Antifragility, Mustachianism, and The Machine – Haberdashery for Uncertain Times

The future is impossible to predict with certainty, which is why oddsmakers exist, weather forecasts include percentages, and we experience time as linear.

There are times though, when the impossibility of prediction becomes more obvious. In the summer of 2020 I wrote a facebook post likening the situation to a singularity.

It seems there may be a grapefruit-sized black hole kicking around in our outer solar system. The metaphor of a singularity seems apt for this moment. As the election countdown shifts from months to days the familiar rules of politics are breaking down.

Physicists attempt to learn about the nature of our early universe by increasing the energy of particles in colliders. Similar weird physics must emerge as particles approach the event horizon of a black hole. Electroweak force?

As we spiral closer to our own event horizon some familiar political forms are becoming unstable. Like protons breaking up into quarks and reforming briefly in exotic forms, the right and left seem poised to split at the seams, and the inexorable gravity of first-past-the-poll plurality voting will bind the remnants, if only briefly, into some bizarre new forms. The conservative pundit class has fractured into classic liberals and authoritarians. Militia groups have fractured into libertarians and authoritarians. The anti-establishment left has fractured as well. Some Russia phobic elements have joined the establishment in an uneasy alliance with the intelligence agencies, while other elements have enough of an anti-establishment charge that they may spin off into third-party territory. As old coalitions fracture, the establishment of both parties is coalescing into a block. Are decades of reductionist fear-mongering and cheapening of the discourse one of the root causes of the current regime? No matter, the masters of the Dark Arts are now briefly aligned with social justice warriors.

And what lies beyond this event horizon? It seems impossible to predict. The alliance between corporate Democrats and Republicans seems potentially stable, but the gravity of our system inexorably pushes us toward two parties.Which seems more likely – that the woke left could find common cause with the bugaloo right to counterbalance a unified establishment party, or that our system itself will fracture?It’s a bad time to make confident predictions.

As of March 2023, with MAGA Republicans continuing to double down, the tension in the system has only increased.

But beyond politics, there are a number of trends playing out at an exponential pace. Humans are famously much better at understanding linear change than exponential change, and the muddy interplay between these exponential trends will no doubt further confound our forecasts.

On the bad side: climate change and inequality both seem to be accelerating, and that’s obviously not good or sustainable. On the good side: solar panels and battery storage are also quickly improving, so electrification may happen more quickly than people realize. On the “too early to say” side: Moore’s Law and now AI careen onwards, promising dazzling breakthroughs, whole industries disrupted, and who knows? Deepfake kidnapping scams or political scandals? Terrorists using AI to probe weaknesses, design weapons and plan attacks? Skynet? Things could go in many different directions.

In my mind there are two large categories of scifi futures: the unitard jumpsuit future of Star Trek, and the fingerless glove futures of post-apocalyptic Kevin Costner movies.

In “Antifragile” Nassim Taleb extols the virtues of the barbell strategy for risk management. To be resilient to the point of antifragility you must prepare for the worst and prepare for the best and forget about median outcomes.

Affix fingerless gloves to your unitard and you will be ready for anything.

So how do we prepare for the two scifi futures?

As I see things, Mr. Money Mustache’s principals are spot-on for a future that continues going well. Basically these are:

1. The most efficient way to make money is not through work but through having money invested.
2. Index funds are the most efficient, rational way to invest money.
3. Over the long term these funds can safely be expected to grow at an average of 5% a year.
4. Therefore, once you have saved 25x your annual living expenses in index funds you can expect to be able to live off of the interest/dividends from your investments at a 4% withdrawal rate in perpetuity.
5. Therefore, a penny saved on your annual cost of living is 25 pennies you don’t need to save for retirement.
6. “Badassity” (basically, the willingness to do tough work or tough out a less than totally idyllic circumstance) can be a substitute for spending money in daily life.
7. Badassity is like a muscle, the more you use it the more you have, whereas the more money you spend, the less you have.
8. Therefore, if you want to hasten retirement (or the option to retire), you should learn to substitute badassity for money wherever you can in your daily life, treat any debt like an emergency, and invest all money you save or earn into index funds until you have 25x of your annual living expenses invested.

There is your prescription for the unitard / humans-mostly-keep-it-together future.

But sometimes the stock market does not increase at all for decades (see: Japan 1990 – present) And sometimes climate change makes your settlements less habitable (see: the vikings) And sometimes your population increases beyond what your environment will continue to support (see: Easter Island) And sometimes Carrington events?

What is the prescription for the fingerless glove future? First let’s talk about the present.

I once read an essay that I cannot find or fully remember now, but it put into my head the metaphor of The Machine, named after a 1909 short fiction story by a  E. M. Forster “The Machine Stops“. The Machine today is what allows you to sit at home in your pajamas and cause all manner of cheaply produced goods to appear at your doorstep.

Above all, The Machine is efficient. It can produce goods at impossibly low costs. The Machine’s efficiency is also the reason why your invested money is better at making money than you are.

The machine takes materials mined deep beneath the earth, transforms them, and delivers them to your doorstep. The machine facilitates and coordinates interactions between customers who need services and those who can provide them. Supremely efficient, all the machine requires is the smallest of cuts. 

So long as growth is possible, the machine can deliver benefits to both customers and shareholders. However, when the potential for growth diminishes, it becomes clear that the primary purpose of the machine is not to deliver value to the customers, but to transfer wealth to the shareholders.

When growth stalls, the enshitification process begins.

A manufacturer might burst onto the scene with a superior product, but once they have successfully created a monopoly moat and their growth hits its inevitable real-world limits, the price will go up or quality will begin to degrade. Often both.

I’m old enough to remember a time when a Google search could re-summon an essay you had read, Facebook was a fun way to figure out who that kid down the hall was, and Netflix ran a prize to help you pick movies you were more likely to enjoy. Now a Google searches work best only if you’ve got your wallet open looking to buy something, Facebook feeds you empty dopamine hits to keep your eyeballs in front of ads, and Netflix has for some reason re-created the experience of wandering dazed through the aisles of mediocre videos at Home Vision Video – and they charge you for the privilege. The Machine is taking your time and attention and converting it into wealth for shareholders.

But the Machine has even more subtle, insidious ways to transfer wealth.

In the universe, entropy increases with time. However, the opposite thing can happen for a limited time on a local level. Living things create little bubbles of lower entropy around themselves. Plants take lower entropy light and CO2, take a cut, and produce order in the form of growth, leaving behind higher entropy oxygen and heat. Herbivores consume these lower entropy plants, use the energy to grow and move, then excrete higher entropy poop. Humans can take lower entropy wood and burn it to do all sorts of useful things, leaving behind higher entropy heat and ash.

I think burning is a good metaphor. Take something of low entropy, burn it, and release something of higher entropy. The difference in entropy between the feedstock and the waste is the maximum harnessable energy.

The Machine burns things too. It burns oil of course, but it also burns more metaphorically. The Machine takes public goods like clean air, human health, and good governance, transforms them into shittier air, shittier health, and dodgier governance, and transfers the difference to shareholders.

Accumulation of wealth for shareholders is the primary purpose of the machine, and it is not only effective, but surprisingly creative at accomplishing that goal. Isn’t it AI that’s supposed to be surprisingly creative?

These days everyone is talking about AI, and the things they fear it might do. These concerns should be taken seriously – not least because complex systems with bad incentive structures and insufficient safeguards have already triggered many of the problems AI naysayers are warning about: erosion of trust, backsliding of democracies, ecological disaster etc… The danger is not new though. The biggest danger with AI is likely that it will continue down the same road we are on, only much more quickly.

Why are we on this road? As Charles Eisenstein explains in Sacred Economics, fundamentally it is because our money is interest-bearing. New money is always accompanied by a larger amount of debt, and thus we lock ourselves into a growth imperative which expects the economy to perpetually defy the second law of thermodynamics. Of course it has not. We demand perpetual growth and what we get is monetization of the commons.

The vast commons, whether of land or of culture, has been cordoned off and sold-all to keep pace with the exponential growth of money. This is the deep reason why we convert forests to timber, songs to intellectual property, and so on. It is why two-thirds of all American meals are now prepared outside the home. It is why herbal folk remedies have given way to pharmaceutical medicines, why child care has become a paid service, why drinking water has been the number-one growth category in beverage sales.

The imperative of perpetual growth implicit in interest-based money is what drives the relentless conversion of life, world, and spirit into money. Completing the vicious circle, the more of life we convert into money, the more we need money to live. Usury, not money, is the proverbial root of all evil.

The Market is The Machine, humanity’s first AI. Monetary policy is the paperclip-maximizing program we have set it to run. Why do we think another AI built on the foundation of The Machine will be an improvement?

Using the metaphor of The Machine, Mustachianism could be boiled down to only two principals:

1. The Machine is the most powerful force thus far created by man.
2. Better to be more of an owner and less of a customer.

But efficiency and resilience ( /antifragility) are usually in tension. The most efficient system will expand most quickly, but it is unlikely to survive an external shock, or even, in the end, contradictions in its own internal logic. AI may just speed things along. On a large scale, like a microbe in a petri dish, an AI boosted Machine might burn too hot, over-consume its feedstock, and collapse in on itself.

Efficiency and antifragility are in tension on a small scale as well. Corporate efficiency experts, (and Mr. Money Mustache) would have you make your purchases just-in-time rather than just-in-case. Don’t buy extra stuff, and don’t keep things around you aren’t using. Preparing for two futures is always going to be less efficient than preparing for one.

One thing that seemed missing from “Antifragile” was any discussion of the ongoing cost of keeping options open. According to Taleb, things which are like washing machines are fragile, while things which are like cats are antifragile. After a hurricane or an earthquake, it is not unreasonable to think that your cat might find a way to survive and then, conceivably, thrive. A washing machine, on the other hand, however robustly built it is, might conceivably survive a disaster, but will never be better for it. A cat can have options, a washing machine cannot.

This is all fine and good, but a cat has to feed regularly or it quickly ceases to be a cat. A washing machine can sit for years locked in a basement and remain a washing machine. A washing machine is delicate, complex, purpose built, and efficient. A cat may be antifragile, but it requires a regular supply of energy to remain a cat, and is never going to do your laundry regardless.

Insofar as I understand it, antifragility essentially comes from placing multiple bets at once. To keep placing each bet, we must continue paying the ante. The ante is the ongoing maintenance cost, the cat food as it were, of antifragility.

So if Mustachianism is the best bet to place in the world of The Machine, what’s a good hedge bet to place for a world without? (And is the ante affordable?)

You’ll want to eat. The stereotypical prepper would have a bunker filled with preserved food and ammunition. If you’ve got a hoard of food and hungry neighbors, you’ll probably end up needing the ammunition. And if the emergency persists, eventually you’ll end wishing you had a garden anyways.

A more pleasant strategy for surviving the apocalypse, in my opinion, would be to make your post-apocalypse life more mundane and your neighbors less hungry.

In “The Art of Not Being Governed” James C. Scott talks about “escape agriculture” – the practice of growing crops which are resistant to appropriation. 

Wheat, rice or corn are crops a government, warlord, or local armed gang would be happy to have you grow. They are standardized, not particularly perishable, will have a fairly predictable yield, will ripen at a predictable time, and after harvesting are usually kept in an easily identifiable central location. Grains are easy to seize.

Peas, on the other hand, ripen continuously. If someone wants to appropriate your crop more than one trip will be necessary. Root crops like potatoes can be left in the ground. If someone wants to steal your harvest, they’ll have to come and dig like a peasant. A patch of manioc can be kept in the woods (in a warmer climate than Maine), the foliage unobtrusive and the roots safely hidden away underground. Now the would-be appropriator needs a shovel, a keen eye, and a level of botanical knowledge.

It seems to me that, in an apocalypse, it would be best to have an array of crops diverse enough to allow most food to be eaten soon after harvesting, rather than being stored centrally. By locking yourself into a certain amount of menial labor, you make your stash less tempting. Increase the entropy in your food supply to make it less appealing. Security through badassity.

Farming isn’t easy work, and, as far as growing seasons go, around this latitude at least, there are no do-overs. It’s probably a good idea to get familiar with farming and the crops you think you would try to grow, and a rough idea of what land you would try to add into cultivation in an emergency. Having a regular practice of buying seed every year will keep you from getting caught flat-footed trying to chase scarce seed with a panicked crowd.

Seed packets usually contain more than the home gardener needs, and I usually can’t stop myself from getting more varieties than I have time to grow. In 2020 I had no problem ramping our garden way up when my work disappeared for a few months. Even then, I had seed left over. In a real emergency there may be some neighbors who would be very happy to have some of that seed themselves. Well-fed neighbors are better neighbors.

Of course, some seeds like squash are easy to save. It’s probably worth it to do a little experimenting with seed-saving. (Google’s Bard claims that the seeds of beans, peas, lettuce, tomatoes, spinach, and carrots are easy to save as well – although some of its detailed instructions were complete nonsense – “Cut the carrots in half lengthwise and then scoop out the seeds with a spoon…”)

Better than a regular seed-buying or seed-saving practice though, is a garden of perennials. Many perennials produce greens much earlier than annuals (think fiddleheads, ramps, and asparagus) so they can help fill in the early part of spring. A wide variety of fruits broadens the direct-picking window. Here in Maine I’ve eaten honeyberries off the bush as early as the first week in June, and medlar as late as the end of December. Learning to propagate, transplant, and graft would allow you to share your perennial resources with neighbors as well.

You’ll want other things as well – shelter, tools, probably some kind of deterrent. The main point is that there will be other things you’ll want. You could make many things yourself, either now or after the apocalypse, but the apocalypse is not here yet, and The Machine can make things much more efficiently than you ever could. Should you buy in advance?

To put it delicately, minimalism is not a particular strength of mine, but I was once an AT thru-hiker, and you quickly learn to think long and hard about what you are going to keep if you have to carry it on your back for 20 miles a day up and down mountains. One lesson I learned as a thru-hiker was to prepare to survive a worst-case scenario, rather than to prepare to be comfortable in it. You don’t need a second jacket layer for very cold weather – if it’s dangerously cold, plan to cook from your sleeping bag, with a hot water bottle at your feet if necessary.

Make your disaster plans as if you’ll have some badassity in extremis. Rather than planning to be totally comfortable after the apocalypse, plan to survive, then get busy.

You’ll want things after the apocalypse, and your neighbors will want things. You could be a useful member of a community by making things. Instead of keeping all these things on hand, you could keep the tools to make things, or even the tools to make the tools to make things on hand.

The metaphor of a seed strikes me as appropriate here. The wood of a tree is a valuable commodity, but there is no wood in a seed. The roots, leaves, and bark work together to power the wood making process, but there are no leaves, roots, or bark in a seed either. The seed holds the instructions for making a tree, some the machinery to start making leaves and roots, and a bit of energy to kick the process off.

A good seed for a few families might look like a small practical reference library, a rudimentary turn-of-the 20th-century workshop, a month’s worth of stored food, and a productive garden. Right now The Machine is happy to sell you all of these things at impossibly cheap prices. It’s probably a good idea to get yourself at least to this seed level while The Machine is still running.

And don’t get plastic crap. If The Machine were to stop you’d find yourself regretting it pretty quickly. The Machine churns out plenty of goods in aluminum, stainless steel, cast iron, and wood. Pay the extra money – it might never be cheaper. “The worker’s control of the means of production” I say to myself as I click “add to cart” and feed The Machine.

Writing this, it occurred to me that, just as in the garden, keeping a perennial might be preferable to saving a seed. Every community would be more resilient if they kept a collection of pre-industrial operations running. Farms of course, but also mills, and other traditional trades and industries. Maybe the signaling value of expensive hand-made goods would be enough to create the demand that could support such a small operation. If not, maybe they should be run as non-profits to provide educational opportunities and a community resilience backstop. But that is a larger problem of collective action, and I’ll set it aside for now.

Keeping a library, old-timey workshop, and emergency food will not be necessary – certainly will not be optimal in a Star-Trek future. Buying them now will only slow your path to Mustachian retirement. But some investments would be useful in either future.

A wood gassifier producing syngas for gasoline engines is not one of these things. It would be great in an extended disaster, at least as long as the machinery all worked. But otherwise it would be a superfluous curiosity and a waste of space. A combination of solar panels and electric tools or vehicles, on the other hand, could help achieve the same self-sufficiency in a disaster but would also deliver cost-savings in a rosier scenario. If a piece of equipment provides you value in both futures, should you buy it right away?

On the one hand, assuming that you are saving 65% of your income and are on a 10-year path to Mustachian retirement, at 5% interest, $100 not saved in year one has an opportunity cost of $163 by year 10. On the other hand, if a $100 efficiency upgrade saves $4.66 in annual expenses, it will decrease your savings target by $116.5. If you make the upgrade in year one, you will also have saved $46.60 by year 10, getting you $163.10 closer to retirement – better than our hypothetical 5% index fund growth. As you get closer to your savings goal, the math becomes even more forgiving.

Put more simply, if an efficiency upgrade will pay for itself in 20 – 21 years then it is within the realm of sensible investments. An upgrade with a quicker payback of course is preferable, and any badassity-based options (such as turning down the thermostat and putting on a sweater) should be undertaken first.

So energy efficiency, electrification, and solar/wind are easy to justify for any future. If you are hedging your bets you’d also want to ask: could this system be made to work without wifi? without the grid? Could it be kept running for 10 years without brand-specific parts? You might want to pay more now for something you know you could keep running. 

Still, appliances do fail. It would probably be good to have at least one small space you can heat efficiently with something like a rocket mass heater or masonry stove.

Speaking of spaces, you might want to take advantage of modern construction equipment and synthetic materials while the petrochemicals are still flowing. It’s much easier to dig a root cellar with a backhoe than by hand, much easier to source polyethylene for earth bags or portland cement for rammed earth. Rolls of plastic are the best bang-for-the-buck waterproofing for an underground structure and should last a long time if covered by a layer of topsoil.

In the end, it comes down to what odds you put on the two futures. If you are confident that the good times will keep flowing on through your grandchildren’s lifetimes, then maybe you are better served dispensing with antifragility and just going full mustachian. If, on the other hand, you’re convinced the apocalypse is only a couple of years away it makes sense to 1. evaluate your mental health and maybe 2. start hoarding seeds, salt, solar panels, ammunition, and gold.

And how much do you believe your actions influence the outcomes? It’s easier to bet on outcomes when the dice are loaded. If you believe The Machine is salvageable you probably should be doing something to support the Elizabeth Warren set, the people who can get The Machine back on the rails. Or perhaps you feel it is a moral imperative to starve The Machine. We might call this the Shaker or Amish solution. (Although another kind of growth is baked into the Amish solution, which is why there are so many more Amish these days than Shakers…) The last, most dangerous group would be the people who believe they can speed along the breakup of The Machine.

In a universe of increasing entropy it’s much easier to break a machine than to keep it running. If you truly knew when and how abruptly The Machine would stop, you might not even fear debt, and you could become a sort of evil anti-mustachian. You might amass fortifications, solar panels, batteries, and an armory of rail guns all on credit, and when you brought the Machine down, no creditor would dare come to collect. Or maybe you are Saudi Arabia, throwing oil billions around while oil is still a thing, hoping to buy enough influence to get some nuclear reactors built in your kingdom before the winds shift much further. We should beware of people who spend other people’s money on dangerous things.

It’s not irrational to make some preparations for chaotic times. A hobby of seed saving and perennial propagation would be a benign eccentricity in a Star Trek future, and potentially very valuable to your community in a disaster. I don’t think we should begrudge anyone of any stripe this kind of small-potato security.

But a few people have benefited so much from The Machine that they can throw billions around. They have gone well beyond Mustachian retirement money and into change-the-world level money. That’s the kind of money that could make a real positive impact bolstering democracy around the globe, averting famine and disease, or building a really swanky bunker in New Zealand.

When we make preparations for an uncertain future we are making bets that reveal our optimism or pessimism. We are also sending very real signals about our values. Whether or not we bet right, the values are unlikely to change. Someone working for the public good today will probably work for the public good tomorrow, whatever tomorrow ends up looking like. By the same token, if someone is preparing to “win” a bloody apocalypse, you can infer that their motives today, and in any future, may be less than selfless.

If you are building a community for the future, start looking for the volunteers and the seed savers now.