Monday, March 9, 2009

Books I've been reading

Wired for War, By P.W. Singer. This is a book about the current and future of robotics warfare. He talks about our military's current systems such as the Predator Drone that can do both surveillance and shoot missiles at targets, all being controlled by remote-controlled human pilots in Nevada who get very, very bored by their duty. He looks in to the future to robotic ambulances that could come rescue wounded soldiers, load them in to a vehicle and carry them back to a hospital.

He looks further into the future towards a paradigm shift he calls the Singularity, a creation of science-fiction writer Vernor Vinge. It's a moment when all the rules of technology change because the artificial intelligence will become so advanced that it, rather than we, will be creating new technologies in a new constant feedback loop.

As a teenager, I read an assortment of novels and short stories by Issac Asimov about robots, all of which were guided by the Three Laws of Robotics:

  1. A robot may not injure a human being, or, through inaction, allow a human being to come to harm.
  2. A robot must obey orders given to it by human beings except where such orders would conflict with the First Law.
  3. A robot must protect its own existence, as long as such protection does not conflict with the First or Second Law.
These laws were hardwired into Asimov's robots, and it was impossible for a robot to do anything outside of them. I liked these stories, and I thought that this would help create a much better future. Alas, several practical considerations get in the way of implementing these laws. One is that technology doesn't yet allow the kind of thinking in a machine that would make such decisions possible. As one programmer put it, "They are in English. How the heck do you program that?" Another is that the military provides most of the funding. They want robots that can kill. Sigh.

Singer is a nerd who thinks robots are really, really cool and that comes through in the text, which sometimes makes a poor juxtaposition with the seriousness of the topic he's covering. However, he does cover a lot of really scary stuff, such as this difficult-to-argue-with point:

Marvin Minsky, who cofounded MIT's artificial intelligence lab, believes that we humans are so bad at writing computer software that it is all but inevitable that the first true AI we create will be "leapingly, screamingly insane."

Another was a quote from Senator John Warner, chairman of the Armed Services Committee, in which he referred to the high casualty rates of past wars:

"...in my judgment, this country will never again permit the armed forces to be engaged in conflicts which inflict the level of casualties we have seen historically. So what do you do? You move toward the unmanned type of military vehicle to carry out missions which are high risk in nature... the driving force is the culture in our country today, which says, 'Hey! If our soldiers want to go to war, so be it. But don't let any of them get hurt.'"

He mandated that by 2010, one-third of all the aircraft designed to attack behind enemy lines be unmanned, and by 2015, one-third of ground combat vehicles be driverless. And, the Congress passed it in to law.

What's most disturbing about Warner's quote is that the electorate of the United States seems to only care about saving American lives. War is fine so long as Americans don't get killed in it, something of an immature attitude in American foreign policy that the robotics revolution is only going to exacerbate. I remember one of Asimov's novels included a robot that actually did kill humans, but only because its programmers were successful at narrowing the definition of human to the residents of a particular planet. That story seemed to be more of an allegory for us than robots, as our country's guarantees of life and liberty are perfectly dispensable when dealing with "bad people" whom we say are associated with terrorism.

Pretty much every advance related to war, whether it be horses, gunpowder, flight or nuclear fission, has increased deaths. I have a difficult time seeing military robots helping matters much.

Predictably Irrational, by Dan Ariely. This is a book about decision-making, focusing especially on times when we allow irrelavent information to cloud our judgment. I enjoyed this book very much, and it wasn't nearly so scary as the death-robot book.

Ariely is a professor of behavioral economics at MIT, and makes his living performing an assortment of experiments to test people's decision-making. One experiment he did was a taste test of beers -- one was a brand-name beer poured without modification, the other was the same stuff but with a teaspoon of balsamic vinegar in it. With blind taste-testing, the balsamic beer usually won, and then the experimenters told the tasters what was in the beer. When the experimenters told the tasters in advance what they were getting, the plain brand-name beer won.

He also explored the issue of "anchoring," when you make choices relative to the choices in front of you as opposed to relating it to what you can afford or what you want. An example: A real estate agent shows you a nice colonial house, a colonial house that needs work, and a contemporary house. Chances are you won't choose the contemporary house because you don't have anything to relate it to. But, you will like the nice colonial house because it's better than the one that needs work, and the fact it compares better to something else makes it more attractive. The real estate agent is trying to steer you towards the nice colonial, and the fixer-upper is just a decoy.

The most interesting part of this book is about the difference in rates of cheating when your reward is cash and when it is a non-cash item. He starts off the chapter with the following conundrum. Your spouse calls you at work and tells you that your daughter needs a red pencil for a school project. You could...

a. Go to the office supply cabinet and take a company red pencil.
b. Take a dime from company petty cash and go to the drug store to buy a red pencil.

If these are your only two choices, almost everybody would pick a, take the pencil. But why? The company is losing just as much with the pencil as with the dime. The experiments he goes on to describe how cash makes us more honest.

The series of experiments his group peformed involved having students take a test with 20 simple math problems. Test-takers got 50 cents for every right answer. The first group had no opportunity to cheat, making them the control group.

They then gave the test to several different groups, each group getting more opportunity to cheat. The first group got to see an answer sheet before they turned it in, thus giving them the opportunity to change answers, the next got to self-report to the number of right answers to the proctor after seeing the answer sheet, and another got to reach in to a jar of dimes in an unsupervised room and take out the amount of money corresponding to the number of answers they got right. A little bit of cheating took place in all of these cases, on average, test takers exaggerated their number of right answers by 2.7 answers. An interesting thing with these groups was that as the ease of cheating increased, the amount of cheating did not. The exaggeration rate stayed right around 2.7 answers.

The reward for the next group of test takers was not money but a token that could then be redeemed for money on the other side of the room. Then the exaggeration doubled, with the takers claiming to have gotten 5.9 more answers right. These tokens were obviously cash equivalents, but they weren't cash. The conclusion that Ariely draws is that when you separate stealing from cash, stealing becomes much easier.

A statistic he uses to make this point is very telling: Employee theft and fraud at the workplace result in about $600 billion in losses each year in the United States. This figure is much higher than the combined financial cost of robbery, burglary, larceny-theft and automobile theft ($16 billion).

He suggests that the directors of Enron, Kenneth Lay and Jeffrey Skilling, would have been unlikely to mug an old lady for $50. But, when it comes to moving around these large, abstract amounts, or miscategorizing them on financial statements, the association with cash and the association with the people who own it becomes so blurred that it becomes easy to rob your company, your employees and shareholders.

The Secular City, by Harvey Cox. This is a library book I have come to own against my will. I read it with great enjoyment, wrote notes about it in my journal and then put it in my bag and took a trip on a Greyhound bus. The bag was put in the undercarriage of the bus and it rained all day. The luggage compartment sprung a leak, soaking the bag and damaging the book to the point where the library made me buy it. Now I am the proud owner of this wrinkly thing. The librarian warned me not to put it on my bookshelf for at least a year for fear that mold might develop and jump to the neighboring books.

Unlike the other two books, which had been published in the past year, The Secular City was published in 1965, and its main purpose is to define the process of secularization as it relates to urbanism. It was a very instructive, interesting book, better than I thought it would be. What I especially liked was how Cox defined the limits of this movement as it is kind of amorphous and easy to associate with anything that isn't affiliated with an "official" ideology.

Secularization, he says, is the process by which people in a city focuses on pragmatism and profanity. Pragmatism is the question "Will it work?" and profanity literally means "outside the temple" or related to "this world." This becomes a helpful way of thinking for how to have a city that one would like to live in, rather than maintaining the power of an official religious group, regardless of its popularity.

It doesn't necessarily have to be come an anti-religious force, however, and Christians make a lot of mistakes when they try to oppose secularization. One example he gave was of churches that tried to organize prayer groups inside large apartment buildings, trying to spiritually connect the neighbors to one another. This effort mostly failed because it was really an attempt to take away the anonymity that the city offered and turn the apartment building into a village, stuck in the middle of a city. The apartment dwellers who wanted to go to church liked the number of choices that the city provided them, and didn't want to get stuck in a narrow social group. The ones who didn't want religion just didn't want religion.

Cox makes it clear that there are many anti-religious forces out there that aren't secular because they attempt to impose their own new ideology. "The rejection of meanings which do not contribute to one's own group is the opposite of secularization." When the pragmatism of modern people is allowed to degenerate into a system that focuses solely on function, what you have is a new religion, just one without gods in the sky. Nazism, he said, was a kind of tribalism in which only those who were "useful" to the new Reich were allowed to survive.

He gives a couple of other examples of "usefulness" going too far -- the business careerist focusing only on success, never on beauty or enjoyment, or the beatnik's resentment of everything established.

Christians and political conservatives alike have spent the past 60 to 70 years wringing their hands over what to do about secularization, and it's usually presented as a big scary force that's responsible for the general decline in church attendance and loosening morals. Every week I come across some essay about the dangers of secularism or of post-modernism. It's easy to write about it as a hostile, anti-God, anti-beauty, anti-life force because it's more fun to get in fights than come up with solutions. In those rare cases when secularism turns in on itself to become a new, aggressive ideology that seeks to impose itself on society, we should oppose it. (One example I can think of is an opinion I've heard that says contraception ought to be required for poor people because their kids cost too much to the taxpayers.)

But really, it's an opportunity for us to be really honest about our faith and live like real people who aren't trying to satisfy religious standards that artificially got placed on us. I'm trying to write a book about this, presently on version number 3 on how traditional religion can do a better job of speaking to those with a pragmatic, functional set of values. It ain't as easy as I thought it would be.

1 comment:

Fr David said...

Kind of along the lines of cultural change - and related in a way to the first book since it is the impact of technology on society is The Transparent Society by David Brin. He's an SF writer and this book is actually a non-fiction work about a theme that he took up in one of his fiction books.