futurehead

The mathematics of fascism?

German historians have long debated the origins of the Nazi disaster.  Part of that debate has revolved around what is called the sonderweig, or “special path" -- the notion that there was something unique in German history going far back that put them on a different path from other nations.  It is a suspiciously comforting thought, because if the Nazi takeover was due to more immediate factors -- was more contingent than inevitable -- then that would suggest that every country, if the wrong things take a turn for the worse, could head to hell rather quickly.

There was also a debate a few years ago the moral culpability of the average German for the Nazis and the Holocaust (prompted by Daniel Goldhagen, Hitler’s Willing Executioners [1996]). I don’t really know enough about German history to have an independent opinion on the matter – but I lean against it.  First, I saw that a lot of historians I respect dismissed that book.  In addition, whatever the merits of that argument, it is not Useful.  If you conclude that the Germans were To Blame, what do you do with that?

I’ve always wondered whether the Nazi takeover of Germany – or any takeover that has at least an element of force – could be modeled mathematically – in other words, if there isn’t an element of game theory at work here.  Somebody oughta look into this, if it hasn’t already been done (probably has been).

This thought is based on an intuition that there is a mathematical dynamic behind a forceful takeover of a civilized nation.  This is based on the sense that a few people willing to kill might prevail upon a much larger, but atomized and passive, group.  It might run something like this:

Initial conditions:

N = Nazis, Hitler’s core supporters who were willing to kill those who oppose them
S = Nazi supporters, who support Nazis enough to inform on others
P = passives – non-supporter, non-opponents of Nazis
O = active opponents of the Nazis, such as the communists

N` = commitment of Nazis to kill opponents
O` = willingness of opponents to risk their lives to oppose Nazis

Now imagine a grid (like that used in the game “life” or in cellular automata) made up of cells, whose state is determined by the state of those around them (how many like individuals they are connected to, since power comes from connectedness & cooperation with others).  Imagine the opponents are red, the passives are white, and the Nazis are black.  Some reds and blacks are darker (more committed), while others are pale. 

Then apply a calculus involving how connected the reds and blacks are to others, and how committed they are, to determine which passives flip to Nazi supporters, and which opponents flip to passives – and which opponents are simply rubbed out.  It is easy to imagine how a tightly clustered, very committed batch of black squares (Nazis) in the middle of a sea of white squares (Passives) with sporatic clusters of Reds could end up flipping the whole field to supporters and passives (and dead people).

Similar-minded experiments have been done on the subject of residential racial segregation.  An economist named Thomas Schelling (a pioneer in looking at "tipping points" and other elements of nonlinear systems) puplished a famous paper in 1971 (“Dynamic Models of Segregation”) in which he found that if you have a population full of people each of whom is not a racist, but just doesn't want to live in a neighborhood that is strongly dominated by a race other than their own, that the individual choices made by that population would lead mathematically to an emergent pattern of almost total residential segregation. 

I don't know why but I just find myself worrying an awful lot in recent years about political stability and the ability of crazily committed and intense insane political minorities to "flip" much larger political systems. . .

Crime and the Law of Large Numbers

Take a poor kid who has committed a crime.  How do we evaluate that ethically as human beings?  Conservatives will say throw the book at 'em.  The liberal attitude (though since Clinton few liberals appear strong enough to ever admit it) is to look at the kid's background and be more forgiving.  I would argue that ethically, we must approach this kind of situation with two minds simultaneously: as a judge, and as a ruler (there's also as a priest, who looks at the sin, and seeks to understand and forgive it, but let's put that aside for now). 

• Judge.  As a judge we look at an individual's actions, and judge them according to our notions of right and wrong.  End of story.
• Ruler.  As a ruler (policymaker) interested in the public good (let's imagine), your goal is to achieve social stability and justice.  From that point of view, you have to recognize that environment leads to crime. 

In other words, faced with an individual who mugged somebody we must be judgmental and censorious (as well as appropriately merciful).  But facing a city full of muggers, we must recognize that, through the law of large numbers, there are environmental factors that CAUSE bad behavior. 

As social policy, we cannot but punish bad behavior in individuals, as if each has complete free will.  (After all, whatever happened to the guy in front of you, there are probably other people with the same history who have NOT become muggers.)  But in making policy that applies to groups, we must act as if each is nothing but a complete product of circumstance.

The law of large numbers says that the larger one's sample, the closer to chance expectation the distribution will be.  If you flip a coin four times and get three heads, or 75% of the time, it is unsurprising.  If, however, you flip a coin 10,000 times, and the rate remains 75%, you can be certain the coin is weighted. 

It's safe to assume that in any human society, a certain percentage of citizens will go bad -- become anti-social, criminals, evil.  Perhaps this is an inevitable product of life; perhaps the multitudinous stresses of daily living, getting along with other people, dealing with injustices large and small, reigning in the ego, absorbing the many small assaults on our self-worth with which each person is confronted, and the contours of each person's inborn personality -- perhaps these factors are, like nodes of sound waves in a concert hall, bound to concatenate in certain places, ensuring that a certain percentage of the population slips into behavior that the society cannot accept. 

Let us say that the estimated percentage of citizens over age 10 who will commit a crime in a given year in the United States is N.  Therefore, according to the law of large numbers, if N% of your town's residents are arrested this year, you can feel as though things are in reasonably good order.  However, if the rate is 3N% or 4N%, and your town or city has more than a few individuals, you must conclude that something systematic is wrong.  It is unlikely to be coincidence that so many bad people have happened to land in your town.  There must be some distinct CAUSE for this unfortunate conglomeration of malevolence.  The racist may ascribe it to the number of blacks or latinos in the city or whatever, and conclude nothing can be done.  But an enlightened person realizes that it must be a result of circumstances, and seeks to change the circumstances.  The fact that so many conservatives argue that nothing can or should be done except bear down with ever-harsher punishments, may be a reflection of a racism implicit in their worldview.  Or just an inability to see things with two minds. 

The AAEAE: an organization I'd like to see

In Washington the other day I walked by a building with a sign announcing that housed inside was the "Society of Association Executives."  I just love that.  If they had embraced their "meta" status as they should have instead of trying to smooth over it, they would have named themselves the Association of Association Executives, or alternately the Association Executive Association.

Which makes me wonder: are they the only association of their kind?  It could be true that there are a number of them.  And, if the executives of those associations one day decided they needed to organize, we might see - yes! - the Association of Association Executive Association Executives. 

If you have trouble parsing that, this might help:

Association of [Association Executive Association] Executives

What a beautiful thing that would be!

Why "Proper Form" is Elusive

As a triathlete, I’m always reading about how important it is to use proper form in running, swimming and biking.  And of course similar advice is routinely handed out in every other sport as well.  But all this advice needs to be taken with a big ol’ grain of salt (see previous post on same theme).  Here’s why:

In mathematics there’s something called the “traveling salesman problem,” which is really very simple.  Imagine 4 cities spread out on a map, various distances away from each other.  The question is, what route should you take so that you visit each of those cities, yet travel the least distance possible?   

When there are only four cities, it’s easy to solve.  But once you start to expand the number of cities, it quickly gets very difficult – with only 25 cities, there are so many possible routes that a computer measuring one million per second would still take 9.8 billion years to search through them all. (I get this from Peter Coveney and Roger Highfield, Frontiers of Complexity [1995], a thrilling and mind-blowing book that I highly recommend to anyone interested in such things).

Mathmaticians call this kind of problem an NP Problem – one that is not solveable in “polynomial time.”  There are a number of real world and mathematical problems that manifest this quality of seeming relatively simple to solve, yet which quickly spiral into unmanageable dimensions as you add more factors.  (Among them is the problem of factoring very large numbers, which is what modern encryption is based on and why it is currently unbreakable through brute force mathematical attacks.  It’s easy to multiply a bunch of prime numbers together and get the product, but it turns out it’s next to impossible to reverse the process.)

Anyway, I think about this whenever someone starts lecturing about “proper form.”  Imagine that you are engineer trying to figure out the most efficient way to move human limbs through water for swimming, or across the ground for running, or through space for kicking, batting, throwing, or whacking golf balls.  You have to take into account the particularities of the specific human body you are looking at:

• The size (and possibly strength, flexibility and density) of each bone that is involved in that movement (for many movements, that would be every major bone in the body)
• The strength, flexibility and elasticity of each muscle involved. 
• You may also need to factor in differences in muscle composition, such as the proportion of fast twitch and slow twitch muscle fibers
• Perhaps other more diffuse differences in metabolism, which could also affect what is the optimum way for a particular person to perform an activity.

I bet that finding the optimum form for a particular person would be an NP problem – a problem with so many variables that no computer could find the best form for you in your lifetime – let alone a single solution that fits everbody (every body).

The way we each tackle this problem seems to be through an approach of evolutionary learning – we flail around more or less randomly (as babies), and eventually through trial and error and the feedback of getting what we want (grabbing a toy when we’re young, getting the ball through the hoop, say, when we’re older), we hone our movements into more and more efficient forms. One way we do this is by trying something, and then adjusting it in different directions to see if our results improve. 

But another thing that Coveney and Highfield talk about is that you can picture the set of all possible solutions to an NP Problem as a topographical landscape.  As you try moving in different directions – adjusting things – you see if you move up hill (toward better solutions) rather than downhill (toward worse ones).  Then, you keep making adjustments until you have gone as high as you can.  The problem is that you might reach the top of a local peak, while there is a much higher peak – a more efficient solution for you – that lies somewhere else in the “solution space.”  And to take advantage of that you'd have to radically restructure and rebuild your stroke from the ground up.  (One person who made the leap and did exactly that is Tiger Woods - an amazing thing.)

So all that is why talk about “proper form” should be taken skeptically.  For your own biomechanics (and biochemistry), there may be one or many completely different forms that will work better than what you are now doing.  That coach who’s yelling at you to change the way you’re doing things is basing that on (best case) nothing more than an approximation of what has been found to be more or less effective by a lot of other people, or (worst case) the current fashion for how things should be done, perhaps based on what has worked for a few elite (and not necessarily typical) athletes.

Of course, there are some basic parameters of physics in every activity.  Also, if you’re a total newbie at something, you’re probably best off listening to a coach to provide some structure to your early flailing.  If you’re reasonably accomplished, the coach will probably just be helping you to try different adjustments in what you’re already doing to see if you can eke out some improvements.

The difficult case would come if you are accomplished at something, self-taught, or a “natural,” and some coach tells you that you’re doing it “all wrong” and tries to get you to completely restructure your form.  Then you have to decide whether to stick with your own body’s evolved solution, or leave that mountaintop and try to find another, higher peak through the textbook solution.  You just can't be sure the peak really is higher.

If you look at successful runners or other athletes or others like pianists, you’ll see a tremendous variety of forms.  There would probably be even more if various trends and fashions didn’t regularly sweep through instruction, standardizing the way people are trained.

Virginia's $3,500 Speeding Tickets

While I was on vacation this August, there was a controversy here in Virginia over a new program to impose Draconian (named after Draco, the first man to write a constitution in ancient Athens - I guess it was kind of harsh) fines of up to $3,500 for speeding tickets.  Some have argued that the fines are too severe, while others have pointed out that the tickets are only given to those who drive more than 20 MPH over the speed limit, which legally constitutes “reckless driving.” 

This law is crazy.  It’s the latest example of Republican willingness to do anything – anything – to avoid raising taxes.  So, in order that their rich constituents won’t have to pay more for the public good, or their ideological simpletons won’t have to re-think their stupid anti-government philosophy, they have instituted a grossly unfair system for raising money that the state needs. Here’s why they’re unfair:

• The writing of speeding tickets is random and capricious.  It’s one thing to impose high penalties on wrongdoing if people have some assurance that the punishment will be meted out relatively consistently, it’s another when that punishment is meted out so seldom and so inconsistently, as is the case with speeding, that it becomes effectively random. Why should a budget gap be filled by random losers in what amounts to a reverse lottery, rather than by everyone evenly through taxes?
• Speed limits have an inconsistent relationship to genuine safety.  It is entirely possible for a person to be driving along a road at a speed that is perfectly safe for the conditions – and yet be in violation of the speed limit, even by more than 20 MPH. Americans, especially in the suburbs, are fond of building overly massive, super-wide, straight, four-lane roads, which look and feel as if they're built like an autobahn -- and then slapping a 35 MPH speed limit on it.  Missed the sign that the limit just dropped 20 MPH?  That'll be 20% of the value of your car, please.
• It gives too much power to individual police officers.  Be REALLY nice to that cop who's pulled you over.  Certainly don't dare to intimate he did anything unfair or unprofessional.  Police already have wide discretion in how they treat civilians.  (Many are fair and professional. Some are not.) Now they will have the power to impose or refrain from imposing fines that for many citizens will be literally ruinous.
  
• The tough-punishment approach doesn’t work.  There’s nothing worse than some jerk driving recklessly fast or just recklessly, endangering the lives of children and everyone else on the roads.  I certainly want people like that punished when I see them.  But this punishment is too loosely related to genuinely bad behavior, and the "tough" punishments that conservatives always love so much never work -- especially if they're not scrupulously fair. What we need are police that use good judgement to catch people who are genuinely posing a threat to others, not capricious revenue-collection agents, making up the Republicans' budgetary gaps through this unjust anti-lottery. 

Several of these points can be illustrated by something that happened to me in 1998.  I was driving along Route 29 through central Virginia from Charlottesville to DC when I was pulled over at a spot in the road where I hadn't noticed that the speed limit suddenly dipped.  I had been driving at a speed that was perfectly safe for the long, straight, empty road I was on.  Later I mentioned to my graduate advisor that I’d gotten a speeding ticket, and he said “let me guess, you were clocked at 59 MPH.”  He was right!  How did he know that?  “They’re always for 59,” he explained, because if the cop had written me up for 60, it would have been wreckless driving, and I (and the officer) would have had to go to court, etc.  Instead, it seemed, Fauquier County preferred to just take my money & run. 

My take on presidential race

Okay my summer hiatus is over, I'm going to start posting again. 

My take on the presidential race so far

Looking over both fields, here are my current observations & predictions about the 2008 presidential race.  They’re a mix of my own thoughts with the observations of others I’ve talked to that I thought were on point:

• One of the frontrunners will win.  As much as the media likes a race, and always manufactures an underdog surge just before or after New Hampshire, the frontrunner always wins these things.  That will be doubly true this year with the compressed primary calendar.  The last guy to win who wasn’t a frontrunner was probably Jimmy Carter. So don't bother picking a pet candidate from the underdog bunch -- they're all running for fun, or for vice president, as someone put it recently.
• The GOP field is very weak. A mormon and two angels of death.  This group has the feel of the 1996 Bob Dole year: doom.  And it’s not just the catastrophic presidency of Bush the Lesser.  It’s just hard to find a president in the bunch.  McCain is lashed to the war, and he’s going down with the ship.  Giuliani is barely more presidential than Larry Flint, and he’s got nothing to offer but more fear and war.  Polls indicate that Americans are deeply suspicious of the mormon religion, and that’s gotta hurt Romney in the GOP primary more than anywhere.

Fred Thompson - I was waiting to see if he would enter, he does look like the only possible president in the bunch.  But, there are a couple of factors that could slow him down:

From his rhetoric in a speech the other day, he’s already making it plain that he too is solidly in camp with the 28% of moron dead-enders in this country who still stupport the Iraq War.

The rap I’m hearing on him is that he is among the emptiest of empty suits.

There’s also the X-factor of the fact that he has cancer – lympoma that was diagnosed 2 years ago but is now in remission.

Hilary Clinton – spare us. She’s hated as if she’s some hardcore liberal – and you don’t even get a liberal out of the bargain, but a mushy centrist.  She’s the candidate of those who think that to win votes you have to move right.  That’s a fundamental error.  I predict she’ll shatter in the early primaries, as Democratic voters reach for someone they think will win, as they did with John Kerry in Iowa.  However, unlike Kerry the early win by Obama or Edwards may not be enough to carry them through the other contests.

Barack Obama –

• First, the guy is good.  I wasn’t really giving the whole Obama craze much attention, until I heard him in an interview on NPR, in which he was asked some very tough questions, including about his racial identity, etc.  His answers were pitch-perfect – thoughtful, genuine, modest and appealing.  He seems like the most thoughtful of the Democrats – of all the candidates actually.
• Second, I’d like to see a little more evidence that he’s actually progressive.  I know he was a community organizer & all, but that doesn’t do it for me.
• Third, what the guy needs to show to prove he can win is some toughness and backbone.  Americans like their presidents to be tough, a touch ruthless even.  He also needs to show some manliness, Americans won’t elect a metrosexual.  Right now, Hilary Clinton’s more manly than he is.
• Of the top three, Obama’s the only one who hasn’t had his “seasoning.”  That’s what the early Virginia colonists used to call the first summer for new immigrants from England, who would die of disease in large proportions during their first summer in Virginia (and were said to be “seasoned” or immune if they survived it).  Clinton and Edwards have both been through the wringer of a presidential campaign and survived. There is a good chance Obama will go down in flames for one reason or another under the brutally intense spotlight that’s about to flip on.
• Edwards supporters are whispering that a black man is still unelectable in this country.  The other people I hear who are most cynical about his chances are – black people.  One guy I know is convinced he’ll be shot.  Others just think he can’t win.  This seems wrong to me.  A lot of the biggest racists are part of the 28% dead-ender set anyway, and other white people who are still racists in some ways will look at many factors in addition to race and could easily imagine voting for a black (call it the Morgan Freeman effect – people can just imagine a black man in the White House now).  That said, on issues of race it may well be that African-Americans’ understanding of the world is more accurate than mine.

• Edwards - I voted for him in the 2004 primary and at this point he’s still my fave.  He’s the only one talking about poverty and other economic issues.  He does have some strikes against him, however:  While charming there’s something about him that makes me just a touch uneasy, I can’t put my finger on it.  Maybe it’s just a California Boy’s unease with the whole Southern Charm thing.  He just doesn’t seem as sincere as Bill Clinton.  Also there’s his vote in favor of the Iraq war (though he has done exactly the right things since turning against it).   Finally, his wife’s cancer is a wildcard.

• Al Gore - Ain’t gonna happen.  Ever since he did the cautious, tie-yourself-in-knots thing in the 2000 election, it’s been so much fun watching what I call “the new WTF Al Gore.” But that's not the vibe of a guy with the bug.

If I had to call it today, I'd say it'll be Clinton vs. Giuliani with Clinton winning.  It will be both a relief to have a sane person in the Oval Office again, but also a grim disappointment that we'll have so little to look forward to under a Hilary Clinton presidency -- 8 more years of triangulation and caution and lack of progress in moving our country forward toward living up to its promise.