The problem with utilitarianism

Considering my posts on game theory, epistemology, etc. one might think I'm sympathetic to Utilitarianism as a fundamental moral philosophy. But I'm not!

The reason is basically the trolley problem:

Utilitarianism is pretty easy when applied to large abstract questions. For example, if you have to choose a public health policy that decreases mortality by 0.1% on 1M and another than decreases mortality by 0.1% on 5M people, of course you would choose the latter. But that quickly breaks down as you get more specific. What if one group is your family and the other is strangers? Would you start giving multipliers like 1 friend is worth 3 strangers? Such calculations are abhorrent. Or absurd -- a trillion shrimp in 100 years vs 1 human life today?

Now let's take not being abhorrent nor absurd as axiomatic for a fundamental moral philosophy. Not in a normative way. I just mean, here we are, and we do deal with good and bad. How do we do it? What do we really use as our deepest foundation? 

When you get down to it, it is the golden rule: Do to others as you would have others do to you. It is not a coincidence that versions of this principle appear not just in Christianity but throughout the history of human wisdom. My favorite (hat tip to Nassim Taleb) is the converse, aka silver rule: don't do to others what you don't want others to do to you. The silver rule, because it applies before action, is more fundamental.

The "greater good" abstraction generally works up to a limit, but if you stick to utilitarianism all the way, you get dystopian outcomes (I'm looking at you "Effective Altruism"). The real foundation of human morality is the silver rule.

The four anti-pillars of Marxism

Those who don't understand history are doomed to repeat it.... And those who do are doomed to watch in horror. 

In recent years, I've watched in horror Marxism seemingly coming back in fashion, as if we're in the 1920s. Too many people lack the practical experience, theoretical context,  or intelligence to know the lessons of the last 100 years. 

The seven pillars of wisdom in Christianity or the seven pillars of Islam are things you should build on. With Marxism, we have some seductive big ideas that you should not build on. I call them anti-pillars.  I won't explain them since many great people have before. I'll just list them as pointers.

1. Economic calculation problem and the fallacy of central planning.  

It is not just about having better computers or more data. The information in an economic system doesn't exist in one place or time, it is distributed. Furthermore, even if you had it all at once, much of it is not explicitly manifest but rather implicit preferences.

2. The labor theory of value is wrong

It is not just labor vs capital, there is also value and, more subtly there is quality.  If I spend an hour on the piano and Mozart spends an hour, it's one hour of labor. But there is a subjective difference in value. And an objective difference in price that others will pay for it. Finally, there is something that is neither objective nor subjective, beyond dialectics, namely quality.

3. The class struggle is a static one dimensional model, but human economic relationships are dynamic and multidimensional. 

This may have been subtle in the 19th century, but in the 21st century, it should completely obvious.  In fact even in the 20th century, class was often cynically replaced by ethnicity or religion by various revolutionary Marxists hungry for power. Lenin used "nations, nationalities and peoples" when he realized Russia was not industrialized enough for Marx's class structure. Ethiopian Marxist-Leninists adopted the same hack 3/4 of a century later. 

4. A political system based on class struggle necessarily evolves into an escalating war of purity. In this process, the most ruthless sociopaths, the self-appointed revolutionary vanguard, always end up as the rulers. Which is why almost every Marxist state converges on a violent totalitarian government.

If you are ever about to be to be seduced, at least know these anti-pillars, and spend a small amount of time understanding them.

12 rules

Apparently everyone does this. So here are mine. Like my lists of books,  if I did this ten times I would generate ten different lists.

1. Burgers: well done. E. Coli.

2. Don't buy the second item on the menu (see my previous post on this). Relatedly for wine: order domestic or same continent. Imports tend to be  over-priced

3. Insurance: only buy insurance for losses you can't afford, or if you have asymmetric information

4. Debit cards: don't use them. Cash or credit are strictly better options.

5. Shoes: one area where it is usually worthwhile to pay the premium. Prioritize quality (comfort, weight, stitching) over quantity.

6. Investments: only make investments without deadlines,   timing is often where hidden uncertainty is greatest, so make sure time contingencies are on your side.

7. Identity: keep your identity small. (h/t Paul Graham)

8. Food:  the food chain is acyclic. don't eat animals that can eat you

9. Butter is always better than margarine. Never use artificial sweeteners.

10. Premature optimization is the root of (almost) all evil in software engineering (h/t Donald Knuth)

11. Never rewrite from scratch (h/t Jamie Zawinski)

12. My phone  does not notify me. I notify it.

Desire and scapegoating

A couple of years ago, I stumbled upon the thoughts of René Girard. It's a pretty rare occasion when something makes you really think about the most basic things in a new way. That tweet led me to reading a bit more about it (thanks Dan!).

People are driven by memetic desire. Beyond our objective needs, what drive us most is wanting what others want. To put it in game theory terms, my utility function is a function of the utility function of others. This most obviously explains things like fashion for example. But also, more deeply, the notion of status in society.

I want something because others want it. This is a self-reinforcing mechanism, and the object of desire can become scarce, so it creates occasional  instability, frenzies of desire, and ultimately violence. This is a fundamental process in all human societies.
 

The way societies deal with it is through scapegoating. As the frustrated desires get stronger and stronger and become unsustainable, societies create scapegoats: some individual or group which is blamed for the inability of the many to satisfy the memetic desire. The frustration of desire reaches a paroxysm of violence on the scapegoat. Scapegoating works as an auto-immune mechanism: by channeling the violence onto the scapegoat, and even institutionalizing it, society avoids self destruction. And societies that don't scapegoat fall apart into chaotic violence. Thus scapegoating is an evolutionary adaptation to memetic desire.

This sheds some light on the institutionalization of violence throughout history, whether it is Mayan human sacrifice or the politics of immigration.

Optimism

A few months ago, I stumbled across a business card left behind on a table. Under the name of the company, it had three words, each followed by a period,  representing, I guess, the three pillars of their "corporate values".  One word struck me: Optimism.   I must have snickered,  because someone asked what was up. I instinctively thought "optimism" was a silly value, but it dawned on me that I had never thought about it explicitly.

Is optimism good?

The question sounds strange because Optimism, today, in American culture, is automatically assumed to be A Good Thing. Like "pro-active". People use that word as if it's synonymous with "good".  E.g. Person A: "Don't do this bad thing." Person B: "It's not bad, it's pro-active!"  Noooo....  Just like sometimes, being pro-active is evil, being optimistic is not automatically good.

Let's define optimism as follows: Having high expectations for a positive outcome. That is to say, compared to most "normal" people's probability distribution of outcomes, yours has more weight on the positive side. Say we both bet on the same horse, and one of us thinks we'll lose and one thinks we'll win.  So when is it good to be the optimist?  I would slice it on three levels:

1. Of course if you turn out to be right, then great... But that just means you got lucky.
2. What if you had to make the same choice over and over again, and on average the optimistic view is more accurate? Great, but that's not really optimism, it's having a better probabilistic model, better foresight.
3. Now what if you believe the same probabilities as everyone else, but you are more willing to take the risky choices and eventually you're better off? You are good at taking the right amount of risk for reward, and if in the long run you are better off (technically i.e. if you are on the efficiency frontier in the risk, reward plane),  then ... well that's good judgement. 

But in all these forms, the optimism is situational! The are plenty of situations where the wise person would take the "pessimistic" position.   Thus, as a fundamental value to live by, "optimism" is actually orthogonal to the things we consider good, truthful etc.

People (including Corporations!) of the world, listen to me: Value luck, foresight or judgement.  Not optimism.  That's just silly.

 

Known unknowns and unknown unknowns

A lot of people made fun of poor Donald Rumsfeld for his infamous quote..

"Reports that say that something hasn't happened are always interesting to me, because as we know, there are known knowns; there are things we know we know. We also know there are known unknowns; that is to say we know there are some things we do not know. But there are also unknown unknowns -- the ones we don't know we don't know."

But actually, he had a very good point. There are things that have uncertainty but we know the shape of the uncertainty. For example, I don't know who will the lottery tomorrow, but we do know the probability of any one person winning, or the distribution of winning amounts. And by the way from that we do know that playing the lottery is one of the dumbest activities known to man, though it is a kind of stupidity that we can harness for good perhaps... but I digress. My point is the outcome of the lottery is a known unknown. Which is different type of ignorance than say, not knowing if God exists. There you don't even have a probability space to support a distribution. Once someone asked to me: What's the probability that God exists? Obviously it was a rhetorical question, and it assumed the answer is "very low" (the question came from an atheist). But then I was like: it could be 0.01% or 99.99% or anything. If you have to choose, it might as well be 42.