Poisoning the well

Certainty is everywhere, fundamentalism is in full bloom. Legions of authorities cloaked in total conviction tell us why we should invade country X, ban The Adventures of Huckleberry Finn in schools, or eat stewed tomatoes; how much brain damage is necessary to justify a plea of diminished capacity; the precise moment when a sperm and an egg must be treated as a human being; and why the stock market will eventually revert to historical returns. A public change of mind is national news.

This is the first paragraph from On Being Certain by Robert A. Burton. I am just a couple of pages in, but this has stopped me short.

The book is ostensibly about the biological origins of the feeling of knowing. How it is separate from “reason” and logic. But this paragraph perfectly encapsulates how the way society frames issues ignores and even rewards unwarranted certainty.

Modern media has an endemic sense of certainty. Journalistic convention is based on an underlying assumption of causation, of the world in front of us as the direct result of something that can be tracked down and explained. Something happened so there must be someone to talk to, or a bang that preceded it.

There’s no way it’s unknowable, or the result of complex interactions we can only tease out with time and after making many assumptions. Dogged by problems of measurement and perception. As a result you get a lot of declarative statements, black and white.

When a professional athlete is doing well, for instance, we are furnished with stories of their extensive workouts. When they do poorly we hear about their troubled childhood and off-court issues. Or maybe they just suck now. There’s little room for underlying randomness, problematic measurements, statistical noise and mean reversion. A cause must be found and responsibility taken.

The issue here is the need for a narrative. As a journalist, narrative is an important tool for grabbing someone’s attention, keeping it, and guiding them through a larger point. Or to highlight something specific and make it memorable.

But what does a narrative need? In this context it almost always entails simplistic cause and effect.

At the end of many news bulletins we get a financial update. We hear how this currency rose, a stockmarket over there fell, after hours trading is stagnant etc. All fair enough, except that they’re often immediately tied to a news hook.

The Yen went down cause a Yeti was spotted in Turkey. The Nasdaq rose cause Dutch tulips were especially vibrant this year. Sunspots.

Narrative is useful for audiences to connect with this kind of abstraction. But there’s no way causation for activities this complex were nailed down in the time between the signal and the news piece, if they ever can be.

These triggers often are big enough to have some association, but how much? How that was figured out is honestly an even better story.

That last line from Burton, about a change of mind being national news is also deserving of unpicking.

It is fair enough that leaders changing their minds about something is news. But the problem is in how it is approached. How often is the story about the change itself rather than what underlay the previous “belief” and how that changed? How good is the information, or, if that hasn’t changed, the mental model that reinterpreted it?

Beliefs often aren’t a binary proposition, especially when it comes to policy. Rather, they are about juggling trade offs, dogged by information asymmetries and stretched resources (mental and otherwise).

But here I am also falling into the trap of beliefs as the function of logic and reason. As I’ve documented here, beliefs have many potential fathers. Perhaps biology is one.

Show me the receipts

A consequence of the last decade-plus of mindless, saturation media is an endless parade of “experts” and “pundits” who have little at stake and pay no price for mistakes.

The very loud people who oppose action on climate change, for instance. Even business lobbies are often add odds with businesses, many of which are already planning for a changed climate.

There’s a brilliant article on VoxEu about the people who do have money on the line:

In our recent paper (Schlenker and Taylor 2019), we look at weather derivatives in financial markets to assess beliefs about climate. We find that traders have been pricing in a warming trend that is closely aligned with the projections of scientific climate models, as well as the observed weather outcomes during that time…

…Futures prices closely follow the predictions of climate scientists, which, on average, appear to have materialised, thus validating the climate models. This close agreement between markets and models implies that traders are taking into consideration the scientific consensus on climate change when making trades. Overall, we find that the market has been accurately pricing in climate change, largely in line with global climate models, and that this began occurring at least since the early 2000s when the weather futures markets were formed.

There are plenty of liquid markets where someone with a contrarian take on climate change or the housing market (etc.) could make a lot of money.

It isn’t enough that you were “right last time”. That could have been a fluke. That might have been one correct prediction after many wrong ones.

In the real world that batting average doesn’t work. So, journalists should ask for receipts.

If an expert or pundit makes a bold claim they need to put their money where their mouth is. Otherwise it’s just hot air.

What causes crime?

There is something we find uncomfortable about a world without obvious answers, which is one reason, of course, why we cling stubbornly to our dominant views of crime. Our thirst for certainty may also explain why a number of theories about crime have, when stated boldly, gained alarming and unjustified popularity.

This quote from Criminal by Tom Gash falls right into the recent theme on this blog.

I’ve just started this book, and I’m not entirely sold on its premise yet. But it does a good job of framing how public policy and discussion can be captured by a false dichotomy. And seemingly because the simple narratives fill a painful void – we don’t know why people commit crimes and it’s probably too complex to ever submit to an single, easily expressed answer.

Gash essentially argues there are currently two dominant explanations for “why” someone commits a crime:

The first I call the ‘Heroes and Villains’ view, because of its moral emphasis and its central premise that those who commit crime must be confronted by the full force of the justice system to avoid society becoming corrupted. The second I call the ‘Victims and Survivors’ view, reflecting the argument that crime is not simply a selfish choice but often one forced by adverse circumstances…

This dichotomy between evildoers and victims of circumstance is quite evident in the world around us. Take the war on drugs and it’s more recent backlash, which has so often been a proxy for deterrence.

I’ve myself strongly argued against deterrence, leaning heavily into arguments about social context, learned norms and identities, and, ultimately, path dependence.

But this is really too simple.

Think of the last time you committed a crime (contrary to our Heroes and Villains view, most of us have) – or, if you are in the minority, think of a recent case you know of. Then consider the forces at work affecting your behaviour. I am certain that you will most likely uncover layer upon layer of possible influences, many considerations interacting in complex ways to facilitate the crime in question.

As Gash notes, “motivation” is too blunt a tool for what we are trying to accomplish. In fact, crime appears to be highly specific and contextual, and we’re ignoring profound economic, social and technological shifts.

There probably is neither one reason or one answer.

As always my emphasis

There probably isn’t one reason

I finally finished The Hidden Half. It’s one of the best books I’ve read in a while, and ties together much of my reading and thinking over the past year or so.

As a recovering determinist, I relish the celebration of uncertainty and the unknown. I’ve written quite a bit as I’ve read along. But here’s one more thought – the implications of uncertainty for silver bullets.

As much as we try to make the world bend to our will, there likely isn’t just one reason for anything. And so there probably isn’t one solution for it either.

…The biggest things are unusual by definition. Unusual things often result from an alignment or interaction of many circumstances – that’s why they turn out big. By their nature, these will be harder to understand. However, this does not mean we have failed to research them as well as reasonably possible: in a world of enigmatic influences, research rigour does not equal nailing down. The best answer might be that there is no answer.

The bigger the thing you’re trying to tackle or explain, the more influences it will likely have. Including ones you can’t see or measure. If you remove any of these jenga blocks, will your notion stand up?

This makes transplanting explanations or “solutions” from one context to another incredibly problematic. Your idea may have “fixed” the problem over there – and that’s a big if. But do you really know why? What about all the factors underlying that?

History is littered with simple solutions to complex problems and we’re all prone to creating panaceas. Modern democracies, especially, incentivise simple explanations rather than waiting, seeing and experimentation.

But the world defies being put in a box.

This is why public policies so often miss or fail entirely. Complex problems have complex causes and likely require nuanced and adaptable solutions. That it’s worked before or fits a particular world view isn’t enough.

…A favourite big thing, a silver bullet, has so many advantages: it’s easier to sell, to describe, to understand, to put into practice. But whether the thing we pick would travel, on its own, to another context is another question. Silver bullets seldom work once, never mind twice.

As I have written previously, what this requires is a little more humility, as well as institutions and a culture that can accept uncertainty and not knowing. Working with best approximations and striving to improve them.

As always my emphasis.