The Failure of Simplistic Solutions to Complex COVID-19 Problems

Bifurcation behaviors of synchronized regions in logistic map ...

At the risk of sounding curmudgeonly, I’d like to take a few moments to address the current state of the commentary surrounding the COVID-19 pandemic.  Of course, I am thoroughly unsurprised by the absolute hash that the American government has managed to make of its response to the virus.  Indeed, in light of what I said earlier, we shouldn’t be astonished that this happened.  The virus would have been fairly easy to deal with in a socially cohesive society that was interested in actually trying to solve the problem, as you had in many East Asian countries, for example. In a society on the verge of social disintegration as American society is, and one in which literally everything becomes a lightning rod for political polarisation, a unified response to the crisis was almost impossible.

As someone who has tried to make a genuine, good faith effort since early December to set aside factionalism and actually let science and data (rather than SCIENCE!) guide my thinking about this virus, seeing the clown show into which the response to it has devolved has grown increasingly galling.  And again, at the peril of seeming like a sorehead, I think that both “sides” of the virus, ahem, discussion are off-base.  Both the progressive Left with its chicken-little-lockdowns-for-ten-years-it’s-all-Trump’s-fault-wear-your-mask-or-you’re-a-murderer karenmongering and the libertarian it’s-just-the-flu-the-virus-is-just-a-secret-plot-to-take-your-freedoms caterwauling are guilty of flooding the internet with bad takes. A lot of us who have been following the development of the virus ever since it was a wee little twinkle in the Wuhan Institute of Virology’s eye know this, however.  What’s not so universally known is why nearly everybody who’s been commenting is wrong.

What it all comes down to is the terminal lack of understanding about complex systems on the part of America’s (and the world’s, for that matter) commentariat.

Now, by complex systems, I don’t mean simply systems that have a lot of moving parts or that are difficult to understand.  Rather, I’m talking about complexity in the sense of dynamical systems in which a change in one variable can result in non-linear transformations of one or more interconnected parts of the system. Most people are used to thinking in linear terms – change one variable, and another changes by a constant amount.  You draw a line on a graph and as the x-axis changes, the y-axis changes according to m, the slope, which is the constant in the line equation.  You gradually increase the force applied to an object and it accelerates to a new velocity according to Newton’s second law of motion.  That sort of thing.  The Newtonian mindset is the typical way of thinking for the average person in their daily life.  However, in non-linear systems, there is not a linear response to inputs.  Instead, you may have positive or negative feedbacks which can lead to enhancement or dampening of the responses on other parts of the system.  Sometimes a system may respond in ways that act to stabilise it, pushing it towards an equilibrium point and reducing the effects of perturbations.  In other cases, the system may respond by increasing system instability, possibly leading to mathematically chaotic behaviour. Typically, these non-linear responses are not predictable, or at least fully so.  Complex systems may even display emergent properties – properties that occur in the system as a whole but which are not present in the individual parts of the system.  Complex systems are not only found in physics, but can describe systems transcending any number of fields from biology to economics to social behaviour.

The course of this pandemic is perfectly suited for analysis as a complex system.  There are so many interacting variables whose behaviour we cannot predict when other parts of the system are perturbed.  What affects the growth and spread of infections?  Does the mutation rate of the virus matter? What about the weather (hot climate or cold, as well as exposure to UV light in sunlight)? What about different genetic susceptibilities among different populations?  What about local population density? Do cultural practices regarding personal space matter? How about differing morbidities across different strains of the virus? What about the overall health level (i.e. pre-existing co-morbidities) of a population? What about any of dozens of other questions that could be asked?  The answer is yes. To all of them. All of these will have some kind of an effect on the local course of the spread and seriousness of infections in a population. And for none of them are we able to directly measure a linear response and make certain predictions.

One example that we can use to make this point is the effectiveness, or lack thereof, of the lockdowns – probably the single feature of this whole process that has gathered the most attention. A lot of hay has been made about the failures of the initial models that showed much higher numbers of deaths than we’ve seen so far.  From this, it is concluded, lockdowns and other remediative efforts were all just a waste of time and served only to destroy the economy.  All along, millions of people were never going to die!  Bring us the heads of the model-makers!

But this isn’t a reasonable response, in light of the complex system we’re dealing with. As one of its many monikers suggests, this is a novel coronavirus, meaning we initially didn’t have much information about it at all. The less information you have about the initial condition of a complex system, the worse your predictions will be about that system’s evolution. There is nothing ulterior or incompetent about this, it’s simply baked into the complex system pie.

First of all, the lower death rates may be due in part to some other part (or parts) of the system than social interactions – we wouldn’t be able to know. However, we do know that lockdowns would have to have had some effect on the spread of the virus simply because altering normal social patterns perturbs the system by altering the typical ways in which respiratory viruses would spread.  Simply put, if fewer people are in contact with each other fewer times and for shorter periods of time, then logic dictates that this would interrupt the spread of the pathogen. This would seem to be such a basic point as to be both inarguable and unnecessary to argue.  How much the lockdowns affected the spread in different regions, and the magnitude of other variables in the system compared to the effects of the lockdowns, is probably impossible to tell with precision. Whether the dampening was worth the economic dislocations the lockdowns caused around the world is another question altogether, but is also likely to be definitively unquantifiable as well.

Yet, every effort to discuss the effects of lockdowns has treated them as a one-size-fits-all system with completely linear solutions. Either lockdowns worked uniformly well or were uniformly a failure. Thus, you see people comparing infection and mortality rates across countries and states without really dealing with any other variables, even in a qualitative sense.  Could some countries have been hit less hard because they locked down earlier than others, thus introducing a perturbation much earlier into a possible chaotic system that is sensitive to initial conditions as its trajectory plays out?  Did masking culture in several East Asian countries dampen the spread so that they didn’t need lockdowns, thus giving the appearance of inutility to lockdowns?  Could Sweden’s infection rate be lower than many other European countries because Swedes, as Scandinavians, already have a form of social distancing in place culturally (large personal spaces) that Spain and Italy don’t?

From the other direction, it’s argued that since the USA’s infection and death rates are so high, this shows that lockdowns don’t work. But once again, how many variables are interacting non-linearly here?  And of course, Andrew Cuomo and other progressive governours sending COVID-19 patients into nursing homes to infect high risk individuals trapped in a closed environment certainly helps to skew this figure as well.

So what can we say about lockdowns?  That they probably worked in some places and not so much in others.  Applying them the way that they have – in a dumbly uniform fashion that doesn’t bother to factor in local potential feedback sources – has been profoundly unscientific in its approach. In many higher risk areas lockdowns likely did contribute to flattening the curve (which, if we’ll recall, was the actual point to them), among other mutually networked factors. Equally like is the notion that now that the risks have been mitigated, the lockdowns are probably not needed anymore.  Either way, flatly stating that “lockdowns didn’t have any effect” is almost assuredly wrong, but so is attributing to them so much magical mitigational power that we have to keep doing them for months or years to come.

This, in fact, has been one of the biggest problems with the narrative pushed by the mainstream media and most of the SCIENCE! types who are calling for endless shutdowns – their superstitious approach that oversimplifies the problem and therefore oversimplifies the proposed solutions.  Because many of these folks, especially found on the progressive Left, don’t have a solid grasp of what actually takes place during a pandemic and all the factors involved, they are prone to treating the pronouncements of proven bad actors like the WHO and Dr. Fauci as a Stone Age savage might the hocus-pocus of a witch doctor. This makes it all the more ironic for them to loudly proclaim their devotion to “the science.”

Seeing as there are no “magic bullet” solutions which are simply going to make COVID-19 go away, what should we be doing about it? Let us bear in mind that even though complex systems are not fully predictable and can react to perturbations in unexpected ways, they are nevertheless not random or senseless.  There can still be a reasonable expectation that disturbing the system in some way will produce a logical effect, even if the magnitude and effects on other parts of the system may not be predictable. This was noted above with respect to the argument the lockdowns have to have had some effect in the direction we would expect.

As a result, we certainly are not powerless in the face of (possibly Chinese manufactured) nature. What can help to reduce viral spread?  Social distancing may have some marginal effect, but it’s starting to look like six feet isn’t sufficient in many cases to prevent human-to-human droplet transfer. But let’s face it, much more social distancing and we’re going to completely disintegrate our already atomising society. The logical alternative is masking when in close contact situations, which is effective at stopping particles which can transfer virus. As noted above, societies which adopted universal mask wearing early on were much less affected than many late adopter Western societies.

Another thing we can do is to encourage as many people as possible to get outside and get active.  Fresh air reduces viral load per volume of air, especially as compared to air being recycled through a building’s HVAC system (which is connected with increased indoor transmission, by the way). Ultraviolet light in sunlight acts as a natural disinfectant to kill viruses. Sunlight also acts to produce vitamin D naturally in the skin (within certain latitudes), which has been suggested to have prophylactic effects against COVID-19. The recorded rates of outdoor transmission are very low as compared to those for indoor transmission. One of the most idiotic aspects of governmental responses throughout this crisis has been to keep people from going outside.  Arresting mothers for taking their kids to the park or lone surfers on a sunny beach is simply stupid and counterproductive.

Since the curve has been flattened, a better alternative to continuing lockdowns is to enact contact tracing protocols which would allow rapid tracing and testing for people who may have come into contact with carriers.  Positive trace cases could then be isolated and receive treatment, hopefully resulting in a transmissional dead end. This would help to minimise the problems associated with super-spreaders. The main problem here is that because most Western governments have been so inept and untrustworthy in their responses to date, a large chunk of the population simply will not be confident that contact tracing would be done without ulterior motives. And don’t even get me started on how you’re going to get many folks to accept any vaccine that may be produced.

Because it’s been in the news ever since the Bad Orange Man mentioned it, the use of hydroxychloroquine (HCQ) should be addressed.  In several small scale studies, a combination of HCQ with azithromycin and zinc supplements indicated successful prophylaxis against COVID-19 in early stages of infection.  Countersignaling this is a recent “registry analysis” (i.e. not an actual clinical study, per se) reported in the Lancet, which suggested no positive effects from HCQ. However, this study reported uses on patients with advanced infections, many with other co-morbid conditions, and did not include zinc as part of the treatment suite.  As such, the Lancet report was largely irrelevant to the results of the previous small scale studies.

Again, none of these things should be treated as simple linear cures for COVID-19. But they can help to perturb the virus’ transmission system in ways that provide negative feedback to push the pandemic back toward the “equilibrium” of no transmission (i.e. the ideal state of stopping the virus). Of course, as we learn more about the virus, we can better understand how the various interacting parts of the overall pandemical system work together, which would give better (but not perfect) predictability for actions we might take.

Nevertheless, we will have to accept the fact that much about this complex system will confound prediction. We may find that it mutates a new and more virulent strain that makes it resurge with a vengeance. We may find that it goes away quietly into that good night. Or it might be a little of both, or neither. The main thrust of any response should involve the humility to accept what we can’t predict and attempt to determine the best negative feedbacks which can observe and use to fight this virus.



One thought on “The Failure of Simplistic Solutions to Complex COVID-19 Problems

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s