If you prefer to listen rather than read, this blog is available as a podcast here. Or if you want to listen to just this post:

Or download the MP3


The last couple of posts (not counting my monthly book review post) have covered the evolution of systems over time, though, as I’m sure you’re aware, the word “systems” covers a lot of territory. Over the last several decades and perhaps longer, there’s been a lot of attention paid to small interpersonal systems. This has led to whole industries devoted to ensuring safety in the workplace, or productivity in the office, and my first post drew on that side of things. But while I’m interested in these sorts of systems, I’m worried (like most of the rest of the country) about much larger systems. Particularly our system of government. 

Certainly we can hope and maybe even expect that improving our system of government should be similar to improving any other system. That the same tactics which work to improve airplane safety might also work to improve government effectiveness as well. But despite whatever optimism we might bring to the process it’s clear that improving a system of government is going to be vastly more difficult than improving any other system. Let’s start with a simple example:

In both of my past posts I used the example of a checklist. A checklist is one of the simplest ways for preventing deviance in a system and it requires a few things to be effective.

First, you have to have some idea of actions which need to be taken or items which need to be reviewed. And ideally, these are things where the answer is a definite “yes” or definite “no”. Either the gust-lock has been removed or it hasn’t. Either the oil is above the fill line or it isn’t. And if the answer is “no” the process for rectifying it should be straightforward. For example removing the gust-lock or adding oil, respectively. 

Second, the process for creating and updating the checklist has to be straightforward, and not prone to disagreement or ambiguity. Everyone should basically agree what goes on the list and what doesn’t and it shouldn’t take large amounts of time to reach agreement or to add the item.

Third, checklists should rectify the mistakes of the past. If a plane crashed because a cargo door was incorrectly secured, then a checklist item saying “Ensure cargo doors are correctly secured” should be added. This way at least you’re not making the same mistake again.

Looking over this list it’s obvious that each of these things becomes much more difficult when you’re talking about a government. If we go back through the list:

First we need a list of actions and the actions need to be definitive. There’s problems on both sides of that mandate. You can imagine an action item “Is there a healthy debate about the issues affecting the nation?” Probably most people agree that that item should be on the list, but even if that’s the case answering the question with a straight yes or no becomes very difficult. If by some miracle there is a consensus, for example if we can definitively answer “no”, as increasingly appears to be the case at the moment. At that point, we still have a problem with the other side of the mandate. Adding “healthy debate” is not as straightforward as adding oil. It doesn’t come in convenient containers at any gas station.

Second on the list was the process of adding to our checklist. Once again this is (understandably) difficult when you’re talking about a system of government. For example, some people feel very strongly that giving women an absolute right to choose whether to continue a pregnancy should definitely be on our checklist. But there are a lot of people who think it’s equally important for that item not to be on the list or for it to be on the list, but in a limited fashion. As we have seen coming to any certain conclusion has been very difficult. And this difficulty pervades everything about systems of governments, from making necessary changes to recovering from deviations, as we will soon see.

Third on our list was rectifying the mistakes of the past. Here we have at least two problems. To begin with, there’s a real lack of data. Nations and systems of governments don’t fail nearly as often as planes crash, and even if we’re talking about minor failures like financial crises they’re still relatively rare events. And when failures do occur the causes of an economic crash are much more difficult to pinpoint than the causes of a plane crash, which is the other problem. Take the Great Depression as an example. Despite decades of study, there’s still considerable disagreement about what caused it, and whether FDR’s policies helped or hurt. It would certainly be nice if there was some “secure cargo door” equivalent we could add to our economic checklist that would prevent the economy from crashing in the same way it did in 1929, but I don’t think there is, or there are many items, and not everyone can agree on them.

The point in going into such depth on this one example is that it’s the simplest example, the one most easily understood and implemented, and yet even this most basic method for preventing  deviance in a system of government ends up being riddled with potential problems. But perhaps having a governmental checklist seems silly to you or perhaps it’s hard to imagine how it would work, so let’s turn to something more concrete, the Amendments to the US Constitution. 

In essence the amendments are a checklist, or at least as close as we’re likely to get when you consider a governmental system in its entirety. And if you consider them in this fashion then the failures I listed above are immediately obvious. To start with, while the amendments are admirably clear, particularly when compared with previous attempts at this sort of thing, they’re not unambiguously clear. What does “freedom of the press” mean in an age of social media. What precisely constitutes “cruel and unusual” punishment? 

Moving on, perhaps the most obvious issue is that we have largely lost the ability to add to this checklist, at least when it comes to anything important or anything which is the tiniest bit controversial. Instead of adding amendments, the current method for changing the constitution mostly involves the Supreme Court broadening the interpretation of what’s already there. I would assume that we can all agree that this is happening, but once you get beyond the mere fact of its existence, deciding whether or not it’s a deviance or how things were always supposed to work, and further, if it is a deviance, whether it’s been normalized, and whether that might actually be a good thing, probably depends a lot on your political ideology. A subject we’ll return to momentarily

Finally there’s the issue of using the amendments to rectify mistakes. Anyone looking at the list of amendments, will quickly realize that while some of them are incredibly farsighted, others, for example the Third Amendment, are targeted towards rectifying very specific mistakes from the time just before the Revolution. And of course the 21st amendment is the greatest example ever of this process, and when combined with the 19th amendment represent the ideal of how this whole thing should probably work. But, if, as I argued above, the process of adding amendments is beyond repair, how do we go about rectifying mistakes which have only been uncovered more recently? Here again the Supreme Court comes into play but to an arguably even greater extent because now the ideology of the court becomes a factor, with some things viewed as unassailable rights or fantastically awful mistakes depending on which judge is speaking. A situation which goes a long way towards explaining why the last few nominations have been so contentious. And also, in my opinion at least, further evidence that this state of affairs is a deviation from how things were originally intended to operate.

As I have argued all of this represents a deviance in the system, particularly if we use the most neutral meaning of the word, i.e. doing things differently from how they have been done in the past. Given this, what are our options for dealing with a given deviance? Broadly speaking there are two we can correct it or we can normalize it. Unfortunately, as I’ve just spent several hundred words demonstrating, correcting it appears to no longer be an option, absent some fairly sweeping changes (for example a constitutional convention.) Which leaves normalizing it. 

This is where we connect the first post in this series with the last one. If you’ll recall in the first one I argued that the normalization of deviance is generally a bad thing, and something you need to continually guard against because, unless checked, it will gradually creep into whatever system you’re using and fatally undermine it. On the other hand, in the second post I showed that, occasionally, normalization of deviance leads to an altogether better system. Certainly you could imagine that as the English parliament grew stronger in the years before the revolution and things inched towards greater democracy, that this could have also been labeled a deviance from how the monarchy was supposed to work. And that further each time one of parliament’s new powers was solidified through usage that it could have been viewed as normalization of that deviance.

Several points jump immediately to mind. The first and perhaps the most petty, is that based on the events of the last few weeks and months I don’t think the UK parliament is the thing that comes to mind for anyone when asked to summon forth examples of well functioning systems of government.

Next, when you get into the history of these deviations to the English system of government you immediately realize how gradual they all were. I don’t think the same can be said of the deviations we’re currently experiencing. Not only are they comparatively rapid, but they’re numerous. A point I’ll return to.

For most people of a conservative bent it’s the rapidity of the change rather than change itself that raises concerns. It is possible to change a system of government suddenly, but it rarely works and it’s always bloody. Some of my readers have questioned where I expect the blood to come from; who I expect to take up arms. And it is a subject which deserves a deeper dive, and one where they probably have something of a point. Nevertheless, it’s important to remember how civilized everyone thought things were before World War I…

Still, there are many people who are probably not comfortable in using deviance, even in its most neutral sense, to describe what’s happening. Everything is just progress, and the faster we progress the better. That most of our attempts to metaphorically keep planes from crashing is better understood as being equivalent to refusing to move from propellor driven engines to jets. This is a valid point, how do we distinguish between harmful deviance and innovative deviance? How can we tell whether our current course will lead to civilizational catastrophe or a communal utopia?

As I alluded to previously, introducing numerous deviances all at once seems particularly fraught if you’re trying to make this evaluation. As has been pointed out, the modern world is fantastic by most measures, but which recent deviation accounts for the innovations we see? Does science or women’s suffrage explain the current technological bounty? I lean towards the first, but it could easily be both, or neither. And if the modern world has problems, which it clearly does, even if these problems don’t pose an existential risk it would be nice to know their source. Is the current increase in suicide cultural? Entirely the fault of the opioid epidemic? Or something else?

The argument people are making is that we’re now smart enough to only deviate in ways that make sense. We’re not doing the equivalent of going into an upside down loop in order to lock our wheels, we’re only doing things that are clearly good ideas. Well, as both I and the original author pointed out, all deviations seem sensible initially, until you’re 300 feet off the ground and about to crash. And frankly even if we are going to go by that standard, do our current deviations actually meet this criteria? Does having completely open borders make sense? Does the increasing number of transgender people make sense? Does Modern Monetary Theory (MMT) make sense? (Probably not, if even Krugman thinks it’s ridiculous.)

It seems worth spending some time on that last one, since it would appear to be something of the platonic example of normalizing deviance. Under any normal financial system one of the checkboxes would be “Do you spend less than you make?” Now I can certainly see an argument that for the government the standard might be somewhat different, perhaps “Is the budget deficit percentage less than the rate of inflation?” But MMT goes way beyond that to “Is inflation at a reasonable level? This would appear to be both a gross deviation from how things have normally been done, and also, by wrapping it in the MMT ideology, one of the more bald faced attempts at normalizing a deviance I’ve ever seen as well. All that said, as I pointed out in the previous episode, there is some chance (I would argue a very small one) that they’re right, that it will in fact work better. That this is one time when we’re not headed for destruction, but when we’re actually pushing through to a new and better system on the other side. But how likely do you actually think that is? And not just with MMT, with any of the things I’ve mentioned?

Still you may have noticed that while I’ve danced around things, I still haven’t answered the fundamental question of how can you tell? How can you know whether the deviance you’re normalizing will lead to civilizational catastrophe or a communal utopia? And I’ve avoided answering it largely because it’s very difficult to tell. However, in closing I will offer some pointers for some things you might want to consider:

  1. Generally, it’s probably not going to lead to catastrophe, but on the other side, it’s NEVER going to lead to a utopia.
  2. Trying numerous radical changes all at once never seems to work. For example, we seem to be combining radically different immigration norms, with extreme changes in culture and extreme government spending all at the same time.
  3. The best deviations are one’s where the benefits are massive and straightforward. For example ending slavery. 
  4. Related to that, it’s also great if they’re easy to understand. In particular I think MMT, whatever its brilliance absolutely fails this test.
  5. Is there an asymmetry between failure and success? Is failure catastrophic, even if it’s unlikely? Is success only marginally better even if it’s nearly certain?

Should you have any other points you feel I should add to this list, or any considerations you think I’m missing I’d be happy to hear about them. But if we take just this list, I don’t see any reason to consider current deviance as anything other than dangerous.

To end where I began, we’ve got an old broken down aircraft. There’s a checklist for keeping it running, but people can’t agree on what the items on the checklist mean. We can’t change the items on the checklist even if we could agree. And there’s a huge debate on what things constitute mistakes and what things constitute progress. The plane is still flying but increasingly the pilots are focused less on flying and more on debating the condition of the plane, and whether the duct tape on the rudder is a bad thing or the latest in aircraft technology. And as one of the passengers, I gotta tell ya, I’m pretty nervous.


This is not the first time I’ve introduced an extended metaphor involving planes. If you want to ensure that these metaphors continue (and I mean really who wouldn’t?) Consider donating