Chapter 2. The Cretan Method

The search for truth is an ancient, and difficult pursuit. Modern society still swims in a sea of lies, in politics, business, and especially the software industry, where grand lies breed like cults. My work is to cut through the lies to develop better theories of the truth. It turns out that lies cause pain, and as we approach truth, we also become happier. In this essay I'll explain how I do this, and provide a series of tools and lessons for your benefit.

Approaching the Truth

Humanity is a powerful eusocial species, and one of our superpowers is the collection of knowledge about the real world. We do this by building models, or theories, refining them through practice, and teaching them to our young. Of all the layers of reality that we can model, the one that affects us most, every day, is the reality of other people. Karl Popper wrote: "We are social creatures to the inmost center of our being."

There is an art to shaping, testing, and applying theories. The truth is a curious thing. Like an irrational number, it exists, is not negotiable, nor subjective, yet unreachable. Truth is an absolute property of 4-dimensional space-time. We can never reach the truth, only build theories that approximate it better and better.

We construct theories apparently out of nothing. We take observations and gut feelings, and the endless legacy of theories delivered to us by past generations. We design new, or improved theories, and we encode them in language and words to be argued, remembered, and shared.

Popper argued that there are essentially two kinds of theory. There are scientific theories, which can be falsified by data or observations, and there are magical theories, which cannot. To put this another way: you cannot ever prove that a theory is true, as the truth can never be reached. You can however try, and fail, to prove a theory wrong. When you remove all the provably wrong, what is left approaches the truth.

All theories are inaccurate to some degree. However, falsifiable, scientific theories have an interesting property that magical theories lack. That is, they can be evolved towards finer and finer approximations of the truth. Here are a set of theories you will recognize:

• The ratio of a circle's circumference to its diameter, Pi, is constant.
• Pi has the value 3.
• Pi has the value 22/7.
• Pi has the value 3.141592.

These are all falsifiable, and cannot be proven true. Take the first theory. We can measure as many circles as we like, and each time we'll get approximately the same answer. We won't get any data that disproves the theory. We won't get any data that proves it. So far, so good. The second theory looks good, and is trivial to nullify. See, wrongness is scientific! The second theory is better, yet as we improve our measuring tools, we'll start to see that Pi looks more like 3.141 than 3.142. The third theory is much better, and it takes a lot of work to nullify it. There is no theory of Pi we can absolutely state is true. We only have increasingly accurate models.

Now here are some magical theories you may recognize: "C++ is great because it has a Standard Template Library," and "Java is great because it stops programmers from making bad mistakes" How can we falsify those theories? Impossible. They are not only not true, they are not even wrong, as Wolfgang Pauli would say.

A scientific theory is always wrong to some degree, and can be improved by reducing its wrongness through testing, observation, and measurement. A magical theory is untrue, and cannot be improved.

A Theory of Lies

To search for the truth, it is worth understanding the nature of deception. The Greeks thought (or pretended) that deceit was an evil spirit called Apate, who escaped when Pandora opened her infamous box. Sun Tzu wrote that "all warfare is based on deception," meaning that deception is a successful strategy in a conflict.

Most minds are honest liars. We can assume that lying is a built-in function of the human mind. The ability to lie, or at least bluff, is essential to develop theories of the world, as all theories are lies in some degree. Only a dysfunctional mind can never lie. When a mind shapes a theory, it remixes the existing theories it knows with new observations, add its approximations, guesses, assumptions, and beliefs, trying to connect those into a consistent story. Most minds do this slowly, over weeks or years. In such theories, the lies are temporary scaffolding that can be fixed over time.

Some minds, in a constant state of war with the rest of humanity, lie strategically, to confuse and manipulate. Such minds can construct theories that are entirely magical, so rapidly it happens in real time, and tell these to the listener, as utterly convincing lies. This is an assault, a weapon of war. We can call such minds "psychopathic", as their goal is to prey upon others. In their theories, the truth is temporary scaffolding, to be replaced over time by fabrications.

The value of the lie, as a weapon, is clear. In a predator-prey relationship, the predator must keep its prey confused and immobile so it can safely feed. With humans, emotional restraint is more effective and less risky than physical restraint. By injecting magical theories into the prey's mind, the predator disrupts the prey's theory process. This is one way cults trap their victims, by injecting large magical theories that disrupt logical thinking.

The first practical lesson is this: everybody lies, as explained by Dr. House (or rather, his script writers). I assume there are honest liars (like me, I assure you), and psychopaths. I'm skeptical to the point of mocking the entire universe unless and until it shows me data. And even then, the best it can do is "wrong", rather than "untrue".

Mind as an Evolved Strategy

Watching my 3-year old son learning, and then playing Minecraft, was a revelation. I've always treated my children as proto-adults, works in progress and yet production ready. If you've not seen Minecraft, this game is a construction simulator that is simple, deep, and social. It is the best model I've seen so far for my ideal software development process.

The old theory of "mind as clay" to be shaped by society has been nullified by twin studies. We know that our minds are largely defined at birth. "Mind as an evolved strategy that is shaped and calibrated by society" seems much more accurate. I'd like to see studies of twins raised in dramatically different cultures. Does growing up in the streets of, say, Kinshasa, calibrate your paranoia rather higher than for instance, London?

Clearly, our mental tools are sharp and functional from a very early age. From about one year old, children begin to interact with people other than their mother, and they mostly do this accurately. If you want to understand a concept like "trust", observe young children. Children start by trusting no-one except their parents and siblings. They extend trust to other adults, when their parents tell them it is OK. They trust other children implicitly, unless there is too much age difference. They do not trust unknown animals unless their parents are with them.

Similarly for a concept like "freedom", which has inspired billions of written words, and yet I've summed up as "the ability to do interesting things with other people." This is obvious from short observation of a typical child, and yet it's profoundly valuable as a working theory. I've talked much more about this in my book "Culture and Empire: Digital Revolution."

The child's world is necessarily simple, and the child's mind is uncluttered with magical theories. We are already powerful theoreticians at a young age. Children are natural scientists.

This brings us a second practical lesson: seek childlike intuition. The tools we need to understand the world are built-in, and evolved by long and relentless competition in the knowledge business. We are good at developing theories until our minds become poisoned by magical thinking.

Pain is Valid Data

One thing about children is how unfiltered they are. When they don't like something, they will tell you. You will get sullen faces, loud complaints, even screaming and crying. It can be frustrating to a parent. It is however fascinating, when compared to how much irritation and pain adults will accept without even blinking.

All theories are smoke, until you test them. This means, apply to reality and observe the results. A good theory works smoothly and almost silently. A poor theory creates what I'll call "friction", expressed as irritation, cost, delay, stress. For example, there are two conflicting theories in my household about how best to get to school:

1. The best way to school is by car (3 minutes' drive).
2. The best way to school is by foot (10 minutes' walk).

This morning, as we were late and felt lazy, we took the car. It took 20 minutes, most of which I spent fuming at the traffic and bad drivers. Then I realized, every moment is an experiment, and pain is valid data, and I was happy again. The car theory is nullified. We can walk again.

When you use a theory, and remember, all theories are wrong in some respects, you will always be able to feel some irritation. Often the irritation is washed out by larger concerns. It could be, "OK, I'm stuck in traffic yet at least I'm protected from the pouring rain". Or, "OK, my boss is an ass to me, yet at least I have a job."

Actual mental pain (as compared to sitting on a piece of broken glass) is a sign of serious friction, like outright lies. Minor irritation indicates small friction, like inaccurate assumptions. Every theory is either scientific, and by definition wrong, and can be improved, or it is magical and irreparable. A magical theory will cause pain when you try to use it for anything serious.

This brings us to lesson three: improve your theories, or discard them. The key indicator of a magical theory is that you cannot remove friction by improving the theory. You either accept the theory totally, or not at all. You can treat magical theories as a form of infectious mental disease.

The Abusive Bond

You will remark that people often stay in abusive relationships far longer than you'd expect. When we started the ZeroMQ project the core developers wanted to use C++ rather than my preference, C. They told me, "sure, it will take you ten years to learn the language, yet it's far more powerful." At the time I didn't have arguments against that. It took them five years and one real open source project to do a U-turn and discard C++.

Abusive attachment is a counter-intuitive mechanism, and worth understanding. We value relationships proportionally to our investment of time, effort, money, affection, resources. In a normal relationship this works two ways: Alexandra and Bob exchange gifts of varying subtlety, and do mental calculations to balance the books. In a "healthy" relationship, the balance is close to zero, and the calculation effort signifies the depth of the relationship. In an "unhealthy" one, there is a severe debt on one side (and the calculation effort will approach 100% of CPU, acting as a DoS attack).

There is a cheating strategy, based on future promises. If Mallory promises a large future payoff, then Bob will invest on that basis. This is the basis for many con games including the Spanish Prisoner aka Nigerian advance-fee fraud. It is the basis for most abusive relationships.

Here is how it works: Mallory makes a large promise to Bob, and offers a small gift as proof of good intentions. Bob responds with gifts, and Mallory accepts them to encourage Bob. At every point, Mallory fails to deliver, always due to tragic outside circumstances. Mallory flatters Bob and acts the victim, and asks for token investments from Bob.

In Bob's mind, the relationship is deepening and gaining value. On the ledger sits a large future promise from Mallory, and all of Bob's investments, and it feels real and deep. Bob develops an increasingly strong attachment to the theory of Mallory's future promises and instead of backing away, makes increasingly large investments.

Mallory moves from blaming third parties to blaming Bob. Everything starts to be his fault. She rewrites history to explain how he is the cause of all her troubles. Bob accepts these debts and the ledger starts to tilt massively against him. The future promise gets lost and forgotten. Bob works overtime to pay off his "debts" and to normalize the relationship, yet just enables progressively worse behavior from Mallory. The more abuse Mallory gives to Bob, the more Bob invests, and the more he invests, the more he values the relationship and is bonded to Mallory. Of course Mallory doesn't give a fig, and has a negative bond to Bob, consisting mainly of scorn and distaste.

This bond can last years, even a lifetime. Seen from the outside, it is incomprehensible and immoral. Yet it is just the price we pay for our eusocial powers: it enables a class of predators, the Mallorys. Mallory isn't always a person: it can be an organization, or a set of magical theories.

There are only two ways out of an abusive bond. One, when Bob has nothing left to offer Mallory, who will then discard Bob like trash, while explaining in great detail to everyone how it was Bob's failure. Two, when Mallory makes a demand that Bob cannot normalize. Bob may wake up at that point, or may self-destruct.

The evolved response to the Spanish Prisoner attack, in our minds, is to discount the future massively. However that is also trivial to beat: simply add enough zeros. Hence modern scams are always promise ridiculously large amounts of money or power.

Theories that are excessively complex, and promise future payoffs are a form of advance fee fraud. When it takes ten years to learn a programming language that promises you "power", you realize C++ users are in an abusive relationship with their language. C++ is the Scientology of programming languages. Java, the sprawling mass religion.

The curious thing is that when you challenge someone who is in the Spanish Prisoner's embrace, they will fight you. To question such a massive, life-threatening investment is felt as an extremely hostile act. Only when the embrace is breaking will the Bobs nod and agree that perhaps they are lost.

This brings us to lesson four: people defend magical theories the strongest. There are very few theories in the software world that are not fraudulent in the same way. Most software is based on magical theories, and most programmers are Bob. We almost totally lack the scientific method in software (the ZeroMQ C4.1 process is a rare attempt at this).

The Role of Emotions

Children may seem highly emotional yet look closely and you see that most children can switch their emotions on and off at will. It's been said, all children are psychopaths. (A better theory is: all psychopaths are childlike.) Children lose this ability as they develop empathy. Emotions are social communication tools, a way to manipulate others into behaving in the way that we want. They are our original, primeval language, displayed in face and body.

This is easy to demonstrate. If someone walks in your way on the sidewalk, you sidestep, smile, or nod, and the nascent irritation (the tiniest ripple, shown perhaps as a raised eyebrow) turns into a tiny pleasant interaction. However the same interaction between two cars can often lead to intense anger in both drivers. The difference is that cage that the car forms around the driver, cutting off verbal and non-verbal communication. It is easier for a car driver and a pedestrian to understand each other than two car drivers.

Lacking any response from the growing irritation, the brain flips into the "fight or flight" response you might have when someone deliberately walks in your way on the sidewalk. Road rage is a basic survival instinct caught in the wrong context. A sharp response to a threat is safer than no response. Except of course, there is no threat.

Some people (the Mallorys I spoke of earlier) retain their childlike ability to flip emotions into adult life. They project fake emotions -- jealousy, hate, fear, anger, self-pity, sadness -- to drive others into Spanish Prisoner embraces. Look, I'm insanely jealous! That shows I love you, now give me more affection!

More interesting though, is our own use of emotions when we're confronted with a poorly-working theory. We will often express the pain and irritation as anger towards others. As shown by my "best way to school" example, emotion is valid as data, yet invalid as a process. Shouting at another driver is not an Aristotelian dialogue.

Emotions make great art, and terrible science. In fact we can measure the amount of magic vs. science in a process by the level of emotions. We noticed this clearly in the ZeroMQ community: as we moved from a magical to a scientific process in early 2011, all emotional arguments disappeared.

I'll explain later how to regulate your own emotions, which I call "Grounding". It is a difficult technique, yet extremely helpful in taking distance. And taking distance from your own experience is, ironically, the best way to understand it clearly. Pain is useful data only when emotions are silent.

To see friction, you can either observe pain and stress in other people, or in yourself. Observing others without responding to their emotions is already hard for non-psychopaths. And psychopaths cannot understand emotions, only read and mimic them. (I guess, without hard data, that they are terrible at science.) To observe your own experiences without engaging with your emotions is extremely hard. Yet there you are, 24/7 inside your own head. If you can develop this skill, you can see friction in practically any situation simply by engaging with it.

This brings us to lesson five: you are your own best instrument.

The Core Process

We've collected enough axioms and theories of theory to talk about the core process itself. If you are a regular reader of my work, you'll know this process already. Like many good theories, it's built on successful practice. The process seems simple:

• Observe friction in a social setting.
• Discover or guess the underlying theories.
• Identify the flaws in those theories.
• Improve the theories, or discard.
• Test your improved theories.
• Repeat until it gets boring.

We've seen that any scientific theory can always be improved. There is no "local maximum", only ever increasing accuracy that you can reach by spending effort and time. What is more surprising to many people (perhaps their mother told them often how special they were) is that this process is entirely mechanical. To put this another way, individual intelligence is somewhat overrated.

To value individual human intelligence is like looking at an ant colony and saying, "look, that ant is super smart!" It is magical thinking. Like ants, we're a collective species and we think in groups, not individually. The smartest ant ever to exist, working alone, is nothing compared to a few "ordinary" ants working together.

Actually, it's even worse than that. Smart people often use their intelligence to compensate for friction. The ability of humans to rationalize even the worst situations can be impressive. If there is a type of individual intelligence I value, it is the ability to sense friction and be annoyed by it. Even that depends on other people: there is no social friction in solitary.

Thus the process I just described is not quite that simple. It only works as a group exercise. I've observed a fairly typical, almost ideal cycle in our thinking process: learn from others, test the new knowledge (play), apply it to real problems (work), then teach this to others. I call this learn-play-work-teach (LPWT) for short.

LPWT has some interesting properties. First, it is innate. You will see this emerge in young children without coaching. To take Minecraft again, the process is: learn from watching YouTube videos of other kids playing, then practice alone to make sure the gained knowledge works, then play the game with other children, then teach other children the things you learned. Children learn Minecraft from other children. No books, no schools. Yet is an extraordinarily rich and deep knowledge.

Second, it works very well. This is how my open source projects build software. After many decades of refining our processes, we arrived where my kids are. We write software in tight LPWT cycles. The software is never perfect, yet it is perfectible, a scientific theory that is always wrong, yet never untrue.

LPWT is orders of magnitude more effective than classic one-way learning. It underlies a whole theory of self-organization around problems that I've addressed several times in my writing.

And LPWT barely feels like work. It is pleasurable, almost addictive. It is a curious thing to see my professional life arc back to a childish view of the world. Rather than this being a natural journey, it seems forced. Indeed, I wonder what happened to me during all those years.

Ah, yes, mass education and mass work, the curses of the industrial era along with mass media. At some point in the 19th century, the old LPWT structures broke down and were replaced by strictly separated schools, universities, and workplaces. We play as children, we learn as youngsters, we work as adults, we retire and die when elderly. Teaching is a kind of work, for a few.

Play, learn, work, die. The four phases of life. The four-lives theory has many problems apart from just making people unhappy. Let me list a few of them:

• It lets us contribute for around 45 years, 60% of our lives. Does that sound like a lot? Through the guise of a "video game", my 7-year old is teaching my 4-year old about architecture, physics, chemistry. It looks to me like LPWT lets us contribute from the age of 3 or 4 to when we die.

• It excludes large segments of the population. The high cost of education (in terms of time) favors those who have nothing better to do, which means young men. It can be very difficult, in many cultures, for women to invest in higher education.

• It favors magical theories over scientific ones. If you are going to spend several years learning knowledge, without any way to test what you are learning in real life, you are already playing Spanish Prisoner. Magical theories will be more attractive, as they can simply lie to you ("yes, learn us and you will become much wealthier!").

• It treats the elderly as waste material. The very concept of "retirement" is the sign of a hate-hate relationship between employee and employer. With LPWT, your age doesn't matter. If you can learn, play, work, and teach at four, you can do this at 84 (assuming you can hold a conversation).

• It cannot handle change and opportunity. If education is a prerequisite for work, then the already-employed cannot learn new things outside the scope of their work, except at a high cost.

• It divides the generations and ages. The working cannot teach the young. The young cannot teach each other. Instead, teaching passes through a layer called "the education system", which decides together with the political elites what knowledge it will teach. I trust you see the scope for monopolies of power and cult creation.

The 4-lives theory of society is visibly inaccurate, and generates endless friction that many of us experience for much of our lives. I think many, if not most, people stuck in this system dream of escape, a return to the freedom they felt as children. Some people do define their own lives. It is rare, and difficult, however. Society tends to frown on the itinerant and the opportunistic.

This brings us to lesson six: society is saturated with magical theories. That nagging voice in the back of your mind that much of your life is profoundly wrong is spot on. Modern technology and cost gravity compensate somewhat, yet life could be so much better.

LPWT is very alive, outside the formal education and employment systems. It is how we learn on-line, it is how mass web forums work. Since everyone can contribute at full speed, it is an extremely powerful collective learning approach. When people use it in anger, as Anonymous did, they become a political force and a serious threat to the established power structures.

Bedtime Stories

"Once upon a time there was a magic castle, high in the mountains..." "What color was it?" "What does it matter? OK, let's say it was white. Anyhow, in this castle lived a princess, all alone..." "All alone? How could a princess live alone? Who made the food? Who cleaned the castle?" "Look, do you want to hear the story or not?" "OK, OK, I'll shut-up."

I speak at many conferences. My style has changed over the years. I used to use slide shows, like most speakers. Neat slides highlighting key points, invoking emotions, telling my carefully constructed story. These days I come without slides, and without a script, and instead of telling my story to the audience as a monologue, I improvise a dialogue with them.

The dialogue is hard work, exhausting, yet it is easier than making slides. I've switched to that format for a simple reason. When I used monologues, perhaps a few people in a hundred would "get it". Everyone would see what I was explaining. Very few would believe me. With a dialogue, I can get half the room, or more, to get it, and react.

The dialogue is an ancient form, almost lost in modern life. We're so enamored of technology that we have forgotten this pattern. Here, watch my video. Click on my links. Friend me on Linked-in. But god forbid I ask you to negate my hypothesis, live, before an audience.

The reason I go to conferences is not to sell, or evangelize. It is to learn, as the dialogue is essentially how we do the "test" part of the core process. Here is my theory, let me explain, you tell me if you have data or observations that falsify it. If I can tell a theory to five expert audiences, and it isn't negated, then it stands. For this process to work, I absolutely have to use the dialogue.

However, and this explains the love of most speakers for their slides, if I'm trying to convince you of a magical theory, then the last thing I want is a dialogue. Instead, I invoke a different childhood scenario, the bedtime story. At bedtime, we accept any fairy tale without question. We don't care what color the castle is, we are half asleep already, and dreaming of the high walls.

This is the power of the monologue: to send your audience into a sleep state where they will accept anything. It's well known by political speakers and preachers, by lecturers and cheap salesmen. The trouble as usual is that we have some immunity to such simple manipulation. We wake up, and then the dream is gone. The best salesmen listen carefully to their public before crafting lies that will appeal to them.

Conference slide shows cheapen the fairy tale further, by reducing it to 2-dimensional visual fast food. Here, a buzzword. There, a cute photo. Here, some memorable words! As entertainment it's fine, even elegant. As a teaching and learning tool it is tragically bad. World-class speakers with decades of experience travel to far locations, just to play text-to-voice synthesizer for 45 minutes. If you go to conferences, I bet you remember random images, yet not the stories. "The best conversations happen in the corridors." That is a statement of failure.

So we come to lesson seven: the best pattern for learning and teaching is the dialogue. More precisely, where one person presents a theory, where others try to negate it, and where spectators applaud and encourage. It sounds confrontational, yet if there is sufficient freedom to participate, there is no need to get it right first time, and thus no shame in being wrong. Indeed, when someone takes your work and improves it, the feeling is one of satisfaction, not embarrassment.

This is how I like to structure my software projects. It's how I organize my workshops. It's how I'd run a conference (and will, in 2015 if all goes well): multi-hour sessions where speakers present their theories (of technology) and ask the audience to negate them. There are many other things broken in the conventional conference model. Meat for another essay.

Simple is not Easy

I've heard complex designs described as "over-engineered." It's more accurate to describe complexity as "under-engineered." for it takes real work to turn complexity into simplicity. I can describe this process, as it is how I write software.

You take an existing theory, and you apply it to new problems. This generates friction, which you can resolve by extending the theory. You can do this over and over, and the theory will become wider and more complex. It is like adding more ingredients to a pizza.

If you are inured to friction, as some people appear to be, then it does not matter how large and messy the pizza gets. You'll just clear space on your mental table. However if you're easily irritated by friction, like me, untamed pizza complexity becomes a problem to solve. You stop adding to it, and you build a new theory on top, an abstraction that has the same effect and yet is significantly simpler.

Instead of "pizza topped with tomatoes, mozzarella, anchovy fillets, and capers", we get "pizza Napoletana". Now we can mix a theory of sizes ("small", "medium", "large", "Americano") with a theory of recipes. It is easier for both customer and restaurant to use such abstractions, to test them against real pizzas ("I did not ask for onion!"), and to improve them.

Karl Popper wrote, "science may be described as the art of systematic over-simplification". Lies tend to hide in complexity, just through a process of elimination. The simpler the theory, the easier it is to negate it. If I explain a theory in ten words, that is much easier to negate than a theory of a thousand pages.

This brings us to lesson eight: simplicity always beats complexity.

Resolving the Gender Gap

The observation that women make better scientists than men is not new. So the gender gap in some areas, like software, is confusing to many people. I've come to believe the answer is that the mass of software engineering is simply not science. It has been this way since I started programming: the vast bulk of software written and used is based on magical theories.

Most software projects fail. This is not engineering. It is a waving of the hands over duck entrails. Oh yes, we all pretend we know what we're doing. This industry is always confident, especially when it has no clue. The truth is, we barely cling to the raft of tenability in a sea of failure.

As I explained already, magical theories are discriminatory. They take time to learn, years of study and practice, spread out over a decade or more of young adult life. Not everyone can make this investment. And not everyone is stupid enough to accept magical theories without a severe "WTF is this?" mental gag reflex. In particular, most young women are too busy building real social networks to dedicate years of their lives to the study of nonsense.

There is a reason the temples of technology are filled with young not-quite employed men. Magical theories are as attractive to young men as they are distasteful to young women. Let's compete on who knows the most arcane facts. Let's see who wins the popular vote. Let's beat each other in argument, not by using the scientific method, rather by quoting magic at each other. My cloak of invisibility beats your sword of fire!

And clerics build pyramids schemes of confusion that promise "just accept this magical theory, and teach it to others, and you too will gain power." This is how cults work. One might argue that at least technocultism absorbs the energy of that stupid-simple slice of society young men represent. However, it does seem a waste.

So we come to lesson nine: real science welcomes everyone, no matter their age, gender, or origins.

Intruder, Identify Thyself!

There are endless magical theories that irritate us. One that struck me was the theory of on-line identity. The Internet has become a tracker of our personal profiles. This is based on a falsehood, which I'll explain. It's a good example of how a few people can use magical theories to profit at the cost of everyone else.

In real life, identity has depth. We start as strangers in a crowd, and we expose our identity gradually, as part of establishing a relationship. I'll tell you mine if you tell me yours. To walk around with our visible name is a strange thing. We do it at conferences and weird parties. It feels forced and unnatural, because it is. Anyone walking in the street wearing a name badge is... someone to avoid.

When we start with identity and then build a social network around it, as happens on the Internet, the results are weird and irritating. It would work if we were all narcissists with an agenda to collect followers. It works for me, for instance, as I decided to make "Pieter Hintjens" my main product, some years ago.

However to operate effectively, I also maintain several other identities. Different crowds are different worlds. The identity-comes-first theory has several large problems:

• We're forced into a "do I use my real name?" choice that we rarely if ever face in real life.
• It gives us no way to divulge our identity gradually on a mutual basis.
• It forces us to use aliases for different contexts.
• It cheapens our relationships to the point where "friend" has lost its meaning.
• It allows centralized brokers to effectively own and tax our relationships.

These problems go beyond irritation into full negation of the theory. The term "Facebook divorce" gets 145M hits on Google.

As with other magical theories, the flattening of identity is also discriminatory. It promotes the trolls and attention-seekers and obscures the modest, the calm, the reasonable. It is the reason we get such inflamed and wasteful arguments about gender and race and politics on our forums. Pseudonyms make no difference. A karma whore would smell as bad, by any other name.

So we come to lesson ten: the core concepts of our Web are magical theories, which cannot be fixed.

Conclusions

This essay is an explanation of a personal approach to learning that I call, lacking much imagination, the Cretan Method. Everything I've written is wrong.

I've explained how society collects and transmits knowledge of the world in the form of theories. Some theories are wrong, and the rest are entirely untrue. We can always improve a wrong theory (also called a scientific theory) by testing it, feeling the friction its wrongness causes, and fixing the theory accordingly. We cannot improve untrue theories (also called magical theories).

We're born as scientists. Society bends most of us into believers of magic, and if we're lucky we escape that, and return to being scientists. Large segments of society are still based on magical theories. Over time, these will die and be replaced by increasingly accurate scientific theories. One example is the learn/work/play trichotomy of modern society. The intuitive, and much more effective approach is to mix learning, working, playing, and teaching into a smooth, continuous cycle.

I've suggested that the best way to test theories of social behavior is to use yourself as an instrument. This means being ready to measure your own experience without emotional responses. I'll explain the technique of regulating your emotions -- Grounding -- in the last section.

And I've suggested that while a monologue (the sales pitch or slide show) is a good way to sell magical theories, the best way to teach a scientific theory is the dialogue, where students search for points of friction with a theory, or even negate it if possible. A free and open dialogue can rapidly negate untrue theories, improve wrong theories, and generate entirely new theoretical abstractions. My belief is that all knowledge work would be best structured as a dialogue.

Finally, we come to a process that could be described as "pain driven development". Take a theory, test it against the real world, observe your own or others' pain, and use that to guide improvements in the theory. It is a mechanical social process that needs no special intelligence. If you really want to flatter the individual, perhaps some people are most sensitive to friction and faster to develop new theoretical models. And others are better at ignoring friction and surviving in a magical landscape.

Grounding

Grounding is to remove all negative emotions from your mind, leaving happiness and clarity. It is one of the most valuable things I've learned in my life. Over time I learned how to ground myself consciously, and rapidly. Unlike meditation, which requires peace and calm, grounding works even when people are deliberately provoking you.

Grounding has echoes in Cognitive Behavior Therapy, a technique developed by therapists who engage with the emotionally extreme minds described as "borderline". I suspect CBT is more useful to therapists, than to their patients. Grounding is simpler, and is meant for self-application in the living room and workplace.

The goal is to remain calm and happy no matter how difficult things become, and in that state, to accurately observe yourself and others. Only with accurate observations can you improve and negate your theories. Emotional turmoil can easily become a cage. So Grounding is a path to freedom and opportunity. It lets you experience the very worst situations as "useful data" rather than "destructive pain".

Let me first describe the Grounding process, and then provide more detail. We start by identifying our strongest negative emotion, and naming it. We then reverse engineer it, using the theory that all emotions are a form of social communication. We then find the root theory that the emotion is based off. We then negate that theory, in our own minds. The emotion then goes away.

This is not easy. Expect years of reading these words and not getting it. Some emotions are trivial to debug and fix. Some are extremely difficult. It all depends on the power of the underlying theories, which can take years to construct and years to dismantle.

Let me do a "Hello, World" walk-through of an emotional moment. Alex is on a date with Brita, and Brita says she has to go home soon. Alex is suddenly filled with self-pity and pain. He feels Brita is deliberately hurting him. He feels angry at her, hides it, and then when she says, "because my cat is sick and I want to check on him", feels guilty about having such a strong response to her words.

Classic Alex. Now let's debug this little story. The guilt comes from fear that she'll have seen his anger and react to it. The anger is a "fight or flight" response to the pain he feels. The pain comes from fear of rejection, and being alone. The self-pity is a cry for help, the emotion a baby feels when it cries for its mother.

Clearly Alex, who is mostly thinking about the possibility of sex with Brita, isn't going to be regulating his emotions much, if at all. Luckily, Brita was lying about her cat (in fact she's married and her husband is waiting for her) and she's also being flooded by doubt, guilt, and fear. She's not really reading Alex at all.

Many years later, a now wiser and less reproductively stressed Alex meets Brita again. They chat and have a coffee. Later, Brita once again says she has to leave. Alex feels the smallest kick of self-pity and conducts a small internal dialogue. "OK, self-pity, what's your theory?" he asks. Self-pity answers, "if you cry, she might give you a hug. Then, sex!" Alex mocks self-pity's theory. "Did that ever work?" Self-pity grudgingly shakes its head and goes away.

That is the general approach. Let me now list the main negative emotions and their classic patterns:

• Self-pity: a cry for adult affection. Self-pity will rapidly turn to anger as our imaginary adult fails to comfort us. The theory is, "I'm under seven years old." This is the easiest emotion to name and debunk.

• Jealousy: a cry for attention in competition with another person. Jealousy will also turn to anger as our target fails to respond. The theory here is, "I can bully this person into spending more time with me". It is harder to debunk, as it is often true for a while. However, a good negation could be, "it's not attractive to act like a child."

• Anger: showing your determination to fight. The underlying theory is, "showing my anger will bend the other person to my will." You can break this theory in various ways depending on the context. For example, "He did it by accident", or "there is absolutely no hope of saving this relationship", or "he enjoys seeing me angry", or "I can resolve this more profitably without risking a physical violence".

• Fear: showing your stress at a looming danger, such as being rejected and alone, or being hurt. The theory is, "if I show fear, mommy or daddy will come and save me." The simplest negation is: "no they won't, you have to survive this alone." A more subtle one is, "the danger you feel is false because X and Y," which applies to most social fears.

• Sadness: showing your stress at losing something precious. The underlying theory is, as with fear, "mommy or daddy will come and make things right again." We can debunk that theory quite easily in most cases. "Will being sad get X back? What are the chances, honestly?"

• Shame: showing our stress at being diminished before other people. The underlying theory is, "we will be ostracized, and then die alone and starving on a rocky hillside." It is easy to debunk, with: "frankly, no-one cares enough to ostracize you", though that will lead to self-pity. Perhaps to be more diplomatic, "everyone is so busy with their own problems that no-one will see your pants are down."

• Guilt: showing our fear of being shamed. The underlying theory is, "if we show how bad we feel about doing bad things, maybe people will forgive us." This is surprisingly easy to debunk, with, "everyone makes mistakes, you just apologize sincerely and carry on", and sometimes, "you didn't actually do anything wrong, the other person is manipulating you."

• Loneliness: showing our hunger to belong in a group or family. The underlying theory is, "people will want us because we look sad." And the negation is, "that works for babies only. For adults, a smile takes you a lot further than a frown."

It may sound strange to conduct an internal dialogue with your emotions. Yet this is the Grounding technique: debunking the emotion by explicitly stating, to yourself, observations and data that negate the underlying theory that the emotion is working on.

When you start, use pen and paper. Your mind won't be able to run the internal dialogue properly. So you search this list for your strongest emotion, and you write that down, and you then ask yourself "where does that come from?" Emotions often trigger other emotions, so you may have to unravel several layers to get at a root theory. When you do this more and more often, you find yourself Grounding yourself on-the-fly, so the layers don't build up.

It can also be helpful to practice this with other people, if they trust you enough to go there. "What are you feeling now?" can be an interesting starting point. If you ever find yourself acting as therapist for an emotionally abusive person, these techniques can save you from the destructive burnout that comes from engaging with unrepentant emotional storms.

More broadly, being grounded lets you meet and talk to anyone without engaging in the emotional games most of us play subconsciously. This includes psychopaths, who attack the emotional surface area of their targets. When grounded, you offer no opportunity for manipulation, and you become a "grey rock" that psychopaths will gloss over.

Thanks

This essay has been many years in the making, and is the result of endless dialogues with friends, and strangers who rapidly became friends. I'd like to thank everyone who's helped me figure these things out, particularly Sachiko, Tuuli, and my mother.