Some kind of Muscovy Duck — the closest picture I have to a Black Swan! The Pantanal, Brazil, 2006
One of the interesting concepts in understanding metacognition is the idea of a ‘Black Swan’ — symbol of a theory developed by Nassim Nicholas Taleb, that describes a Black Swan event as (from Wikipedia) an occurrence or design that comes as a surprise, has a major effect, and is often inappropriately rationalized after the fact with the benefit of hindsight. The rise of the Internet is considered a prime example. Black Swan events come from the notion of the ‘unknown unknown’ part of metacognition. Ostensibly, you don’t know they’re out there, and you wouldn’t necessarily know what you were looking for it, even if you were looking.
Taleb’s a smart guy, and it’s a great idea. He is not only a professor,but a financial guru who has been a trader himself, and so has developed a whole theory around Black Swans, which he’s written about in an eponymous book. Taleb’s theories revolve mostly around probabilistic representations of Black Swans, and probability is a good way to understand them — especially when they come from the outside. Yet any Black Swan has to be, at some level, a created object — designed, so to speak — and the dynamics of how something is created that one has little knowledge of, as we’ve covered in this blog, can be very complex. But it can be understood.
There’s also a way of understanding Black Swans in the empathetic evolution space, though. I’ve already talked about Chris Voss’ excellent book, Never Split the Difference, on negotiating. In his world, a Black Swan is a idea or insight inside a negotiation that leads to paradigm-changing advantage and resolution of a problem, coming from Voss’ notion of Tactical Empathy. Regardless, here’s the takeaway — it’s still a designed solution, and as such, has to follow the rules I’m going to lay out below.
So let’s start with a definition that utilizes what we know about empathy, and empathetic relational modes. A Black Swan is a manifestation of a design solution that was not considered or known about inside a larger, general solution space that profoundly changes our mental models regarding those families of solutions. That means that Black Swans have the potential to be major empathy ladders, as they disrupt belief systems and force organizations to modify their social structures, create more duplex information paths, and become more data driven.
In order to give some kind of graphical representation to all of this, let’s consider how different social structures cover design spaces, and then we’ll have some idea of how Black Swans work. Below is a plot of solution spaces for a very siloed company.
Let’s say this is a company staffed by Mechanical Engineers (MEs), represented by the X axis, and Electrical Engineers (EEs), represented by the Y axis.
The MEs in this company typically generate a range of solutions with paradigms that they’re educated with — namely things like heat, mechanical motion, fluid motion, and so on. (Thermodynamics, Dynamics, Fluid Dynamics, etc.) EEs have a different set of tricks up their sleeve — electricity, magnetism, microprocessors, integrated circuits (ICs). On and on. A customer approaches this company for a solution, and either it’s ‘All ME, all the time’ or ‘All EE, all the time.’
The best that such a company can provide is an additive solution — a EE control system laid on top of an ME primary solution, for example. This was the state of the world when I became an engineer back in 1982 in the steel mills of Cleveland. We had an ME solution — a rolling mill with handscrews at the top. And we worked with an outside firm using computers to run the electric motors to turn the handscrews. The mill itself was old — probably 20 to 30 years old, and had originally had people turning handscrews on the top. The consulting company’s job was to retrofit those drives (which had by that time already been replaced with electric motors) with computer control.
Obviously, this is a pretty non-empathetic paradigm. There’s no desire, with this kind of organizational structure, for MEs and EEs to trade information. There are strong In-group/Out-group dynamics, and that’s decided by your title and degree. Any completed system will be, at best, an additive manifestation of an EE and an ME part– what I call functionally ‘meta-linear’. But just because the EEs and MEs don’t talk, it doesn’t mean there aren’t buried, unplanned-for synergies inside the final design. And these effects may be potentially negative. Perhaps the material selected by the MEs messes magnetically with the ICs picked by the EEs. Those synergies are also potentially nonlinear, and will only be discovered by running the solution in its environment, through a range of operating conditions. And when we have those buried synergies, what can pop up? A Black Swan! Something that, upon failure of the system, people will look in hindsight and say “boy, we should have anticipated that!”
On top of that, we have a whole big hunk of design space that, because of the limited awareness of our potential solutions, remains uncharted. We can’t even know what’s out there because we have no people helping us peer around and tell us. That’s a giant lake for a Black Swan that might be spawned by a competitor. So we have two avenues for the genesis of such a bird: the first are the potentially negative, likely nonlinear synergies coming from meta-nonlinear actions inside our design space, driven by lack of awareness; and second, the probabilistic actions from companies and phenomena outside our design space. Not good.
Let’s modify our company a little, and give some overlap in knowledge between our EEs and MEs. This is functionally equivalent to what we do now in engineering education — have students take classes where they get both a smidgen of information and insight into other disciplines. We call this type of experience ‘multidisciplinary’, which, at some level, it is. But it is not particularly well-integrated.
Now our MEs know something about EE world, and the EEs also know something about the constraints faced by MEs. While far from perfect, this has the advantage of eliminating at least some of the negative synergies that plague the design space above. Hopefully, because of explicit knowledge declaration, the MEs know not to build certain things out of magnetic materials that might mess up the EEs’ circuits. And the EEs have an understanding of how MEs might make things so that they don’t ground a circuit on something where it might be designed to break so that one doesn’t see catastrophic failure. We’re still not seeing any particular positive synergies in design, because the MEs and EEs are only peripherally talking to each other. We still have meta-linear designs, because we still have an Authoritarian/Legalistic social structure — our MEs and EEs are all hanging off the General Manager’s primary node. Any integration is going to happen by the person at the top of the org. chart.
We still have our large and empty Design Solution lake that external Black Swans are swimming in. They’re out there, and we’re blissfully unaware that the big quacker might be coming to bite us in the butt. Until he does, of course.
Now let’s introduce some higher-level rational empathy into our social structure, between our MEs and EEs. That yields a design space map that looks like this:
With our Big Blue Empathy Arrow, we’ve added duplex interchange between our EEs and MEs. Information is now flowing freely back and forth between both parties in the organization. This duplex information is not constrained to just a one-off back-and-forth. Instead, we end up with a system that generates shared coherences between the two groups — but not so much on a predictable timescale. We have transients of indeterminate length to arrive at shared solutions. The engineers talk back and forth until they agree — that’s the standard. Not once, not twice, but enough to reach the goal of information coherence. In the process, both sides become much more educated on consequences of the others’ actions. And through mapping of paradigms from one group to another, we have the potential for a truly nonlinear breakthrough — our very own Black Swan design. A Black Swan design may be a budding-off of an old paradigm, or an entirely new jump in performance.
With empathy, and its indeterminate transients and changing social structure not only composed of a formal hierarchy, but also independently generated relationships between actors in our organization, we have now created a meta-nonlinear system. Right off the mark, our organization is creating organizational conditions for multi-solution, blended designs.
This meta-nonlinear system’s behavior can be understood with what mathematicians call bifurcation theory. In the case of budding, it’s the potential result of a pitchfork bifurcation. In the case of a performance jump, it’s a Hopf bifurcation. No longer just in the realm of probability theory, we have a deterministic map that gives insights into how our Black Swan came into being. It’s still very difficult to predict — even with simple nonlinear equations, mathematicians still struggle with coming up with useful estimations of when these phenomena occur. But these mental models can facilitate understanding exactly how multi-solution, breakthrough thinking manifests itself.
Well-known examples of the first of the two types might be the iPod. Look at the family diversity involved in that exercise. And a jump in performance? How about the Tesla, firing all wheels with their own motors. By eliminating the heavy drivetrain and engine inside the vehicle, you get performance that looks like the video below:
We’re still not out of the woods with those probabilistic Black Swans on the outside. They’re still swimming around in the big white Design Solution lake that our own, empathetic, integrated team may start to be aware of. But we’ve got a start, because now our company is starting to populate the Unknown Unknown part of the design space with our own ideas, claiming the space and making it known.
How do we manage the external Black Swan part of the lake? Let’s take one larger step into interconnected networks.
We still primarily staff our company with EEs and MEs. There’s some overlap in training between our two specialties. But what we’ve added is now an assortment of constituencies that also fill the Design Space. It’s hard to convey on a 2-D plot, but we don’t even need our other parts of the network to be engineers. In fact, we’d like it if they weren’t. It might be nice if they understood the laws of physics. But that’s not even necessary. The laws of physics constrain the larger space for all designs (the Laws of Physics, after all, are The LAW!!) but any design engineer will tell you that it’s often not the case that a successful (or unsuccessful) design was constrained by the Second Law of Thermodynamics. It might be that you were trying to sell an item with a color that was out of style. Or you named your car ‘Nova’ (no va) in a Spanish-speaking market.
If we populate that space with specialists up and down the Spiral, we add cognitive diversity to our expanded network. And as far as guiding ourselves in how we pick constituent groups for our expanded network, it’s all about appropriate scaffolding. Topical information, delivered by subject matter experts, matters, of course. But there also needs to be a focus on picking different thinking modes. Government regulators might supply some legalistic V-meme information. Non-profit groups might give insight into potential rising concerns inside that community, as well as lifecycle pathways to future sustainability challenges. Style mavens might fill in the arbitrary Authoritarian niche, identifying trend-setting celebrities and give some insight into whether straight lines or curves are going to be de rigueur in the next design season. Customers, through direct interviews, focus groups, and beta testing, are going to give us valuable feedback on actual use.
Slowly but surely, our Design Space fills up. And the robustness of that ‘filling in the lake’ is directly related to the level of empathetic evolution and its direct consequence –appropriate trust –given each of the constituencies. If we evolve our own company to be self-aware, (like Clint Eastwood said, “a man’s got to know his limitations,”) we know how much those networks know, and the level of certainty that they know stuff. Through that network, we can now come to terms with our “known knowns”, and count on them for our “known unknowns.” It’s impossible to ever completely map out our “unknown unknowns” out of existence, which is where those potentially negative probabilistic Black Swans reside. But with our complex network that fills the Design Solution space, we’re definitely denying them habitat.
The organization we’ve described has also been talked about before — the Wise Organization. By creating a large, empathetically connected, duplex network of diverse stakeholders, with a core of self-aware professionals, we continue to hone what we know, and what we don’t know. And over time, that creates the basis for your organization to create its own breakthrough designs.
Takeaway: Black Swans are avoided with robust, interconnected design communities that focus on diversity of all kinds. The added benefit is that such communities are much more likely to generate breakthrough designs themselves, creating Black Swans with negative consequences for their competitors!
This is also worth a read — this short article by Taleb himself regarding Ten Principles for a Black Swan-proof World. There is just a ton of wisdom in this, and worthy of how the social physics discussed in this blog directly map to Taleb’s insights. I just love it.