Book Review - Rory Sutherland - Alchemy
Rory Sutherland’s “Alchemy”1 is an unusual book. It advocates to use less logic, but “some magic” in finding solutions to problems. What’s interesting about it, is that it might actually make sense.
On The use and abuse of reason
Sutherland stars with presenting some reasons, why always being logic can get you in a worse situation. For example, when you are wholly predictable, people learn to hack you. You are not a good strategist, if the other side can anticipate every move you make. The application of logic will get anyone to the same place. In negotiations, it pays to have a reputation of being somewhat irrational. When your business partner fears that you will immediately leave the table, when they make a low offer, they might resist doing so.
Entrepreneurship and science are also areas, where people fool around and sometimes make great innovations, which were not predictable before they were made. Antibiotics and the smartphone are good examples of that. But unfortunately, smart people are often invested in appearing smart to other smart people So they will probably rewrite the real history of an invention to sound like a nice, linear story.
If you are a technocrat, you’ll generally have achieved your status by explaining things in reverse; the plausible post-rationalisation is the stock-in-trade of the commentariat. Unfortunately, it is difficult for such people to avoid the trap of assuming that the same skills that can explain the past can be used to predict the future. Like a criminal investigation, what looks neat and logical when viewed with hindsight is usually much messier in real time.
For Sutherland, finding something that works without necessarily knowing why it works is just a normal process.
Evolution, too, is a haphazard process that discovers what can survive in a world where some things are predictable but others aren’t. It works because each gene reaps the rewards and costs from its lucky or unlucky mistakes, but it doesn’t care a damn about reasons.
Given that anyone that uses logic and the same assumptions should reach the same conclusions and solutions, Sutherland makes an interesting conjecture: when we have been trying to solve problems for a long-time with “logic” and mathematical models, but failed, maybe these problems a logic-proof:
Similarly, if you expose every one of the world’s problems to ostensibly logical solutions, those that can easily be solved by logic will rapidly disappear, and all that will be left are the ones that are logic-proof – those where, for whatever reason, the logical answer does not work. Most political, business, foreign policy and, I strongly suspect, marital problems seem to be of this type. This isn’t the Middle Ages, which had too many alchemists and not enough scientists. Now it’s the other way around; people who are very good at deploying and displaying conventional, deductive logic are everywhere, and they’re usually busily engaged in trying to apply some theory or model to something in order to optimise it. Much of the time, this is a good thing. I don’t want a conceptual artist in charge of air-traffic control, for instance. However, we now unfortunately fetishise logic to such an extent that we are increasingly blind to its failings.
A good example of such failings is the misapplication of mathematical assumptions to “repeated games”, which are a staple of economic theory. For example, take a repeated game, where you gain 50% of what you have, when you win on a coin flip, but lose 40% of what you have, when you don’t. You calculate the expected value (which is like “simulating” what happens when thousands of people do it thousands of time and average the outcome), which is positive. But unfortunately, assumptions about the independence of the individual trials do not hold. And that means that the ensemble perspective is not equivalent to the time-series. The most likely outcome for you individually is to go broke, but some will get fantastically rich. This distinction has implication for the behavioral sciences, because it means that some purported biases might in fact be only so from an ensemble, but not from a time-series perspective. What happens on average when a thousand people do something once is not a reliable guide to what happens when one person does something a thousand times. In this sense, 10x1 is not always 1x10.
This is often disregarded: for example, investments in railways often assume that 40 people saving an hour ten times a year is the same as one commuter saving an hour 400 times a year. But the latter is a life-changer.
But it is conventional and easy to disregard that distinction: “Many people pretend to be data driven, but lacking insights about the inner workings of the model they use or the context of the data, they box themselves in.”
For Sutherland,
“Logic should be a tool, not a rule. At its worst, neo-liberalism takes a dynamic system like free market capitalism, which is capable of spectacular creativity and ingenuity, and reduces it to a boring exercise in ‘how we can buy these widgets 10 per cent cheaper’. It has also propelled a narrow-minded technocratic caste into power, who achieve the appearance of expert certainty by ignoring large parts of what makes markets so interesting. The psychological complexity of human behaviour is reduced to a narrow set of assumptions about what people want, which means they design a world for logical rather than psycho-logical people.”
Why magic still exists
Sutherlands states that there is room for “magic”, for solutions that work, but do not necessarily make sense in the first place. A major reason this is that “We don’t value things; we value their meaning. What they are is not determined by the laws of physics, but what they mean is determined by the laws of psychology.” A good example of that is Prussia that managed to bring rich families to swap their precious gold jewelry for iron jewelry, which was stamped. This became a badge of honor, because rich people could show that they were patriotic and rich enough to give their jewels away. They pulled the same trick with the potato, which was despised by the population. Once the king planted potatoes in a royal garden and posted guards, it became highly demanded. The same trick works today with regional branding, like “Champagne”, “Parmigiano” or “Schwarzwälder Schinken”. That conveys a sense of scarcity for which people are eager to pay.
Signalling
Signalling, is the need to send reliable indications of commitment and intent, which can inspire confidence and trust. Cooperation is not possible, unless a mechanism is in place to prevent deception and cheating. Often, some degree of efficiency has to be sacrificed in order to convey trustworthiness and build a reputation.
Studying four years to pass the cab driver exam in London might be inefficient in times of GPS, but putting that much effort into becoming a driver makes it highly unlikely that a driver will defraud you. Nature also uses costly signalling, like the peacock, to inform potential mates about genetic qualities.
When a company creates a brand, it makes an investment in a recognizable identity that promises long-term or repeat relationship. Sutherland hypothesizes, that this might be what makes capitalism work. Creating a brand shows that a company intends to do repeat business and customers can differentiate products and give feedback. Sutherlands tells the story of Soviet ships breaking apart, a problem that was not solved, until the companies manufacturing the rivets had to stamp their name on the rivets. This closed the possibility of meeting quotas by simply lowering the quality of the product. For him, branding isn’t just something to add to great products, it’s essential to their existence. Without it, there is a race to the bottom.
Subconscious Hacking
Because “[we] often cannot alter subconscious processes through a direct logical act of will - we instead have to tinker with those things we can control to influence those things we cannot or manipulate our environment to create conditions conducive to an emotional state which we cannot will into being.” In short, we have to hack ourselves. A powerful tool to do so, Sutherland argues, are placebos. “For something to be effective as a self-administered drug, it has to involve an element of illogicality, waste, unpleasantness, effort or costliness.” There are placebos for bravery, like uniforms, medals and marching. There are also beautifying placebos. Trillions of dollars are spent on beautifying products like designer cloth, cosmetics, etc. Globally, more is spent of female beauty than on education! If you think about what alternatives that money could go to, maybe considering pride as a sin had something to it. Anyway, using such placebos serve any combination of signalling to the opposite sex for reproductive fitness, to other females for status, or to oneself.
For Sutherland, placebos are just another form of mood-altering goods we consume: alcohol, coffee, tea, tobacco, entertainment, drugs.
After all, it’s not just the purchase of cosmetics that can be explained in this way – I would contend that a large proportion of consumerism is designed to achieve the same thing. In fact, much luxury goods expenditure can only be explained in this way – either people are seeking to impress each other, or they are seeking to impress themselves. Is almost everything a mood-altering substance?
Satisficing
We are not looking for optimal solution, instead we are satisficing. That is what decision makers do in situations, where an optimal solution cannot be determined. Problems are often either computationally intractable or we lack information, both of which precludes mathematical optimization. Decision makers can find satisficing solutions either by finding optimal solutions for a simplified world or by finding satisfactory solutions for a more realistic world.
We often try to find solutions that will not get us into too much trouble. Brands, in this view, are a great way for consumers to aid finding satisficing solutions: They do not choose Brand A over Brand B because they think Brand A is better, but because they are more certain that it is does not stink horribly. Knowing that people will pay a disproportionately high premium for the elimination of a small degree of uncertainty might explain a lot of consumers’ decisions. “When people snarkily criticise brand preference with the phrase, ‘you’re just paying for the name’, it seems perfectly reasonable to reply, ‘Yes, and what’s wrong with that?’”
Psychophysics
Psychophysics “is essentially the study of how the neurobiology of perception varies among different species, and how what we see, hear, taste and feel differs from ‘objective’ reality.”
The lesson is that we are dealing with the perception of things, not with things themselves. For example, a TV screen has only 3 colors. Blue, red, and green. This cannot mix to make yellow. But our eyes only have receptors for these three colors and stimulating them with a specific combination of green and red makes us perceive it like it is yellow light.
We act on what we perceive. Which creates a rather interesting situation. Because, we cannot describe someone’s behavior on what we see, or what we think they see, because what determines their behavior is what they think they are seeing. In this sense, perception may map neatly on to behavior, but reality does not map neatly on perception.
For example, companies had to learn through repeated failure that costumers prefer sales and “rebate as event” to consistently low prices. It can also lead to grave errors. For example, environmentalists seems not content with people doing the right things, for example reducing CO2 emissions, but they insist that people must do the right thing for the right reason, for example out of altruism or out of fear of a climate catastrophe, but not out of a profit motive. They also insist that the right things are done in the right way. Using natural gas or nuclear power to get results is verboten. This limits their effectiveness.
How to be an alchemist
For Sutherland, there are ample opportunities to act like an alchemist. For example, there is some research that shows that people succeed at logical question, when they are framed as “fraud detection”, but fail at the same task, when they are expressed as pure logic.
Given that people use satisficing answers that don’t get some fired, we repeat the same stuff over and over. Allowing ourselves and civil servants to be a little bit foolish might help us to not make stupid mistakes.
The UBI is for Sutherland a good example of reframing a political thought in a way that makes it more attractive. It is a form of redistribution, but strangely many people that are identified with the economical right, like Milton Friedman, favor(ed) it. Sutherland hypothesized that:
Perhaps protests against wealth redistribution are essentially, like most political opinions, merely an attempt to add a rational veneer to an emotional predisposition. People on the right instinctively dislike most welfare programmes, but UBI is paid equally and indiscriminately to all, which means there is no incentive for claimants to exaggerate their own misfortunes in order to benefit. UBI also preserves differential incentives to work: if one man lies in bed all day and his neighbour goes out to a job every morning, the worker will be richer than the layabout in proportion to his effort. Finally, UBI does not allow the ruling political party to bribe its own supporters at the expense of people who don’t vote for it.
Finding solutions that solve a problem in another frame that makes people like the solution more, leaves a wide-open field for political alchemists.
Rory’s “Rules of Alchemy”
Sutherland provides a few short rules to make some “alchemy” happen:
The opposite of a good idea can also be a good idea.
Don’t design for average.
It doesn’t pay to be logical if everyone else is being logical.
The nature of our attention affects the nature of our experience.
A flower is simply a weed with an advertising budget.
The problem with logic is that it kills off magic.
A good guess which stands up to observation is still science. So is a lucky accident.
Test counterintuitive things only because no one else will.
Solving problems using rationality is like playing golf with only one club.
Dare to be trivial.
If there were a logical answer, we would have found it."
Summary
Alchemy is a refreshing and entertaining book with a lot of food for thought. Sutherland doesn’t claim to have all the answers, but he ask some really great qustions.