Next time you think you are creating something new …
Conway’s law says that we can only build things that are organised the way our organisation works.[1] The understanding of organisation that we bring to design and architecture is gained by seeing what works around here. But that raises the mother of all questions: how do we know what works and what it is that works? In this blog, we keep exploring the difference between work and real work, between outcomes and seeing what is really the result, between impact and understanding our role in the system.
Most people understand that the official line, the management structure and organisation chart, the set of incentives dreamed up by HR have little or nothing to do with how things work. People find workarounds to allow them to do real work no matter what the official line is.[2] What we are pursuing here is the notion that Conway’s law is about the way things work, not the way things are supposed to work.
We can only marvel at the architecture of a termite mound. But the termite mound is the result of the termites doing the real work that they need to do. It is not an imposed structural form within which they get busy. Staying with ants,[3] we can think about stigmergy, with a massive hat tip to Steven Brewis at BT Research.
Stigmergy: A mechanism of spontaneous, indirect coordination between agents or actions, where the trace left in the environment by an action stimulates the performance of a subsequent action. Wiktionary.
Footpaths
I think that this metaphor is now embedded in organisational folklore. If people cut across the grass and push through the hedges in the car park, don’t put up a fence to stop them, build a path where people actually want to walk. Don’t worry about the reason that they want to walk that way, there’s a reason that these are known as ‘desire lines’.[4]
The footpath is a stigmergy mechanism. It indicates to new arrivals that people walk that way, and the stronger the trace of the footpath, the more likely it is that people will follow it. One caveat is that there is a chance that the people using the path are not your target population: for instance, schoolkids cutting through the hospital car park.
With this metaphor as an image in mind, and the slightly more technical and less visible notion of ants laying pheromone trails wherever they go, we can think about what people actually do in an organisation as distinct from what they are supposed to do. To the ethnographer there are paths and rituals and a nexus here and there. Not to mention rhizomes. To any passionate observer there is an intricate wealth of unexplained detail, a courtly dance that no-one teaches and everyone follows.
And if we unearth these things that are more real than the HR niceties, we will find some structures, that will absolutely be as architectural as a termite mound. They are the traces of real work.
Another caveat about this ethnography. I was studying the denizens of the night clubs in a Midlands binge-drinking town. I suspected that the girls in particular were under-age and I asked for police support in asking some discreet questions about what parents knew and about real ages. (False ID could at that time be obtained from Spain for £20.) In the event, the police found someone using someone else’s passport as ID and treated it with Plod-sized boots, so that we lost all hope of getting any facts. Think about this: enforcing rules that no-one follows is deemed much more important than being able to see what is happening. Asking why comes nowhere.
Philip has a colleague whose story illustrates this question of stigmergic paths. He was a high-flying (sorry) rep for Rolls Royce, in a new city every day. He had a PA who, despite time differences, would ring him at his hotel each morning and tell him where he was and what he needed to do next. When the PA went on maternity leave this RR colleague could no longer do his job and had to leave.
And in a similar vein, our friend Christoph does Occupational Therapy, typically trying to help people put their lives back together when they have burnt out at work. Christoph gets incredibly irate at the way corporates can destroy people’s lives while blaming them for poor performance. His record is treating 13 consecutive appointees to the same role, each one going off work on long term sick leave before the company concerned was prepared to acknowledge that there might be something wrong with the role.[5]
This effect of the trace of real work is visible when we look, and we know that we can get the architecture of work seriously wrong. We don’t tend to consider the Conway’s law question and ask about what we can and can’t design from where we stand.[6] The breezy entrepreneurial view (“we can invent the future”, “we can disrupt you out of existence”, etc) is subtly two edged. To the extent that it is true, it implies that the people concerned are already organised differently. If you are organised differently, then many conclusions people may want to draw from their experience and many rules and regulations will not apply.[7]
I was invited to work with the (then) very innovative Durham PCT, in particular to develop a Viable Systems Model model of the local health economy. Before we could get started, they were to have an inspection by the CEO of another PCT. This CEO said it was only possible to inspect if Durham were organised more like her own PCT! So, they reorganised! Really. These were the first PCT staff I had ever come across who had a grip on the job that the PCTs were supposed to do and they were neutralised (neutered, even!) in order to assess their work.
Different architecture
I have been watching a film of democratic schools in the Netherlands called School Circles.[8] The kids, from small to large, are all concerned with how decisions about their school get made, not as a theoretical and stupid piece of sociology, still less as educational theory, but as an intensely practical day to day question about getting their school experience the way they want it. I was charmed by the children signing in an out on their school flexitime. “We are supposed to be at school for at least four hours in a day and at least 27 hours in a week.”
As I watch the film and look at how the kids can distinguish between consent and agreement, I can’t help think of the bitter partisan wrangling, both sides of the Atlantic, and how these supposedly grown-up politicians don’t have a tiny fraction of the skills that they need. But we must keep that as an architectural or at least stigmergic question. The idiot politicians take the paths that they do because those are the paths that they take, and those paths were created before them by people who in some twisted way got the results that they wanted.
Small children are able to devise ways to get on together and for everyone to gain from the mutual experience. Cynical adults think (despite what they say!) that it is a zero-sum game and they have the whip hand. The contrast between these is just excruciating to focus on, but it shows the meat of our argument about Conway. You can only create things from your experience of how the world works. Especially if you are an alpha male.
Indeed, the person or group with the most nuanced view of the world is likely to be on the wrong end of a power gradient. The distinctions are necessary for survival and a chance at happiness. The stereotypical alpha male needs to hold onto far fewer distinctions in order to blunder about successfully.
Those who don’t know history are doomed to repeat it. Edmund Burke
We think we are creating the future to be different from the present. Conway’s Law says that structurally, architecturally, this cannot be true. Burke seems to be saying that if we really understand where we are coming from we may see how to sidestep our doom. If we see how our past produced our present in unexpected ways, then we are already operating from some new truths.
plus ça change, plus c’est la même chose. Jean-Baptiste Alphonse Karr
Where is the leverage point?
We all believe we understand how control in organisations happens. There is something about the belief rather than the mechanism that condemns us to repetition. There is precious little observation in an ethnographic sense or with a Goethean spirit takes place in organisations. You might almost be forgiven for thinking that people really don’t want to know. And that is precisely Burke’s point. The history of organisations is that they are thought to be resistant to change!
In the School Circles film, resistance to change is not visible. Rather, what is clear is that everyone in the film thinks that they can get things organised in a more sympathetic way if they put in the time to bring up their ideas and support them through the process. They will either implement their ideas or learn, genuinely, why implementing them is not a good idea. I note however, that they do this within what we would have to call an architecture.[9] So we will have to limit ourselves to a tentative conclusion such as: some architectures produce stuckness beliefs and some architectures are much more freeing-up of change.
In the most dynamic organisations, ones where you can feel the potential, there is a different set of cultural assumptions about colleagues. They constitute the context in the spirit of Ubuntu: it takes a village to raise a child. Authority, either positional or earned, is not so much of an issue because there is a genuine diversity of thinking and feeling, all of which is respected as a potential source of wisdom. We don’t get the egos and the celebs and the bullies and the manipulators and the prima donnas because there is much less point or reward in being those things. Those ways of behaving are also roles that belong to an architecture that we don’t question.
It is a commonplace among consultants that when you find someone in an organisation you are consulting to who is particularly aware and insightful about what is going on, that person is probably about to leave. It is the stance of being half in and half out that allows them to see. They are part of the organisation and not part of it simultaneously. In this situation we can see the membership of the organisation produces a blindness towards the architecture and the beliefs about architecture that are embodied. This is the leverage conundrum. It seems to me that change is something that cannot be decided upon with any of the normal decision mechanisms, precisely because those mechanisms are already hostage: part of the problem.
The standard decision architecture is all about judging people. But judging people no matter how wrapped up in warm language is where all the standard projections happen. My failings (many!) appear as your faults. The question of what the dance looks like is hidden by the fact that I can see you spoiling the pattern.
There is a suitably weird anthroposophical discipline called eurythmy. It really is a discipline: a three-year course to learn it. In eurythmy you literally do an organisational dance and study the patterns. Sometimes there are some large wooden balls that get passed from hand to hand as part of the dance. In my first outing in a circle of fifty people, the balls flowed until there was a hiccup and the pattern broke. Instead of having one ball in a dynamic, pass it along sort of way, I had three. The same hiccup next time. But here’s the point. It was a breakdown of the pattern that was observed. Probably I messed up but it could well have been someone else. And it is about the maintenance of the dynamic pattern, not about me or anyone else having two left feet. At its height you can feel the pattern and the way it can only be maintain dynamically. Oh, and you get some poetry to establish the rhythm and the mood as you dance. And no, I am not an anthropof.
What do we need to observe?
The fully systemic insight is this. We need to observe, closely and in detail, without getting caught up in existing stories and post-hoc justifications of what happens. What do we need to observe? Well the patterns that maintain the system invariants, the things that remain the same when all else changes. With a nod to POSIWID, not only can no-one tell what those patterns are, but it is a safe bet that all your guesses will get derailed at every opportunity. My experience is that the closer you get to what the organisation actually does, the louder everyone screams at you and the more threatening they become. Which becomes a method of sorts!
[1] Pedantically, of course he didn’t say about how the organisation works, but how it communicates, the communication structures of the organisation. Combine that with McCullough’s law, and you constrain the decisions that can be seen/made, and from there it’s a mighty small hop to ‘work’ (or ‘not work’, as the case may more often be).
[2] Our water heater (a ‘geyser’) had its annual inspection yesterday and exhibited unusual behavior under test loads. The official line had to be that it was unsafe and that we’d been informed but had declined to let the certified gas professional turn it off. In actuality, he said that he was obliged to categorise it as unsafe and to set remedial wheels in motion, but that he’d leave it running and it’d be ok…
[3] In formal taxonomies, termites are not ants. But of course, those taxonomies are imposed on the world and no real beast is obliged to stay within its section of the tree… My favourite example is not the platypus, but those frogs from across a South American mountain range. The adjacent frogs can mate and breed (same species) but frogs from one end of the mountain range aren’t compatible with frogs from the other end (different species).
[4] The flip side is Chesterton’s Fence. If you don’t see the sense of an existing barrier, err on the side of observation rather than demolition.
[5] Barry Oshry uses a model he calls Tops, Middles, and Bottoms. Each of those positions has its stereotypical pathologies that create and exacerbate stress. (It’s not an org chart — we occupy all of these positions, albeit in different contexts).
[6] And McCullough, of course, what we can and cannot decide from where we stand.
[7] At the risk of introducing yet another Big Name™, this is akin to what John Boyd meant by the Orient phase of his OODA loop.
[8] It’s on Vimeo. https://vimeo.com/ondemand/schoolcircles/278922931
[9] Nick Obolensky’s work on the hard and soft factors needed for self-organising systems comes to mind. There is always an architecture.
