than the state department and the US military combined. It’s nothing to clap about. It’s actually a calamity.”⁹

The rise of the networked society and increasing social complexity has prompted new research that has clarified our understanding of how highly connected societies function. In a 1998 Nature article, Duncan Watts and Steve Strogatz articulated a mathematical model for what they called “small-world networks.”¹⁰ Strogatz and Watts applied their model to a sociological explanation of how our world works. Their research explains what has been called “small-world phenomena” (such as the idea that we’re all six degrees of separation away from actor Kevin Bacon), but also covers the rapid spread of viruses, the connectivity of the Internet, and gene networks.¹¹

The basic idea is that small-world networks consist of clusters that are weakly linked through a small number of network agents. In social terms, this means that people are related to each other through relatively small, tight clusters, but a few members of these clusters are weakly connected to other clusters. It’s through these weak links that information, viruses, and connections in general occur.

The rise of small-world networks and what are called scale-free networks (like the Internet) changes the notion of influence in society. Paul Adams, who previously led Google’s social research team and is currently working at Facebook, counters what he calls “the myth of the ‘influentials.’”¹² Adams argues, “Trying to find highly influential people is a risky strategy.”¹³ This is because the success of any highly connected individual fluctuates wildly, and it’s not clear how we actually measure influence consistently.

During the first-generation social labs we ran, looking for influentials was our convening strategy. In the case of the Bhavishya Alliance, we spent millions of dollars and hundreds of hours interviewing people across India in an effort to find the right “diverse but influential” participants from the entire field of child malnutrition.

Commenting on Watts, Adams writes, “The most important factor was not whether there were influential people but whether there was a critical mass of easily influenced people who were connected to other people who were easy to influence. When this critical mass of connected people didn’t exist, not even the most influential people could get an idea to spread widely.”¹⁴

My colleague Mustafa Suleyman, in thinking about our convening strategy, had the breakthrough idea of simply broadcasting an invitation through our networks and seeing who turned up. After attempting our original strategy unsuccessfully while convening the Finance Innovation Lab (very few people from the finance sector we spoke to wanted to join an effort to change it), we eventually sent out an email invitation to five thousand targeted people. This resulted in running three events where three hundred people turned up. These people eventually coalesced into a network that formed the participants of the Finance Lab. It cost a fraction of what our Bhavishya convening strategy cost and took a month, compared to two years of searching high and low. This is what I refer to as open convening.

Interestingly, for the Finance Innovation Lab, this strategy yielded a group that is not influential in Millsian terms and has instead created more of an inside-out dynamic, where key mainstream actors in the financial sector were relatively wary of participating fully in this group. The question of what impact this group of hundreds of people will have on the financial system is still open and in play.

The question of how to constitute a dream team to respond to a complex social challenge therefore remains open. Developments such as the rise of the networked society, more sophisticated understandings of power, and the actual evidence base for who has influence in society means that people who would not be considered elite in Millsian terms are now able, more than in any other period in history, to determine the shape of their own lives.

This shift in how power is approached represents one of the most significant differences between first-generation social labs and next-generation labs.

2ND REQUIREMENT: DESIGN AN ITERATIVE PROCESS

Here’s a thought experiment. Imagine that you’re given a task, either by your boss, a client, or a minister. The task is to get a group of people who are standing in a valley to climb a mountain. You’re free to use whatever means necessary in order to achieve this task, and the assumption is that people will stay on top of the mountain after they’ve been moved. What would you do?

Perhaps you simply ask people to go to the top of the mountain. What would happen? The first thing that would happen is that people would ask you why you want them to go. Soon after that, at least one person would ask the feared question, “Who are you to tell us where to go?” What do you do then? Well, you need a strategy.

A bad strategy would be to simply describe the desired outcomes, without providing any information about how to achieve them.¹⁵ A competent strategy would propose how to get the group from the valley to the top of the mountain, with a clear grasp of the practical difficulties and taking them into account. A planning-based approach would exhaustively document every step of the journey, logframe it, and then hire a bunch of people to execute it. The longer the journey, the more uncertain the terrain and the higher the probability that the group following a planning-based approach will fail.

A Statist or Civil Society strategy would consider a number of BAU options. A developmental strategy would involve convincing the group that it had to move for its own good, and it would probably take years. A humanitarian strategy might be to tell the group, to the dramatic backbeat of helicopter blades, that they are about to be swept away by a tsunami. A security strategy would involve riot police and vans that would forcibly move people at gunpoint. If a battle-space strategy is being used, the group is assumed to be hostile because all other strategies have been tried, and it’s questionable how many of the group would survive the air war prior to the ground assault in order to be forced up the mountain.

In the last two instances, the solutions are unlikely to be easy or sustainable. It’s likely that once the threat of force recedes, the group would simply steal back to the valley after dark. Or worse, they would return and lace the road with IEDs to deter further assaults. Of course, each of these approaches gets increasingly expensive, requiring more resources and resulting in more damage.

A market-oriented approach would be to wait until others had created a market for the journey. Or if a client were willing to take the risk and foot the bill, a marketing communications agency would design an advertising campaign to convince the group that it was cool to get to the top where winners lived like kings and that only losers lived in the valley. Then the task would be to set up a limo service to take those who could afford it to the top as quickly and painlessly as possible, while selling food, supplies, and equipment to those too poor to get the A-list treatment. And so on.

In base terms, what is required to get the group from the valley to the top of the mountain is energy. Gravity means that a certain amount of energy is required to move the mass of the group. This minimal energy must come from somewhere. If the group decides it’s going to drive up the hill or take helicopters, then the amount of energy required goes up immensely. The most sustainable solution to the challenge is, of course, that people decide for themselves to get to the top of the mountain, collaborate with each other, and get there using their own locomotion—to walk up. This is a subsistence strategy and the only sustainable one. As a solution, it does not require a helping industry, a military-industrial complex, or an energy sector to drill the Arctic. And it doesn’t produce industrial-scale environmental waste. At most, the requirements are food, water, and suitable clothes.

In other words, there are more and less effective strategies for getting the group from the valley to the mountaintop. Sometimes, the scale, complexity, and abstraction of our challenges blind us to the simple verities, such as “Behind the world’s most difficult problems are people—groups of people who don’t get along together. You can blame crime, war, drugs, greed, poverty, capitalism or the collective unconscious. The bottom line is that people cause our problems.”¹⁶

What is certain, however, in all these scenarios, is that if the group does not want to move, it’s difficult and expensive (in all senses of the word) to move it—with no guarantee that the group will stay moved. What’s also certain is that if this group decided, truly decided, it was going to get to the top of the mountain, it would take a lot to stop them, as countless epic journeys and expeditions literally demonstrate.

The task of supporting this group to move of its own volition is typically what we, in our work, call process facilitation because it focuses on the process of a group self-determining where it wants to go and then inviting a facilitator to help it get there. The role of the facilitator then becomes to support the group and deal with the how of the journey, issues such as leadership (is anyone leading?), decision making, conflict resolution, and clarifying purpose (for when the journey gets really tough).

A good strategy in the context of complexity would include an iterative process. The simplest form of an iterative process is trial, error, observation, and reflection. You try something out, wait to see what happens, and then make another move based on what you’ve learned. The more complex the challenge, the more sophisticated a search strategy needs to be to find the way through the terrain, but the core essence of any iterative process is the same. Of course the ideal trial is one that is fail-safe—one that results in no lasting damage if it doesn’t succeed.

The notion of a process is philosophically oriented toward social constructivism. The idea behind social constructivism is that social processes produce everything we usually assume as having an essential or objective character. Social constructivism arose out of a school of thought labeled the sociology of scientific knowledge (SSK) that studies science as a social activity. The more radical proponents of this school would argue that scientific or technological labs are also social labs, in that social processes constitute them and determine what happens in them.

3RD REQUIREMENT: ACTIVELY CREATE SYSTEMIC SPACES

Crane Stookey runs an experiential education program called the Nova Scotia Sea School. It is based on the Container Principle: the wisdom of no escape.

Stookey explains how this works:

The image that best describes this principle is the stone polisher, the can that turns and tumbles the rocks we found at the beach until they turn into gems. The rocks don’t get out until they’re done, the friction between them, the chaos of their movement, is what polishes them, and in the end the process reveals their natural inherent brilliance. We don’t paint colours on them, we trust what’s there.¹⁷

The work then of process, in the context of the Sea School, is to create an environment—a container or a space—that lends itself to experiential learning. Social labs are space in the same sense. Process is used not in order to engineer a pre-determined set of outcomes, but rather, to create a container within which strategy can emerge from the friction of diverse participants working together as a team.

This notion is echoed by Steven Johnson in his book, Where Good Ideas Come From: “Some environments squelch new ideas; some environments seem to breed them effortlessly.”¹⁸

John LeCarre once wrote, “The desk is a dangerous place from which to view the world.” Yet, so many of our efforts to address complex challenges are born, live, and die at desks and boardroom tables. These spaces are by their nature homogenous and static, especially when contrasted with spaces that are actively produced and are dynamic, like the Sea School.

Static spaces are designed to support static organizational structures where relationships are carved into org charts, facts are written in stone, access is controlled, and people come to work and do the same thing day in and day out. The furniture is bolted down, the doors are shut, and the world is a computer screen. Such spaces are designed for control, Soviet in spirit, and dominated by a set of unchanging dispositions.

In practice, unless you’re an architect, urban planner, or interior designer, you probably work in an inherited space, that is, a space over which you exercise very little active control. The world, in static space, is not something we actively construct. Perhaps we can pin some postcards to a wall or put photos on our desks, but that’s largely the extent to which modern working space is actively shaped by those within it.

Henri Lefebvre attempts to articulate a “unitary theory” of physical, mental, and social space, which serves to diagnose our current condition.¹⁹ For our purposes, Lefebvre makes three main claims. The first is “Social space is indistinguishable from mental space (as defined by philosophers and mathematicians) and physical space (as defined by practico-sensory activity).” The second is that social space is a social product. Third, “This act of creation is a process [and] … every social space is the outcome of a process with many aspects and many contributing currents, signifying and non-signifying, perceived and directly experienced, practical and theoretical.”²⁰

These three claims by Lefebvre can be used to understand how BAU habitus operates. BAU is the result of historical processes, which have produced a number of spaces. These BAU spaces—development, humanitarian, security, and battle spaces—can all be thought of as examples of what Bourdieu calls fields, or what Lefebvre calls dominated space, which are “invariably the realization of a master’s project.” All societies produce their spaces according to Lefebvre. BAU spaces are the products of a particular society—ours—that values episteme and techne over phronesis. Our spaces reflect this bias, and, hence, BAU spaces are hostile to the activities of phronesis.

We therefore require the production of a new, systemic space supportive of phronesis and of emergent strategy. Lefebvre comments, “A new space cannot be born (produced) unless it accentuates differences.”²¹ This space is in actuality a heterogeneous space, one that allows for the informal to exist with the formal. It’s a space that’s externally oriented but can also turn inward when the need arises, one that is supportive of diversity and difference.

Finally, it’s clear from the countless examples of the kinds of spaces that engender creativity, innovation, and problem solving, that such spaces must be autonomous, allowing high degrees of freedom. Systemic action, therefore, requires a particular space to support it and a particular organizational form that is actively designed to be systemic in nature. This space is what I refer to as a social laboratory.