Big Think

Subscribe now

Erik Hollnagel gives major academic dad vibes. He has a basic HTML site, has published a million pieces of writing, and has some great ideas. This past week I’ve been reading a bunch of his stuff. A few ideas that have stuck with me:

• Dynamic non-events (Business-as-Usual)

• Efficiency-thoroughness tradeoffs (The ETTO Principle)

• Safety I vs. Safety II (Zero-Accident vs. Maximum Non-Accidents)

What is a non-event? It’s when something doesn’t happen. For example, say a warehouse goes a day with no forklift accidents. An accident is an event. The absence of an accident is a non-event. If, every day, there are 1000 forklift maneuvers, and 0 of those maneuvers result in an accident, there are 1000 non-events. This string of non-events is known as a dynamic non-event. In other words, business-as-usual. It’s a non-event because there are no events. And it’s dynamic because that can change.

This takes us to the ETTO Principle – the tradeoff between efficiency and thoroughness. Using the warehouse and forklift accidents as an example. Maximum efficiency means the forklifts operate as fast as possible, wreaking havoc along the way, but excelling at moving boxes to and fro. Maximum thoroughness means the warehouse stops using forklifts, thus reducing the number of forklift accidents to zero.  

And that’s why the traditional philosophy of safety (What Hollnagel calls Safety I) pits workers and employers against each other. Workers want thoroughness (they want to be safe). Employers want efficiency (they want to be productive). According to ETTO, these things conflict. But Hollnagel came up with a different formulation of safety, which he calls Safety II. In the warehouse language, the goal is to perform as many accident-less forklift maneuvers as possible. The result is the same: improved safety. But the underlying philosophy is different.

“…to manage safety we need to know what happens when ‘nothing’ happens.”

-Erik Hollnagel

Sarah Perry is a researcher and writer. Last week, she gave a great talk to the SoP cohort. The topic was “The Generalizability Crisis”. It’s similar to the problem of replicability in the social sciences. Generalizability (G) and Replicability (R) are the two most common failure points in studies of psychology, economics, behavioral economics, sociology, etc. 

An R failure means that the results of a given study could not be found again under the same conditions in a duplicate study. This means the findings of the original study are suspicious. A famous example of this is Daniel Kahneman’s research on priming (A.K.A. nudges). It turns out the experiments proving priming did not replicate. There is no scientific basis. Yet, many policymakers continue to believe that they can use slight changes in physical environments to “nudge” people into certain behaviors.

A G failure means that the study is messed up by the use of vague words, like ‘depression’ or ‘happiness’. There is no set definition. There is no set method or unit of measurement. Researchers use different measurements and they interpret results using slightly different definitions. So comparing any two social science studies is like comparing apples and oranges.

In 1873, the Canadian Labour Union was founded in Ontario. It was founded by 46 local unions. Due to several factors, it fell apart after just a few years. The Union’s mission was “to agitate such questions as may be for the benefit of the working classes…”. This is pretty standard. But the idea of a union of unions appears radical. It’s certainly logistically complex – which makes me wonder: would there be unique benefits that made it worthwhile?

My summer project is about workplace safety, so it’s been impossible not to get sucked into the topic of unions. Organized labor has been responsible for many of the improvements in workplace safety. It seems that workers can’t change much on their own, especially when they’re replaceable. Unions enable workers to improve their bargaining power. So why didn’t a union of unions work? Wouldn’t this just have increased workers’ bargaining power even more?

A hint might lie with who is at the negotiating table. Historically, unions have been comprised of either:

1. Workers under one large employer, such as Toyota, or

2. Workers in a certain trade or industry (that consists of multiple employers).

Type A. is known as a Company Union. Type B. is known as a Trade Union. What we haven’t seen yet, however, is a General Union. A union of workers across companies AND industries. In part, this is because it’s difficult for that group to find a counterparty to negotiate with. Another factor is logistics. In 1873, geography was a massive constraint. In fact, the Canadian Labour Union consisted only of companies in Ontario. It was never a national organization. 

One of the slogans of early Marxist movements was “Workers of the World, Unite!”. It was translated into dozens of languages. But again, this union of unions never formed. Who could it have negotiated with, if it did form? What could it have achieved, that local unions can’t?

I’m starting to think that some version of this union of unions could exist today, for a few reasons. New technologies have broken us out of geographical constraints and language barriers. There are larger, more mature national and international organizations which a General Union could influence, lobby, or negotiate with. Covid-19 put the spotlight on public health and the “leanness” of our global supply chain. Gig workers, contractors, entrepreneurs, and small businesses form the long tail of workers – and they lack the time, resources, and critical mass to form unions on their own. 

This Union of the Future isn’t here, yet.

A common misconception: survival of the fittest. We think evolution is driven by the most successful members of the gene pool. In reality, it’s driven by the exit of the least successful members. Fitness is not about being at the top. Fitness is about not being at the bottom. Of course, in terms of reproduction, it’s better to be at the top. But the first order of business is escaping the bottom 20%. So, a less misguiding expression is “non-survival of the least fit”.

Flourishing is about maximizing human potential. There’s some value in looking at humans as trees (I’m biased). To flourish, we abandon our less successful branches in favor of our more promising ones. The process of flourishing is growth enabled by sacrifice. Goblin mode is like the opposite of flourishing – holding on to everything. 

Fitness (as commonly conceived) and flourishing have a sinister side. When scaled to the species level, they no longer work. We should not treat humanity as a tree. Individuals and groups cannot be lopped off by a megalomaniac, self-appointed socio-arborist. It’s more productive to think of ourselves as a forest, where the trees tend to themselves without impeding the flourishing of others.

A colleague this week shared with me that his experience as an entrepreneur was much more rote or scripted than he expected. Operating in the world of venture capital meant that he had to do pitches. Many elements of the startup culture felt like a caricature. Earlier in life, he’d witnessed a different kind of entrepreneur – one who had built a company from the ground up without much (if any) outside investment.

Today, this is known as “bootstrapping”. Defined as: growing a company without giving up equity (ownership in the company) to investors. Equity stays within the company, between its owners and operators. Bootstrapping is often talked about, and praised, but rarely done. Taking on venture capital eliminates a great deal of risk, but it also reduces your optionality. You have to conform to the norms of startup culture.

Similarly, there are two ways of building a social network. The old-fashioned way and the new way. By old fashion, I mean making friends in person and getting exclusive means of communication (like a physical address or regular cafe). The new way is social media. It’s a supercharged and fast way of building a network of friends and acquaintances. But like venture capital, it limits your behavior to a set of norms. The norms of social media platforms. You might make more friends, more quickly, but your network might not feel as real as it would otherwise.

Subscribe now

My SoP project page looks nicer.

I have a new personal site.

Did you know that porcupines float?