Agile through the eyes of a Biologist
Some interesting parallels I’ve noticed between Agile and nature.
I have been a Scrum Master for 10 years now, but when I was young, I studied Biology at Utrecht University. The highpoint of my studies was working with Orangutans in the forests of Sumatra, where I spent about 7 months in a remote research camp following the Orangutans from sunrise to sunset, in one of the most beautiful jungles left in the world.
Those were long days of solitude, immersed in the mindfulness of nature, and in silent communion with a solitary Orangutan, peacefully going about her daily routine: looking for fruit trees, eating, everything in slow motion. And a lot of sleeping.
Until a given tree, usually a fig tree, suddenly produced millions of fruit, allowing multiple Orangutans to gather together to feed at that one tree. In those circumstances, the Orangutan suddenly becomes a social animal, with individuals interacting extensively with each other.
Fights, sex, children playing together. I found it mesmerising.
That’s how I discovered that social organisation in apes is determined by environmental factors. You can literally explain the way apes will organise themselves by the place they live.
Social systems
Orangutans live in an environment where there is a lack of food. They have to range far and wide to collect their daily food. There is just not enough food to allow Orangutans to form social groups. So they live solitary lives. Until every now and then that one tree comes along.
On the other side of the world, in Ethiopia, Geladas form huge groups that spend their days grazing on vast grassy savannas, almost the way herds of antelopes graze in other parts of Africa. Yes, you read that correctly, they graze. Their main food staple is grasses. Again, a result of their environment. Because they graze, food is ample and there is more than enough to allow big groups to form.
Sticking together helps them to deal with predators. As does the fact they live in open spaces, where many eyes easily keep a watch of their surroundings. But more importantly for their social organisation, come evenings, Gelada retreat from the grassy savannas to spend their nights on cliffs, safe from predators.
The cliffs offer a limited amount of room and the result is that hierarchies form as individuals fight to secure the safest places on the cliff, those in the steepest parts and farthest away from the edges. The cliffs also guarantee group formation, as individuals will all be competing for space at the same limited places.
And so on.
But what does all this have to do with Agile?
To me, first of all, it reflects very much the Cynefin framework which helps us determine which Agile methodology best fits a specific situation. Just like with our apes and their environment, it is the environment of the organisation that will determine which methodology is best fitted to successfully applying Agile in the organisation. That is the reason Scrum is presented as a framework, because it must always be tweaked to fit a given team.
With this in mind, it is an interesting idea to look at an organisation the way an ecologist looks at an ecosystem. An ecologist will model the ecosystem step by step, continuously probing, collecting data and testing and tweaking the model, gradually making sense of more and more of it. This is the way to manage the complexity of the system.
When introducing Agile to an organisation we are dealing with similar complexity, so it makes sense to use a similar approach. To me, this also corroborates the idea that applying a readymade scaling framework is a rather obviously over-optimistic approach.
In this sense, ecology is system thinking applied to a biological setting, in which empiricism drives the way a model of reality slowly evolves. It is an evolution.
Evolution
Of course, there is no way I could skip mentioning evolution and Darwin’s theory of adaptation.
Evolution is nature’s version of empiricism. Small changes (adaptations) are released into the wild to see if they are viable. Individuals with viable adaptations survive and get to pass on those changes to their offspring. The rest die out. Over millions of years, we see the world change.
It all sounds very neat, but it is actually total chaos. The direction of evolution over millions of years may be clear, but at any given moment it is a mess of accidents, coincidences and lucky breaks. This is because, in nature, variation is purely random and survival callously dispassionate. You may have the best adaptations available but you may still be the hapless one to be hit by random lightning, meaning that all those wonderful adaptations die with you. Think dinosaurs and meteorites here. Or you may be a completely hopeless species but living in safe isolation, allowing you to survive. Think dodos here.
Considering all this, it is amazing how evolution has led to all that we see in nature today. To me, it underlines how amazingly effective it is.
The truth is that empiricism is a shortcut for intelligence. Let me explain. If I play chess on my phone my concentration is often minimal. To win, all I have to do is undo my moves until I get the right one that finally leads me to checkmate. This is empiricism. In this case rather voluntary stupidity, but I think you get the drift?
In that sense, it is comforting to realise that evolution similarly is powered by mistakes. Variation arises mainly from mistakes made when genes are copied. Without these mistakes, there would be no evolution.
Evolution is not terribly viable for Agile and the challenges we face applying it, as it takes so very long. Fortunately, in our case, as human beings, our advantage is that we can direct empiricism.
For Agile, with its breathtaking pragmatism, empiricism is the ultimate way to control complexity. It allows us to accept our limitations, to accept that there is a limit to the amount of complexity we can manage. With empiricism, we can fall back to our strengths: our playful, unconcerned state of mind, in which we find flow and our highest levels of creativity, knowing that we can go back and pick the best results. It leads us to things like evidence-based management, experimentation, double-loop learning, the things we discuss in our retrospectives. It is the reason we value failing fast so much.
To me as a biologist Scrum Master, everything happens in the context of evolution. The evolution of my team. The evolution of the team within the evolution of the company. My own evolution within the evolution of the team within the evolution of the company.
Yes, it can get a bit complicated, but being aware of that time dimension is an essential and solid point of reference. Being able to visualise the flow of things in time is crucial. Empiricism is not possible without a sense of time. Only those who know their past can consciously grow into something new. And avoid the mistakes of the past. System thinking becomes second nature.
I sometimes imagine a god at the beginning of creation, drawing-board and ruler in hand, designing the whole of existence. The whole project, again and again, sinks in its own complexity. It keeps running out of his hands, like water. After repeated failures, he suddenly has an idea and thinks “oh, but what if I figure some simple rules (genetics, gravity, speed of light) and a simple process (time) and sit back and enjoy the show?
Simple rules
Back in the 1930’s scientists studying birds were so amazed by the way some bird species were able to form fascinating moving shapes when they flock together, that they suggested the birds must have psychic powers to be able to create those patterns.
The scientific name for this phenomenon is murmurations. Craig Reynolds (who I’m convinced should have been a Biologist) managed to simulate these patterns by creating digital algorithms that used only 3 simple rules:
Separation — avoid crowding neighbours,
Alignment — go the same way as your neighbours,
Cohesion — try to keep the same distance from your neighbours.
The key here is that the individual birds are autonomous, all busy with their own goals. The 3 rules are constraints that simply allow them to pursue those goals, dare I say it, as a team. What the purpose is science hasn’t been able to prove definitely yet. But one thing is made clear:
Simple rules lead to complex behaviour
For us today, especially amongst Agilists, this is a concept that we are very familiar with. But I like to imagine computer scientists before Craig, trying to somehow create a program that would micromanage every single bird in an attempt to create those patterns. I am struck by the idea that that is precisely what we try to do with Waterfall, as we try to prepare and control a complete project upfront.
Another beautiful example of this kind of organisation is ants and bees, which form huge complex colonies in which individuals function autonomously but are directed by a queen. How do they do that? The queen is sequestered somewhere in a hole while her subjects are off experiencing all the challenges of life, so how does the queen know how to guide them?
Biologists have discovered that the queen gathers information from her subjects, allowing her to oversee the whole. And she is able to influence her subjects by secreting pheromones, basically simple chemical messages. The queen’s pheromones can be seen as simple rules, which she uses to elicit and orchestrate complex behaviour in the subjects.
At the same time, and this is where it gets interesting, the worker ants and bees are not dumb automatons, but proper, independent individuals, able to deal creatively with any situation they run into. So the queen provides the high-level vision and her subjects take care of shaping it in the context of their specific situation.
Simple rules lead to complex behaviour.
Such a simple principle and yet so powerful. So far removed from our process-driven industrial heritage, yet seemingly so… natural.
To me, this principle represents the huge challenge we face with Agile. That of stepping away from what comes naturally to us; control & command and the idea you can design a whole process; and embracing a scary and completely new way of thinking about things, an organic and counter-intuitive way.
It is in fact one of my favourite things about Scrum: the focus on self-management, as opposed to good old command & control. Usually understood in the context of the team, to me, self-management is a fundamental element for any form of organisation. Ideally, Agile leaders provide vision and a few well thought-out simple rules, leaving the rest to the people involved.
This is sufficient to lead to that elusive balance between alignment and autonomy.
Which brings me to my next subject, distributed control.
Distributed control
Those same simple rules we were talking about, form the basis for another kind of counter-intuitive organisation that we see throughout nature: that of distributed control.
A striking example of the phenomenon of distributed control can be found in jellyfish, in which it happens in a single organism. A jellyfish doesn’t have a central brain. Instead, different parts are directed by separate small independent brains. Unlike the ant and bee colonies, but similar to murmurations, there isn’t a single higher authority brain to orchestrate. What you have are separate areas of control collaborating with each other to manage the organism as a whole.
To me, this reminds me so very much of the kind of collective intelligence we are trying to achieve when we scale organisations. It suggests that if we would give teams more room, they might be able to manage much more than just themselves.
It also suggests that maybe we take scaling too far. I mean, if even a centralised control structure might be unnecessary, how much more simple would that make scaling?
Maybe the main challenge with scaling is not scaling itself, but the fact that we lose focus of the problem we are trying to solve: Scaling is about the alignment of empowered autonomy. Managing dependencies between self-managing units. Period.
Everything else is just our very human selves trying to placate our fears of losing control by keeping our processes and structure. Our stubborn inability to abandon our linear way of thinking. The kind of thinking which leads us to huge process-heavy systems, like SAFe.
These thoughts are not just some kind of utopia. There are organisations that are already experimenting with these kinds of distributed control.
Take Haier, a global international corporation, which introduced the concept of entrepreneurial microenterprise, basically breaking up the whole corporation into 4000+ mini-enterprises and eliminating 10000+ middle management positions. Each mini-enterprise is completely independent and self-managing, receiving only minimal direction from a central management. Today Haier is the world’s number one appliance maker and one of the most innovative organisations in the world.
Or Morning Star, the world's largest tomatoes processor where they basically have no management.
If you find these examples inspiring and would like to know more, I would definitely recommend reading Humanocracy. I got these two examples from there.
However far you want to take it, I think the important thing to realise is that just like with Agile, it is the organisations that adopt these new ideas first that are going to have a massive advantage. It is not a choice, it is the new competitive baseline. The next step in evolution.
Conclusion
In this piece, I have described ideas and concepts we are familiar with. From social organisation and evolution to simple rules and distributed control, nature shows us different ways to organise ourselves to better manage complexity. In an organic way, without control.
That lack of control, that is the common thread in this article, and possibly the main challenge we face with Agile.
It is ironic to think that in Biology we faced a similar struggle, in which for more than 200 years two camps fought each other ruthlessly. On the one side the camp for intelligent design, which argued that the complexity of nature could only have been made possible by an intelligent designer. On the other side the camp for evolution, championed amongst others by one of my favourite writers, Richard Dawkins. Maybe there is something for us to learn there?
I hope to have stirred your interest in this kind of cross-pollination of different kinds of knowledge. What other ideas and connections are still out there, waiting to be discovered and inspire us into new directions?
To me, it is a vindication of my indulgence in reading an eclectic variety of subjects that often seem completely irrelevant to the matter at hand.