2007/04/01

Puzzles vs. Mysteries

While I was on vacation in Roatan, I read a fascinating article by Malcolm Gladwell, the author of the books Tipping Point and Blink. Both are very interesting and worth reading, although you may not agree with everything Mr. Gladwell states, which is fine. They should both make you think, which is better than blindly agreeing with him anyway, and I think he would agree.

Anyway, in this New Yorker article, Mr. Gladwell talks about the difference between a puzzle and a mystery. He seems to base much of this on information from Gregory Treverton, who is or was a national security expert of some sort. I googled him to actually add some facts to this blog, which i don't normally bother to do, and he has a number of books and papers you can read.

Anyway, I won't try to explain the difference in detail because I think it is done well in the article. To paraphrase, a puzzle is something that has a solution, and if you have enough information, you can solve it. An example from the article is Bin Laden's whereabouts. We don't know where he is, but he is obviously someplace and if we had the correct information we could find him.

Mystery's are more complicated. No matter how much information we have, we don't know whether Iran will kill the British sailors or what would happen in Iraq after we overthrew Saddam. The article itself argues that Enron, which seems to have created an intentionally convoluted accounting system to hide it's lack of revenue/cash flow, was a mystery although you think it's a puzzle. The information was hidden and convoluted, but it was there if you asked the proper questions. Then Mr. Gladwell gives examples of people who did, including a college study group and a reporter for Fortune, I think. Anyway, the most telling fact from the article, and the one that proves to me people who value companies and recommend stocks are not as qualified as we like to believe, is this: Enron didn't pay taxes for 4 of it's last 5 years because it didn't have any money. Their annual reports used an accounting practice of booking future revenue (which was I believe based on a best guess of the future cost of energy), but they didn't receive any money. The IRS only cares about actual money. Enron didn't have any for 4 out of 5 years! They didn't pay any taxes! They still recieved awards and were modelled as a leading company of brilliant thinkers. It wasn't a mystery though, it was a puzzle. The piece of information someone needed from Enron to realize the company was in trouble was the tax statement. That is explained much better in the article, and I highly recommend you take the time to read it. REgardless, their accounting firm should be out of business and their CFO/President/CEO should all be in jail because of their ineptitude or lies, or both. The documentary "The Smartest Guys in the Room" is an interesting overview of the entire fiasco.

Anyway, I just wanted to throw out my opinion on the mystery concept as it relates to current events. The problem becomes how do you solve a mystery, how do you know you are asking the right questions, asking them of the right people, and getting honest answers? The Iraqi was seems like a good example. Let's forget any ineptitude or lies or both from our leaders and assume, as did happen, we were going to invade Iraq (this way we can ignore all the infighting, the outing of CIA agents, bad intelligence, etc, and focus on the post battle problems which is really what we need to resolve now). The mysteries around this seem to be as follows:
  1. If we pull out immediately, what will happen? Nobody knows for sure.
  2. If we make a timetable and pull out in 18 months, what will happen? nobody knows.
  3. If we don't make a timetable, what will happen? Nobody knows.

My point is, it seems with mysteries you have to deal with a lot of unknowns. How do we deal with unknowns, how do we resolve these issues? Well we all deal with unknowns in our lives. We deal with unknowns around our jobs, our personal lives, our families. What do we do? We try to focus as best we can and make a decision. Some of us are more detailed with this (job = money, I want to save to travel therefore this job is best now), others just wing it (I want to travel, so I'll quit my job and worry about retiring later).

I want a little more detail, clarity, and forethought on the mystery of Iraq from our government than I am seeing, and I am including both democrats and republicans. Ultimately, it seems to me we are dealing with tradeoffs and percentages, but we never really discus it this way. The political rhetoric involves making Iraqis safer, or brining home our troops, or saving money, or bringing democracy to the region (except in Palenstine, we apparently don't support their elected leaders or agree they are even a nation - that's another blog). Anyway, there are a lot of variables. Let's take a step back and start with 1 assumption: given the state of the world right now, we are bettter off with a stable Iraq that produces oil and isnt' a trouble making state. That's a bit vague, but I think you get the concept. What are the mysteries:

  1. Will Iraq be more likely to be stable in 2 years if we leave our troops there?
  2. If the answer above is yes, how much more likely is it to be stable in 2 years than now, or in 1 year? We then need to balance this percent by the cost of the war in both dollars and bodies. Remember the cost in bodies is not only deaths but ongoing medical treatment for the rest of the soldiers lives. I am not in favor of the war, but i am strongly in favor of supporting them when they return.

As I have argued before, any argument must begin with the same assumption. I believe we need to ignore why we are in the war when discussing what to do. Sure, people should be tried for treason but that's another story. So if you agree we want a stable Iraq, we need to decide how likely this will be in 2 years. Also how likely is it that this is possible. And does stable require us to have a democracy? That's anothe subtelty. So this analysis gets more and more complicated, but it can be broken down and analyzed. Let me repeat that: it can be broken down and analyzed. Anything can. If you can agree to our current state and our desired state, then define the potential scenarios are at each step, the only debate becomes the probability of each scenario. This would be political but informative, and also allow us to make a reasoned decision. I don't believe this is happening. The rhetoric is just that, feeling based speeches to get votes. It's not logical, and I believe we moved into this war based on feelings. We should not get out of it for the same reasons.

This type of analysis was part of my college education in operations research. I won't bore you with details because I still don't truly understand it, but the mathematical part of the class focused on giant probability matrices. One class was called "Stochastic Equations: The study of probabilistic systems as time goes to infinite." I got a D- in it, and believe me, it was a pity D- because I never got above 20% on any homework or test. Fortunately, my glorious alma mater doesn't give pluses or minus so it looks like a solid D on my transcript - rock on MIT! In another class we reviewed studies that showed how inconsistent we are when we make decisions based on feelings. I'll try to find examples of this, but you can identify these by making a list of priorities. For example, a person might prioritize saving for retirement, saving for education, then money for going out. However, in a given month, they will use money they set aside for retirement in order to go to a 5 star restaraunts. It's not logical. People are not logical. Commercials attempt to manipulate us, politicians manipulate us, religion, friends, family. Sometimes, although it's heartless, our best decisions should be analytical. It may be a mystery, but by breaking it down as much as possible, we can at least make the best decision and, hopefully, be prepared for all the possible alternate scenarios. This didn't happen in Iraq (obviously). Let's not let it happen again.

No comments: