On July 3 we had our third PNI hangout. Again it was a useful lively meeting on the topic of complexity in PNI. It is already becoming a bit of a habit to approach PNI hangouts “the PNI way” and start sharing stories after a short intro of new attendees. So that’s what we did this time. The same holds for the way this blog post is formatted: as is fitting with PNI anonymity, we won’t share names.

We started by talking about complexity and how people see it.

Visible and invisible

A said: Over the years I’ve learned that complexity is everywhere, but it is not equally visible everywhere.

  • Biologists, for example, take complexity for granted. It’s such an integral part of biology that biologists don’t talk about it. For example, what Marshall McLuhan calls “the medium” is the complex mechanisms through which actions return (in often changed form) to acting (or non-acting) biological agents.
  • In the physical sciences and math, people mention complexity a lot. There are the people that work with non-linear models, the agent-based ones and the various types of systems people.
  • In the social sciences, people talk about interactions and feedback, but they usually don’t mention complexity (even though it’s often implied). People in social sciences usually discover systems thinking first, and that leads them to complexity.

B responded: Yes, and when people demand or expect to talk about complexity, it can be perceived as a bit of an insult to people in some fields, because it has already been there forever (though not always with the same terminology). That is sometimes a barrier to talking about complexity. Some people are like, this is all new, and other people are like, no, it’s not.

C said: In the humanities, it is not something that is generally included. But when you get into epistemology, it’s in there too.

What is a system?

A said: Some ten or even fifteen years ago I made a list of what people mean by “system.” It got confusing after seven or eight perceptions.

  • If you ask IT people, it’s some piece of hardware
  • If you ask programmers, its some piece of software
  • If you ask OEM’s, its a product
  • If you ask consultants, it’s a model of reality
  • If you ask biologists, it’s an ecology
  • If you ask medical people, it’s a body
  • If you ask humanistic social scientists, it’s a group
  • If you ask some mathematicians, it’s a simulation of agents; if you ask others it’s …. any number of things

The list becomes more and more useless as it becomes longer.

B responded: A system is a decision. It’s all about where you put the boundary, and that depends on where you’re coming from and where you’re trying to go.

Interaction and iteration

Flock of white storks


B said: I look at complexity in PNI from two sides. From one side, stories form complex patterns in human groups. Complex patterns feature an interplay of interaction and iteration (which is really just interaction over time), and working with stories is one of the primary ways in which people interact with each other in an iterative way.

A responded:  For me the link between narrative and complexity is about how emergent behavior and properties of social systems arise. What you said about interaction and iteration reminds me of some good work on this from the ’80’s and ’90’s. In particular by Hejl (1984, building on Beer, Maturana, Foerster, Varela, Checkland and others) and Goodwin and Reason (1999).

For example, Goodwin and Reason mention six principles of complexity science: rich patterns of interconnection (among diverse components); cycles of iteration (defined by constraints, giving rise to order); emergence (impossible to predict from the characteristics of the interconnected components); holism (the whole being more than the sum of its parts); fluctuations (patterns that emerge from apparent randomness); and the “edge” of chaos (more properly the zone of emergence, where there is a mixture of nascent unorder and patternless miasma).  

Language is also a Complex Adaptive System. In the work of Luc Steels, an example is given of old Dutch words that survived in the Afrikaans language (South Africa) while they have long disappeared from contemporary Dutch.

C responded: Stories are social actions. This is important because stories are meaning-laden. The process of meaning making is important for us as people, and we do that through sharing stories as social actions. It is in the interaction and iterations of those actions that meaning is negotiated. The stories are not the reasons we do it; they are the vehicles by which we do it.

B said: Story interactions take place at three levels: inside stories (that’s story form, like intra-cellular mechanics), at the boundaries between stories and the rest of life (that’s story function, like the cell membrane), and outside the story proper (that’s story phenomenon, like cells forming organs and some cells traveling around the body).

Story iterations take place as stories repeat and flow around people and society over time. When you work with stories you use them to keep your finger on the pulse of the complex patterns that flow through the stories.

I don’t know how to say that well yet.

A replied: This PNI Institute is all about talking about and working on things we don’t know how to say well yet.

D said: I can provide an example to illustrate. We are working on a project that aims to bring together knowledge from patients with rare diseases. These patients have knowledge that nobody else might have, not even people in the medical profession. I recently spoke to one member of a group of patients about the life stories we had collected and he said:

It looks like the story project we are doing is like collecting pieces of a puzzle, but we don’t know yet what picture will be on the puzzle. It might even be that the image on the puzzle changes while the pieces are collected.

A said: That’s what emergence is about.

History and hysteresis

B said: From the other side (going back to the “from both sides” thing), all complex patterns have stories, so making sense of complex patterns requires understanding the stories of how they take place.

Complex patterns have a contingency, a historicity about them that determines not only their shape but also their future. The phenomenon of hysteresis is relevant – that if you try to go back to a previous state you can’t. You can go back to something similar (in some ways) but you can’t go back to anything precise. (Wikipedia says: “Hysteresis is the time-based dependence of a system’s output on current and past inputs. The dependence arises because the history affects the value of an internal state.”)

An example in ecology is desertification, where a pattern might repeat near a boundary, but after it crosses the boundary it can’t get back. That’s kind of like bifurcation, but it has an iterative aspect to it as well.

A responded: There is a lot of order to be found even in the complex aspects of systems. There are patterns, like seasons, that are hard to change, or they might be habits, things that resist to being moved from their orbits. It seems a lot of people who talk about complex adaptive systems assume complex means it’s not ordered, but in fact the patterns in there are a form of order. These patterns are a blessing in disguise, because they bring resilience to the system. Systems with classical forms of order are brittle and break easily. Systems with complex patterned order resist change while being capable of handling disturbances and veer back. This poses the question how we can follow patterns over time. There is still some PNI theory lacking there.

C responded: I have a question about your use of the words “history” and “states.” It’s related to what you said about hysteresis. Can you ever go back to a story? Can you roll back a story? Can you go back to a previous pattern? In PNI work, when you ask people to tell the story about an organisation on one day, then they tell you the same story another day, the story has changed.  I sometimes feel a sense of loss about this, that the original stories were lost. We worked on a safety project where we wanted to record the telling of stories as videos. In the process of telling the story, every time it was told it changed. There was a magic in the earlier versions that was never repeated. I thought it was because people were forgetting, but I now think it was because the story itself was changing.

A replied: You’ve just formulated the core of PNI, that the story changes by telling it. You can’t step into the same river twice, and you can’t tell the same story twice.

C asked: Is it even possible to tell the same story twice?

Everybody else said: no.

A said: This is related to resilience.

The end of resilience

B replied: Yes, but there is an invisible point (possibly a moving point) where you fall out of resilience.

A said: PNI should help people observe a bifurcation point. We should help them find it. By bringing a new narrative, the new pattern will have a new narrative. For example, people might say that, “We are no longer a company for X, we are a company for Y.”

B said: It’s true that complex patterns are resilient, but resilience doesn’t guarantee longevity. If you have to choose between putting your energy into saving a complex pattern and saving an ordered pattern, you should put your energy into the complex pattern. Because even though complex patterns are resilient, there is an invisible point at which that resilience breaks and can never be fixed (like with the desertification example above). Because the system is opaque, or at least cloudy, you need to err on the side of caution – it works and works and works, until it suddenly doesn’t. An ordered system, on the other hand, can always be rebuilt, even if it takes a lot of resources.

Complexity and values

A pointed out:  Yes, and there are examples where complex patterns have to sacrifice themselves to save other complex patterns. In this event that just happened in Tunisia, where there was a gunman on the beach, one man actually walked towards the gunman to protect his wife and child. Sometimes one complex pattern does that for another complex pattern.

That’s what innovators do. They say, “I might get hurt, but I want to create a new system.” Innovation has more to do with values than it does with anything else. It’s like what we said above about the puzzle and not knowing what will be on it. It’s not that somebody had a vision of what would be on the puzzle; it’s that somebody left the old system of values and jumped into the unknown. They didn’t know what would happen next, but they knew what was guiding them.

B responded: That’s the problem with books on complexity. They don’t want to tell people that it’s about values, because nobody wants to buy books like that. They want to make it seem like complexity is controllable. It is, but only partially.

A said: Still, there are things you can do to influence complex patterns. You can nurture small alternative patterns that might eat or disrupt the dominant pattern. People do that with skunkworks. I saw some research that said people in a new company need to be outside one day driving or flying distance, or the new culture won’t take. It won’t establish a new narrative pattern.

PNI and complex patterns

C asked: How does PNI  help people establish a new narrative pattern? I think one way is, PNI can help the leader hear the stories.  Often leaders in large organisations, consciously or unconsciously, choose not to hear the stories. Or because of the large scale, they can’t hear them. That’s one of the simplest things PNI can do, help the stories be heard. For leaders and for anyone in the organisational structure.

People want want to hear the stories for a reason. To update their mental models, to get information, to create change. That’s also where the participatory element comes in, because just hearing the stories is not sufficient. Hearing is an interaction that creates an iteration.

A responded: Suppose you are director of a firm with 10K people, and you want to know who are the 3 people with a new value system. The mistake people make is, they say, I want to make these people director of X. I say, no, it’s better to say, I want to hear more stories of this and less of that, but without tying it to an individual person.

I also say that if the director has doubts about the project, they can kill it for a week, and then see if stories surface because of that. The reaction tells you whether the project should continue. You can tap in to diversity of perspectives about project. Whatever you do, whatever story you return to the organisation, you get new stories.

C asked: Isn’t the process of describing a pattern a language act, which means that it’s a story?

B answered: I think the whole project is a story.

C said: But there are so many different layers. The process of having your finger on the pulse changes the story of the pattern you are detecting.

B said: I remember a time when somebody showed me an organisational chart, and I said, “Where are your trick stairways?” I meant, where are your connections, so that the hierarchical elements of the organisation are not the only thing you can rely on. I think PNI is partly about creating trick stairways.

A said: Yes, instead of a formal organisational chart you get a fine maze of story.

Where do stories live?

A said: I think stories are like cats that visit us and live with us, but also walk around and visit other people. Sometimes people think that collecting stories is about measuring individual people, but it’s not. Because stories move among people, collecting stories is about understanding the whole environment in which the stories live.

A responded: Yes, you can tell a story but find out that it comes back to you changed. It’s like a virus, it has genes. The genes evolve with the hosts, and with 16 hosts it evolves 16 times.

B said: It’s like the stories form a layer that is tethered to us but floats above and forms its own complex patterns.

A comfortable distance

C said: How can PNI help to make sense of that layer, and to use that layer to improve things? One way I see is that PNI gives a comfortable distance for leaders. It is a social action to listen to stories, and to prompt for stories. When people share stories, they want to know that story has been received. They also want to know HOW it has been received. There is a social exchange to it. Sometimes leaders are afraid of this exchange. By listening to stories (rather than opinions) a distance comes in, because leaders can receive the stories and reflect without having to do anything right away.

This is also a danger of PNI, because the leader doesn’t have to do anything other than hear the stories. People might think PNI is not about doing anything lasting, but just lip service. That’s why the return phase is so important. People need to do something to close the loop. But using stories puts leaders into a position where they can look at what is happening in an externalized way.

A replied: Collection without return is manipulation.

C said: Yes, and that’s true whether it’s intentional or unintentional.

A said: Collection with return has less of that danger.

C replied: And the prerequisite for return is courage.

A elaborated: If you extract stories, you think of them as YOUR stories, and there is a risk of manipulation. But if you have participation, you say these are OUR stories, and we ALL have this space for reflection. Then you get an ecology. You also get anti-stories, but that’s healthy. You end up in a situation – like that puzzle again – where you say, we don’t know where we want to go, but we want to go away from here because X is happening.

B said: The layer of stories that floats above us is an epiphenomenon, an emergent property. This relates to the comfortable distance; because the epiphenomenon is connected but not identical, it is easier to work with, more indirectly. I’d like to do more with that later.

C remarked: This call has been different than our previous ones. In those we traded experiences, but in this one we covered some new ground. That’s great. We’ve stumbled onto things that need to be explored more. Maybe some more topics for the hangout will come out of this.

Looking inside

B said: I did think of one story to tell from my experience, related to PNI and complexity. It was a project where two firms were merging. The two firms had two very different cultures related to complexity. One was go-with-the-flow, flat structure, comfortable with uncertainty. The other was hierarchical and hated uncertainty and complexity. The project was canceled, mainly because the hierarchical group couldn’t stand the complexity of the project itself (too much contamination with crazy ideas). I think of that project as a challenge to PNI. The approach needs to handle a project like that – of differing reactions to complexity – more effectively. I don’t have a good solution for that yet.

A responded: The way you run a PNI project for one of these companies needs to depend to some extent on whether it’s an internal or external problem. A lot of people I meet at clients say PNI is great for external views, say of customers, but they don’t want to look internally. They are like, “We can look at what’s out there, but let’s not look at ourselves.” I’ve seen some research that says the more powerful the hierarchy is in an organisation, the stronger the informal or complex networks become, because people need to find a way to get their work done in spite of the unnatural degree of hierarchy. So I wonder how we can use PNI to help people make that work.

C said: I liked what you said about “let’s not look at ourselves.” I wonder if there is an anxiety about looking inward that creates that fear response, and I wonder what we can do as PNI practitioners to help people get past that.


In summary, everyone enjoyed the call immensely. We agreed that we should talk about complexity again sometime, since there is so much depth to the topic.

Come to our next Hangout!

Our next call will be on Friday, July 17, at:

  • California 7 00
  • New York 10 00
  • London 15 00
  • Amsterdam/South Africa 16 00
  • Melbourne 24 00

The topic will be PNI and Power Differences. The Google Hangout we are using is at: