I hope this email finds you as well as can be expected. Amidst the uncertainty manifesting from the multitude of dynamics that are currently playing out across the world, many people seem to be struggling to make sense of what is happening. As paradoxical as this sounds, there is both an overload of information and a deficit of information, depending on where you do or don’t look, and to who you do or don’t listen.
And so tonight, now that the kids have gone to bed, I’ve decided to dedicate a few hours to seeing if I might be able help you weave together some coherence. Consider it a complexity practitioner’s perspective on current global events. Here goes…
Five years ago in my first book I referenced this article by Adam Garfinkle in the magazine American Interest. Titled ‘What’s Going On’, Garfinkle posed what he described as the ‘uber question of our time’: ‘What the heck is going on, anyway?’
“It is now banal in the extreme to say that we are living in a rapidly changing world, and it can be misleading, too. The challenge is to understand how the world is changing, not how fast it is changing”.
This is my job tonight, down here in this great southern land, to see if I can help you understand how (and why) everything is happening in the way that it is. Rather than focussing on what should or shouldn’t be done, I’d just like to help you make sense of what’s going on.
Because right now, the world’s largest ever en-masse sense making activity is taking place. More than seven billion people are currently trying to understand what the heck is going on.
To put it succinctly, what’s going on is the result of non-linear (i.e. multiplicative) dynamics in an incredibly connected, complex network. In this network, we (the global human population) are the nodes and the novel coronavirus (SARS CoV 2) is the pathogen that connects—or has the potential to connect—one node to another. (If you’d like a detailed, technical explanation of this, refer to Albert-Laszlo Barabasi’s appropriate chapter on spreading phenomena here.)
Of greatest alarm—but of no surprise to folk who understand the exponential growth function and the possibilities for complex networks to rapidly scale in size and speed by order of magnitude—is the current spread rate of the virus, which seems to be around R2-2.5 (this means that every individual infected will on average pass it on to another two to two and half people). Whilst this doesn’t sound like it’s very much, it explains why the number of cases double, and then double again, and then double again, and so on and so forth.
If you’re still grappling to get your head around the realities of exponential growth, this thought experiment might help.
Imagine Sydney Harbour has been drained of all of its of water, and you are tasked with refilling it using only a teaspoon. But the catch is, you get to fill it up at an exponential rate. This means that you first tip in one teaspoon, followed by two teaspoons, followed by four teaspoons, followed by 16 teaspoons—and so on and so forth.
If you were able to complete the task over the course of a week (no sleeping, mind you), and you started on Monday morning, frantically tipping in and tipping in, every observation from Monday morning through to Saturday afternoon would confirm that nothing is happening, because you wouldn’t see any water building up. In fact, you’d probably convince yourself that the whole pursuit is a waste of time. It’s only at some point on Saturday evening that you notice the very first signs of water pooling across the bottom of the harbour. And so you keep going. But even by Sunday lunchtime there still wouldn’t be much water in the harbour. And then it all happens at once. By Sunday night the Harbour is full. Welcome to the world of non-linear dynamics.
The real concern here is that many people mistake the early stages of multiplicative/exponential growth for additive/linear growth, and fail to truly consider how much things will change as the ‘knee’ or the ‘kick’ in the curve—the bit where it starts going upwards very quickly—is passed. Many people have been lulled into a false sense of security, just like the infamous turkey the night before Thanksgiving.
By now—at least if you’re a dedicated reader—you’re already familiar with the basic concept of networks, which I unpacked in all of their glory last time. Even if you’re not, you’ll still most-likely have a sense of how our lives—not only socially, but also biologically and technologically—are our networks. At every level, networks are at play. We are networks ourselves, and we belong to networks.
In recent times however, the technological and social networks to which we belong have become larger, more connected and denser. In addition, completely new technological and social networks have emerged. Not only this, but these networks have connected with many other networks, meaning we have a network of networks, forming an ecosystem of networks. These increases all mean one thing: our lives are irrevocably enmeshed in complex networks and ecosystems.
Many possibilities: good, bad, or whatever
This complexity brings with it many possibilities. Critically, these possibilities are not good and nor are they bad; they just are.
These possibilities include being able to follow one billion different Instagram users, being able to quickly search for and find nearly an infinite amount of information (such as how many Instagram users there are), and being able to travel in relatively quick time to anywhere on the globe. Again, these possibilities are essentially agnostic: they are what they are. They become good, bad, or whatever else it is they become only once they are given a context.
For example, one billion users of Instagram is great for Facebook (Instagram’s owners), but it sucks when you see the degradation of your favourite mountain hut occurring as a result of masses of Instagramers flocking to the next must-capture location. Likewise, information used with good intent can generative, but used with malice, it can be destructive. Thus, the many possibilities of complex networks are entirely contingent upon context.
Of critical significance to our current context of coronavirus pandemic is that the greater the number of nodes and connections in the network, the greater the number will be of possible outcomes. There’s no debating this, it’s just a mathematical property of networks (see Metcalfe’s Law).
The problem for us in this current context is that the greater the number of possible outcomes, the more difficult it becomes to successfully predict which specific outcome will eventuate. This means there are two prices to pay for the benefits of having a vast number of upside possibilities to entertain: the first being that there is the same number of downsides (remember, every possibility is agnostic, and the upside or downside only manifests once the possibilities are contextualised), and the second being that you can’t confidently predict which possibility will eventuate, which leads to chronic uncertainty.
In terms of our current context i.e. SARS CoV 2 and the value we place on human life, we now face the stark reality of a rapidly changing context. Up until a few months so ago many of us revelled in the upside opportunity that cheap air travel and flexible border controls affords. But now we must face the downside possibility emerging from this ecosystem of networks in which we live our lives: a global pandemic that results in many deaths and a protracted economic slow-down.
Globalisation has played a critical role here: by opening borders to free trade, globalisation has dramatically increased global complexity in a very short period of time. This increased complexity has been turbocharged by many more factors, including but not limited to the internet, social media, and the aforementioned affordable travel and cross-border movement. Adding fuel to the turbo is Francis Fukuyama’s End of History and the ascension of western neoliberal democracy, with its strong beliefs about individual freedoms. (The significance of our beliefs about the individual will become apparent when we discuss what we need to do to survive this crisis.)
Critically, and somewhat paradoxically, as every day goes by, the uncertainty of this significant downside possibility decreases. In other words, as every day goes by and the pandemic’s intensity increases, the number of possibilities decrease and the likelihood that this global pandemic does result in both many deaths and a protracted economic slow-down increases. Thus, within the over-arching context of chronic uncertainty, certainty begins to emerge. But it’s a certainty we most certainly want to avoid, at all costs.
This is why a coordinated global response to mitigate the spread of SARS CoV 2 is critical at this very moment. The reality of being part of a large, densely connected system is that if you desire to maintain the integrity of that system, everyone must act at the individual level as if they could—just by themselves, in their lower order, localised space—protect the system.
Borrowing from the work of Michael Puett and Confucian philosophy, subjunctive ritual allows us to break the patterns of individualism that we have all fallen into and to act in the interest of the higher order of the system that we all belong to. The subjunctive action—also referred to as acting otherwise—that we must all do to protect our system is social distancing. There are still way too many people out there at the moment, at least here in Australia, who don’t appreciate the gravity of the situation.
If only half or two-thirds of people in communities and nations of the world act as if, the system will not be adequately protected, and containment efforts will fail. This is basically what happened when the lockdown of the Lombardy region was announced on March 8: not everyone stayed in, and many people fled the region to return to their families in other parts of the country. This is why the entire country was ordered into lockdown the following day. Not everyone acted as if.
In complexity speak, this is what we refer to as lower-order activity, or a fine (as opposed to coarse) level of granularity. It is through the lower order—each of us acting as if—that the higher order emerges. Arnie did a great job of this in his gorgeous and wonderfully human (and equine) tweet from yesterday. He was acting as if. He was acknowledging where he has agency to act, and where you have agency to act, too.
In this context, the higher order that emerges from the individual acting as if in their lower order is simple: survival through this crisis. If you think you can’t do it, you’re wrong. It might be uncomfortable, and you probably don’t have a miniature horse named Whiskey and a miniature donkey named Lulu to cuddle, but that’s the sacrifice you have to make. We all have to make this sacrifice, for an unknown period of time.
In short, each individual person, each individual household, each individual community, and each individual nation must act as if they could—just by themselves in their lower order, localised context—protect the system. This must occur now, in this critical phase, during peak uncertainty. Everyone must suspend their individual needs and wants, and act for the betterment of our system.
To further increase our understanding of this situation we must briefly delve into a world that very few do. I’ve written about the Gaussian and Paretian worlds we inhabit on numerous occasions—most recently here—but to really understand what’s going on at the moment, you must get your head around this.
In that piece I wrote about the problems associated with working in uncertain spaces. I described these spaces as problem spaces, and noted that the uncertain nature of them means that you either cannot be aware of all the variables in the problem space, and even if you can, calculating the number of possible outcomes from the interactions between the variables becomes impossible. It is for these reasons that the uncertainty is so chronic.
Think of our current coronavirus pandemic as a problem space where there are any number of variables—namely ones relating to individual, community and societal scales—which are either not yet understood, or are changing so rapidly that they never can be understand (until after the fact). Yes, this is a reality of the connected, complex networked world we live in: the variables themselves interact and influence each other, which means goodbye to our ability to fully understand our problem space.
It was also in that last piece that I introduced to you the notion of probability distributions (of problem spaces), and the different types of distributions that exist. Whilst there are a number of different distribution types—and noting that I studied geography and cultural anthropology, and not math—I simplified these into two key types: Gaussian distributions and Paretian distributions, with the former named after German mathematician Carl Friedrich Gauss and the latter Italian economist Vilfred Pareto.
The distinction between these two distributional types has been popularised by Nassim Nicholas Taleb’s metaphorical worlds of Mediocristan and Extremistan, with Medicoristan representing the Gaussian world and Extremistan representing the Paretian world. According to Taleb in The Black Swan, the key distinction between the two worlds is as follows:
“Mediocristan has a lot of variations, not a single one of which is extreme; Extremistan has few variations, but those that take place are extreme”.
Mediocristan vs Extremistan
non-scalable vs scalable
mild randomness vs extreme randomness
typical member is mediocre vs no typical member
winner gets a small slice vs winner takes all
historical vs modern
subject to dampening vs subject to acceleration
physical vs informational
many small events vs a few huge events
easy to predict vs hard to predict
history crawls vs history jumps
Gaussian distributions vs Pareto distributions
As you scan these distinctions, my hunch is that you’re now just beginning to realise that the pandemic situation we currently find ourselves in is consistent with the exemplars associated with Extremistan on the right-hand side of the list.
Our current coronavirus context places us squarely in the most uncertain depths of Extremistan.
Not surprisingly, Taleb was onto it in the earlier stages of the viral outbreak, writing this note from the New England Complex Systems Institute with his colleagues Joseph Norman and Yaneer Bar-Yam on January 26, 2020. The authors make the following points:
“Clearly, we are dealing with an extreme fat-tailed process owing to an increased connectivity, which increases the spreading in a non linear way. Fat tailed processes have special attributes, making conventional risk-management approaches inadequate”.
The authors go on to reference the General Precautionary Principle, as follows:
“The general (non-naive) precautionary principle delineates conditions where actions must be taken to reduce the risk of ruin, and traditional cost-benefit analyses must not be used. These are ruin problems where, over time, exposure to tail events leads to a certain eventual extinction. While there is a high probability for humanity surviving a single such event, over time, there is eventually zero probability of surviving repeated exposure to such events. While repeated risks can be taken by individuals with a limited life expectancy, ruin exposures must never be taken at the systemic and collective level. In technical terms, the precautionary principle applies when traditional statistical averages are invalid because risks are not ergodic”.
In short, what the authors are saying—and what the network and complexity sciences confirm—is that delaying and not coordinating a global response to this pandemic gives much greater possibility to it resulting in many deaths and a protracted, significant, economic slow-down.
But let me give it to you straight from their mouths (so to speak):
“Standard individual-scale policy approaches such as isolation, contact tracing and monitoring are rapidly (computationally) overwhelmed in the face of mass infection, and thus also cannot be relied upon to stop a pandemic. Multi scale population approaches including drastically pruning contact networks using collective boundaries and social behaviour change, and community self-monitoring, are essential.
Together, these observations lead to the necessity of a precautionary approach to current and potential pandemic outbreaks that must include constraining mobility patterns in the early stages of an outbreak, especially when little is known about the true parameters of the pathogen.
It will cost something to reduce mobility in the short term, but to fail to do so will eventually cost everything—if not from this event, then one in the future. Outbreaks are inevitable, but an appropriately precautionary response can mitigate systemic risk to the globe at large. But policy- and decision-makers must act swiftly and avoid the fallacy that to have an appropriate respect for uncertainty in the face of a possible irreversible catastrophe amounts to “paranoia”, or the converse a belief that nothing can be done”.
It is my very real concern, as it is with many other folk, that many governments have fallen and indeed continue to fall victim to the fallacy that respecting uncertainty amounts to paranoia, and that it must be avoided at all costs.
From my perspective, the Australian response in the past fortnight has also become—admittedly after a very promising start—a victim to this fallacy.
If my argument still fails the cogency test, consider this very recent example from 28 February 2020 of this fallacy in play regarding the pandemic. It’s written by Cass Sunstein, a highly influential US academic who is well-known for both his behavioural economics book Nudge (co-written with Richard Thaler) and for his role with the Obama administration. Sunstein refers to ‘probability neglect’—which is essentially the propensity for an individual to focus on the outcomes of a significant event, but not on the likelihood of that event occurring —and says of people and their concern over the coronavirus:
“A lot of people are more scared than they have any reason to be. They have an exaggerated sense of their own personal risk”.
However, Sunstein is wrong, wrong, wrong.
Just to clarify, behavioural economics is not necessarily bad per se (although some clever people I know think it is), rather, in the context of highly connected, complex systems exhibiting rapid non-linear dynamics, its application poses asymmetrical risks which can amplify exponentially, and could overwhelm a country’s healthcare system.
And somewhat scarily, it seems that the whole ‘herd immunity’ play has come out of the work of the British government’s ‘nudge unit’. For a while there, it seemed that down here in Australia we were also on a similar trajectory (both the federal and many state governments have behavioural economics teams), although it’s hard to tell exactly what’s going on here at the moment. (The whole herd immunity play is what complexity practitioners would refer to as an experiment that is not ‘safe-to-fail’, or, if you prefer swear words, a potentially massive clusterf@ck of epic proportions.)
If I have still not managed to shine some light on the benefits of common-sense during this pandemic, please check out this article from Rutgers University’s Harry Crane.
I’m paraphrasing Crane here:
“First, survive: prepare for the worst case scenario by considering all possibilities.
Second, protect: consider the most plausible bad-case scenarios within the realm of possibilities, and hedge against as many as are reasonable.
Third: remember that when it comes to making decisions under uncertainty, being accurate is the absolute last thing one should care about”.
Or Taleb, again from only three days ago:
“Collective safety may require excessive individual risk avoidance, even if it conflicts with an individual’s own interest and benefits. It may require an individual to worry about risks that are comparatively insignificant… Hence one must ‘panic’ individually (i.e., produce what seems to be an exaggerated response) in order to avoid systemic problems, even where the immediate individual payoff does not appear to warrant it”.
This is probably the best thing I can leave you with; it’s a simple heuristic that can scaffold you through each day as you try and make sense of this seemingly never-ending unfolding of dynamics:
Remember that what will feel stupid today will not feel stupid tomorrow.
I hope it gives you all the courage that we need right now—despite how stupid you might feel—to act as if. You’re doing it for the greater good.
A long day’s night
After a full day of work yesterday, I started writing at 7:30 pm. It’s now 6:14 am. I’ll send this off to Amelia shortly, and she’ll post it soon enough. I’m feeling tired, and my first conference call for the day is in two hour’s time.
But I know there are so many people around the world right now—especially the frontline healthcare workers—who are infinitely more tired than I am, so it doesn’t matter. We all must act as if. This is but one of my contributions to acting as if.
Distracted, and admittedly a little frustrated, I now find myself scrolling through photos on my phone of my two kids. My son is turning two in a week’s time, and my daughter seven a few days after that. They both have blonde hair. They’ll be awake in an hour or two. I adore them.
Remember, if we can’t work together in a crisis, if we can’t act as if for the greater good, then there’s no point to any of this. Not my writing, I mean, but to life. There’s no point.
So we need to work together. We will recover from this, but we will immediately face new uncertainties: decarbonising our economies over the next decade is a herculean task compared to this current crisis. Having had first hand experience of the bushfires down here over the past summer, I can assure you that an ever-warming climate and its consequences is no fun. There’s only misery if we keep going down that path. If we don’t get through this current crisis, we don’t stand a chance for what will be a much bigger test. We can do it.
I write to you all with much love.