Hacker News new | past | comments | ask | show | jobs | submit login
Thomas Schelling has died (washingtonpost.com)
151 points by nabla9 on Dec 14, 2016 | hide | past | favorite | 48 comments

The beginning of PART II in his book "The Strategy of Conflict" is good intro to what he was thinking.



On the strategy of pure conflict - the zero-sum games - game theory has yielded important insight and advice. But on the strategy of action where conflict is mixed with mutual dependence - the nonzero-sum games involved in wars and threats of war, strikes, negotiations, criminal deterrence, class war, race war, price war, and blackmail; maneuvering in a bureaucracy or in a traffic jam; and the coercion of one's own children traditional game theory has not yielded comparable insight or advice. These are the "games" in which, though the element of conflict provides the dramatic interest, mutual dependence is part of the logical structure and demands some kind of collaboration or mutual accommodation - tacit, if not explicit - even if only in the avoidance of mutual disaster. These are also games in which, though secrecy may play a strategic role, there is some essential need for the signaling of intentions and the meeting of minds. Finally, they are games in which what one player can do to avert mutual damage affects what another player will do to avert it, so that it is not always an advantage to possess initiative, knowledge, or freedom of choice.

I can't remember who recommended "The Strategy of Conflict" to me, but it's an incredible read -- for people with or without a background in game theory or international relations. It shows how important commitment is, how powerful it can be to have your hands tied. Very highly recommended.

I hope fans of Schelling will meet up to celebrate his passing together (at a time and place not agreed upon beforehand.)

Is that typical writing in the book? That is incredibly clear while still being reasonably dense.

That is very typical of Schelling. Clear, but you need to read it over and over.

The Schelling model for racial segregation is a great little agent-based model demonstrating one complex systems idea. There's a nice browser version here: http://nifty.stanford.edu/2014/mccown-schelling-model-segreg...

Vi Hart & Nicky Case wrote a web page demonstrating the same effect: http://ncase.me/polygons/

Neat. I think I will show this to my more SJW-ish friends, simply to demonstrate that there is more than one model that could account for population self-segregation other than someone being hateful or racist.

I'm not sure it's as relevant as you seem to think. For one, the distribution of the population was never initially "random". But second, the idea of boiling it down to "how satisfied are they" being the same as "there's enough people like them around them" is pretty close to what people are saying is happening anyway: sure there is "self-segregation" happening because people want to be close to their families, but "white flight" is a thing too.

Racism comes in a huge variety of different hues, but when one population is segregated in the nice neighborhoods and another is segregated around the not-so-nice, I find it difficult to accept that self-satisfaction with the number of similar neighbors is much of a driving factor of anything.

Don't misunderstand- I am certainly not claiming that racism isn't an issue!

My only goal is to try and get them to see the world as something more complex than the 'agree with everything we say or you're literally Hitler' dichotomy that currently appears to be popular in elements of the modern left.

(Perhaps I find this particularly frustrating, as I am left-leaning myself.)

The concept that reasonable people can disagree is something that I think everyone, regardless of political persuasion, should keep in mind. And part of learning that is seeing how multiple models, even simple toy ones like this example, can explain large-scale behaviors in ways other than one's currently-held belief, whatever that may be.

That's not the lesson I got there - the key points are

1. Small personal biases lead to large collective biases

2. Unless people are unhappy and actively willing to move when segregated prior; segregation is the steady state. i.e. White flight exists, but the inverse does not.

I don't disagree. I think your first point is particularly salient. What I want them to understand is shades of degree.

The question then is "do the different reasons for population segregation lead to different results?"

A community of people sharing a common language and heritage may just lead to pleasant dinner parties for them, and an eventual cultural blending for the kids and grand-kids.

A community segregated around a common feeling of oppression or discrimination may end up fomenting hatred and violence in response.

The segregation isn't the problem, the reasons are.

Yes! And I want them to understand that. Many of them aren't there yet.

Don’t be a jerk about it though - just because there may be non-racist reasons for segregation to have occurred doesn’t mean that there weren’t outright racist reasons too: sundown towns were a thing in the supposedly “enlightened” north-east for a start & undoing the effects of those racist policies doesn’t happen overnight.

I agree with what you're saying.

My biggest desire is simply to get them to understand that reasonable people can just plain disagree on something, and that that doesn't necessarily make either one 'bad', because the real world is more complex than any one mental model.

The existence of segregation is repeatedly cited as PROOF of racist oppression and anyone who says otherwise denounced as racist. The fact that segregation can easily emerge even with no oppression at all means that segregation alone is not sufficient to prove that any injustice has been committed at all much less that injustice/oppression/racism is the primary cause of the segregation.

Cited as evidence perhaps. I would like to see these repeated assertions of proof.

Have a conversation with left-leaning, progressive college student.

I wouldn't even have ever heard the term SJW, or be aware of how modern progressivism is so different from the liberal attitudes of my youth, if I didn't have younger family still in college.

However, there is a difference between explicit segregation (black people can't use this bathroom or school, redlining) and the type that arises naturally from interactions between people.

Housing segregation is extremely complex and multifaceted, I'm not sure anyone ever claimed otherwise.

Most of the causes are probably due to ripple effects of historic policies.

I had the privilege of lunching with Tom semi-regularly in the academic years 1979-81. He was remarkably gracious and at least seemingly humble, and just an all-around nice guy. (Others at our table were Dick Zeckhauser and Chris DeMuth.)

Tom wasn't exactly my inspiration to "Think deeply about simple things"; I got that mantra from Arnold Ross. But he was perhaps the best I ever knew at putting that principle into practice.

One of the great ones has left us.


For those who are interested, Schelling's work was also foundational to David Friedman's (son of Milton) work on libertarian anarchist theory.

From the conclusion of http://www.daviddfriedman.com/Academic/Property/Property.htm...

"The central project of this essay has been to give an account of rights, especially property rights, that is both amoral and alegal—an account that would explain the sort of behavior we associate with rights even in a world lacking law, law enforcement, and feelings of moral obligation.[28] I have tried first to explain how, with no legal system to enforce contracts, it might still be possible to contract out of a Hobbesian state of nature, and then to show how the same analysis can be used to understand in what sense a civil order, such as our own society, is different from a Hobbesian state of nature. Having offered answers to those questions, I then tried to show how we might get from the state of nature to something like the present society, and to use the analysis to partially explain the puzzling similarity between actual rules, just rules, and efficient rules.

If my analysis is correct, civil order is an elaborate Schelling point, maintained by the same forces that maintain simpler Schelling points in a state of nature. Property ownership[29] is alterable by contract because Schelling points are altered by the making of contracts. Legal rules are in large part a superstructure erected upon an underlying structure of self-enforcing rights."

When I asked my dad - https://en.m.wikipedia.org/wiki/Des_Ball - what the most important books were to shaping his thoughts, The Strategy of Conflict was number one.

Not a good year for strategists.

Sorry you lost your father. That's really rough- I've been there (years ago). Find some peace for the holidays- it can get kind of rough if you don't prepare.

Thank you. Fortunately I have a close family.

Schelling's works are necessary reads. They are important, but every time I look at them I come away thinking of them more as codifications than novel concepts. He did put names to previously unnamed dynamics (game theory) but when I read history I cannot believe that people pre-Schelling didn't already know and work with these concepts daily. I'd say that he did great service in documenting and understanding the dynamics. His work allows everyone to up their game. But I read his work more as I would read the work of a psychiatrist 'discovering' a new disease. The disease has always been there and now the doc is documenting it in a way so that other docs may recognize and discuss it more easily.

And Black-Scholes just codified what seasoned options traders had understood for centuries. The reason it was important and rewarded is exactly that codification, which enables rapid understanding to a wider audience.

Discovering something new is exciting and important. Explaining that discovery in a way usable by the rest of humanity is ground-breaking.

Not sure what you have read.

But, try re-reading his article "Hockey Helmets, Daylight Savings, and other Binary Choices".

You will find something new about collective choices.

On global warming Schelling has an interesting take. He claims that GW has potentially disastrous consequences to developing nations but the danger to developed nations has been exaggerated. That developed nations, in reducing emissions, will bear most of the cost while the developing world will get most of the benefits.

That seems very short sighted in a couple of ways. First the notion that people suffering from the effects of ecology destruction and the aftermath that ensues just hang around and suffer and keep the suffering contained. Secondly the notion that there's nothing to be gained by combating global warming for the entire world. More people die from the effects of pollution than the combined deaths from homicide, war, car accidents and suicide per year.

The fact that "developed nations, in reducing emissions, will bear most of the cost while the developing world will get most of the benefits" does not mean it's not a good idea to reduce emissions, it means that an agreement between the two is unlikely to happen. It is an empirical claim, not a normative claim.

Along those lines, this video has a good interview at 33-36 minutes in: https://youtu.be/8v1kU7Rc9c4?t=33m35s ("Before the Flood")

India has 300 million people without electricity, and they're doing exactly what the US and other nations did years ago: focus on improving their quality of life, using inexpensive coal and fuels.

> More people die from the effects of pollution than the combined deaths from homicide, war, car accidents and suicide per year.

Pollution is not the same thing as climate change.

Imagine a Venn diagram with Pollution one circle and Climate Change in another and the intersection is Reducing Emissions. The title being "Intersection affects both".

Yes, I understand that reducing pollution by, for example, reducing emissions from coal-fired power plants will also affect climate change (assuming you believe reducing CO2 emissions will do that). That doesn't change the fact that pollution causing people to get sick or die is not the same as climate change causing people to get sick or die. If we want to fight pollution because it kills people, and help with climate change as a side effect, that's one thing; but the post I responded to was basically arguing that we should fight climate change and hope that it reduces pollution as a side effect, based on the fact that pollution kills people. That argument makes no sense.

Orly? No sense? The more issues that can be shown to affect people directly that can be traced back to emissions the better.

> The more issues that can be shown to affect people directly that can be traced back to emissions the better.

I would have thought that "pollution kills people" was already reason enough.

Or, to look at it from the other side, if a country like China isn't going to reduce its emissions in order to stop killing people right now, why would they bother to do it to reduce some hypothetical effect on climate change over the next century? And if the answer is "well, they say they will", why would we believe them?

Remember that economics is rather bad at pricing things like huge population migrations. "Bearing the cost" does not include an a huge array of expected outcomes that humans will find negative but are difficult to plug into a toy economic model.


Reminds me of the Coursera course on model thinking: https://www.coursera.org/learn/model-thinking

The segregation model was one of the models (which I though eas pretty neat), the course is basically an introduction to a bunch of different models to think about the world. Highly recommended.

Original title is better: Thomas Schelling has died. His ideas shaped the Cold War and the world.

Usually I can see why titles are edited. This one makes no sense to me. After reading the article I can certainly see why he was important, but a household name he ain't.

I wonder if the theory "Nash equilibrium" was influenced by him, or even possibly vice-versa.

No, Schelling's contributions to game theory were not at a level of formalism.

His uses of game theory was however at the highest level of strategic thinking.

Without looking it up, I don't think the timing was right. He once said his work during WW2 was one of the highlights o his life, but while I forget the details, I think it was conventional economies rather than the strategy stuff for which he became famous.

Wikipedia says the von Neumann version of the NashEq was published in 1944 when Schelling was just graduating. And also that Cournot published a version in 1838...

Technically, it's true that:

-- The Von Neumann Min-Max Theorem is a special case of the existence of Nash Equilibria. -- Proving the original Min-Max Theorem isn't any any easier to do directly than is proving the existence of Nash Equilibria in general.

That said, my dissertation was an extension of the Min-Max Theorem, and I basically didn't think about Nash Equilibria once the whole time I was working on it.

THAT said, my dissertation was an independent proof of the theorem of Mertens and Neymann. Their proof seemed closer to the Fixed-Point Theorem kinds of techniques that Nash used.

Anyhow, I'm pretty sure that Tom himself didn't engage in that kind of mathematics.

He was such an intellectual powerhouse.

Unfortunately the Kindle version of The Strategy of Conflict is unavailable.

But you can buy it on iBooks. Just did.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact