Query: by what objective criteria do we determine whether a political decision is rational?

I propose that the key elements -- necessary but not sufficient -- are (where "you" refers collectively to everyone involved in the decisionmaking process):

• you must use only documented reasoning processes:
  • use the best known process(es) for a given class of problem
  • state clearly which particular process(es) you use
  • document any new processes you use
• you must make every reasonable effort to verify that:
  • your inputs are reasonably accurate, and
  • there are no other reasoning processes which might be better suited to this class of problem, and
  • there are no significant flaws in in your application of the reasoning processes you are using, and
  • there are no significant inputs you are ignoring

If an argument satisfies all of these requirements, it is at least provisionally rational. If it fails any one of them, then it's not rational and needs to be corrected or discarded.

This is not a circular definition (defining "rationality" by referring to "reasonable" things, where "reasonable" depends on people being "rational"); it is more like a recursive algorithm, where large ambiguous problems are split up into smaller and smaller sub-problems until we get to a size where the ambiguity is negligible.

This is not one great moral principle; it is more like a self-modifying working process (subject to rational criticism and therefore improvable over time -- optimization by successive approximation). It is an attempt to apply the processes of science (or at least the same reasoning which arrived at those processes) to political discourse.

So... can we agree on this?

This is a hugely, vastly, mindbogglingly trimmed-down version of what I originally posted. All comments prior to 2010-08-26 20:52 (EDT) refer to that version, which I have reposted here for comparison purposes and for the morbidly curious. (It got voted down to negative 6. Twice.)

A word's denotation is our conscious definition of it.  You can think of this as the set of things in the world with membership in the category defined by that word; or as a set of rules defining such a set.  (Logicians call the former the category's extension into the world.)

A word's connotation can mean the emotional coloring of the word.  AI geeks may think of it as a set of pairs, of other concepts that get activated or inhibited by that word, and the changes to the odds of recalling each of those concepts.

When we think analytically about a word - for instance, when writing legislation - we use its denotation.  But when we are in values/judgement mode - for instance, when deciding what to legislate about, or when voting - we use its denotation less and its connotation more.

This denotative-connotative gap can cause people to behave less rationally when they become more rational.  People who think and act emotionally are at least consistent.  Train them to think analytically, and they will choose goals using connotation but pursue them using denotation.  That's like hiring a Russian speaker to manage your affairs because he's smarter than you, but you have to give him instructions via Google translate.  Not always a win.

Consider the word "human".  It has wonderful connotations, to humans.  Human nature, humane treatment, the human condition, what it means to be human.  Often the connotations are normative rather than descriptive; behaviors we call "inhumane" are done only by humans.  The denotation is bare by comparison:  Featherless biped.  Homo sapiens, as defined by 3 billion base pairs of DNA.

Cryonics scales very well. People who argue from the perspective that cryonics is costly are probably not aware of this fact. Even assuming you needed to come up with the lump sum all at once rather than steadily pay into life insurance, the fact is that most people would be able to afford it if most people wanted it. There are some basic physical reasons why this is the case.

So long as you keep the shape constant, for any given container the surface area is based on a square law while the volume is calculated as a cube law. For example with a simple cube shaped object, one side squared times 6 is the surface area; one side cubed is the volume. Spheres, domes, and cylinders are just more efficient variants on this theme. For any constant shape, if volume is multiplied by 1000, surface area only goes up by 100 times.

Surface area is where heat gains entry. Thus if you have a huge container holding cryogenic goods (humans in this case) it costs less per unit volume (human) than is the case with a smaller container that is equally well insulated. A way to understand why this works is to realize that you only have to insulate and cool the outside edge -- the inside does not collect any new heat. In short, by multiplying by a thousand patients, you can have a tenth of the thermal transfer to overcome per patient with no change in r-value.

But you aren't limited to using equal thickness of insulation. You can use thicker insulation, but get a much smaller proportional effect on total surface area when you use bigger container volumes. Imagine the difference between a marble sized freezer and a house-sized freezer. What happens when you add an extra foot of insulation to the surface of each? Surface area is impacted much as diameter is -- i.e. more significantly in the case of the smaller freezer than the larger one. The outer edge of the insulation is where it begins collecting heat. With a truly gigantic freezer, you could add an entire meter (or more) of insulation without it having a significant proportional impact on surface area, compared to how much surface area it already has. (This is one reason cheaper materials can be used to construct large tanks -- they can be applied in thicker layers.)

Another factor to take into account is that liquid nitrogen, the super-cheap coolant used by cryonics facilities around the world, is vastly cheaper (more than a factor of 10) when purchased in huge quantities of several tons. The scaling factors for storage tanks and high-capacity tanker trucks are a big part of the reason for this. CI has used bulk purchasing as a mechanism for getting their prices down to $100 per patient per year for their newer tanks. They are actually storing 3,000 gallons of the stuff and using it slowly over time, which implies there is a boiloff rate associated with the 3,000 gallon tank in addition to the tanks.

The conclusion I get from this is that there is a very strong self-interested case (as well as the altruistic case) to be made for the promotion of megascale cryonics towards the mainstream, as opposed to small independently run units for a few of us die-hard futurists. People who say they won't sign up for cost reasons may actually (if they are sincere) be reachable at a later date. To deal with such people's objections and make sure they remain reachable, it might be smart to get them to agree with some particular hypothetical price point at which they would feel it is justified. In large enough quantities, it is conceivable that indefinite storage costs would be as low as $50 per person, or 50 cents per year.

That is much cheaper than saving a life any other way. Of course there's still the risk that it might not work. However, given a sufficient chance of it working it could still be morally superior to other life saving strategies that cost more money. It also has inherent ecological advantages over other forms of life-saving in that it temporarily reduces the active population, giving the environment a chance to recover and green tech more time to take hold so that they can be supported sustainably and comfortably. And we might consider the advent of life-health extension in the future to be a reason to think  it a qualitatively better form of life-saving.

Note: This article only looks directly at cooling energy costs; construction and ongoing maintenance do not necessarily scale as dramatically. The same goes for stabilization (which I view as a separate though indispensable enterprise). Both of these do have obvious scaling factors however. Other issues to consider are defense and reliability. Given the large storage mass involved, preventing temperature fluctuations without being at the exact boiling temperature of LN2 is feasible; it could be both highly failsafe and use the ideal cryonics temperature of -135C rather than the -196C that LN2 boiloff as a temperature regulation mechanism requires. Feel free to raise further issues in the comments.

I've been asked to start a thread in order to continue a debate I started in the comments of an otherwise-unrelated post. I started to write a post on that topic, found myself introducing my work by way of explanation, and then realized that this was a sub-topic all its own which is of substantial relevance to at least one of the replies to my comments in that post -- and a much better topic for a first-ever post/thread .

So I'm going to write that introductory post first, and then start another thread specifically on the topic under debate.

A little while ago, I argued with a friend of mine over the efficiency of the Chinese government. I admitted he was clearly better informed on the subject than I. At one point, however, he claimed that the Chinese government executed fewer people than the US government. This statement is flat-out wrong; China executes ten times as many people as the US, if not far more. It's a blatant lie. I called him on it, and he copped to it. The outcome is besides the point. Why does it matter that he lied? In this case, it provides weak evidence that the basics of his claim were wrong, that he knew the point he was arguing was, at least on some level, incorrect.

The fact that a person is willing to lie indefensibly in order to support their side of an argument shows that they have put "winning" the argument at the top of their priorities. Furthermore, they've decided, based on the evidence they have available, that lying was a more effective way to advance their argument than telling the truth. While exceptions obviously exist, if you believe that lying to a reasonably intelligent audience is the best way of advancing your claim, this suggests that you know your claim is ill-founded, even if you don't admit this fact to yourself.

The Lisbon treaty was finally ratified last Tuesday, in a most wonderfully disdainful signing cermony.

I take it that everyone on the list is emotionally overwhelmed by this, one of the most important political events in recent history. The world's largest economy has taken a firm step towards statehood; the ramifications of this will be felt across the world. People will die who would have lived; people will live who would have died: the body count is much affected. The potential implications for AI alone (think political singleton, research funding priorities) are huge. Depending on your opinion of the consequences, you are probably dumped into a dark ditch of despair or swimming in a limitless ocean of triumphant glee.

If neither is the case... why not?

I stumbled upon an article called The New Nostradamus, reporting of a game-theoretic model which predicts political outcomes with startling effectiveness. The results are very impressive. However, the site hosting the article is unfamiliar to me, so I'm not certain of the article's verity, but a quick Google seems to support the claims, at least on a superficial skimming. Here's his TED talk. The model seems almost too good to be true, though. Anybody know more?

Some choice bits from the article:

The claim:

In fact, the professor says that a computer model he built and has perfected over the last 25 years can predict the outcome of virtually any international conflict, provided the basic input is accurate. What’s more, his predictions are alarmingly specific. His fans include at least one current presidential hopeful, a gaggle of Fortune 500 companies, the CIA, and the Department of Defense.

The results:

The criticism rankles him, because, to his mind, the proof is right there on the page. “I’ve published a lot of forecasting papers over the years,” he says. “Papers that are about things that had not yet happened when the paper was published but would happen within some reasonable amount of time. There’s a track record that I can point to.” And indeed there is. Bueno de Mesquita has made a slew of uncannily accurate predictions—more than 2,000, on subjects ranging from the terrorist threat to America to the peace process in Northern Ireland—that would seem to prove him right.

[...]

To verify the accuracy of his model, the CIA set up a kind of forecasting face-off that pit predictions from his model against those of Langley’s more traditional in-house intelligence analysts and area specialists. “We tested Bueno de Mesquita’s model on scores of issues that were conducted in real time—that is, the forecasts were made before the events actually happened,” says Stanley Feder, a former high-level CIA analyst. “We found the model to be accurate 90 percent of the time,” he wrote. Another study evaluating Bueno de Mesquita’s real-time forecasts of 21 policy decisions in the European community concluded that “the probability that the predicted outcome was what indeed occurred was an astounding 97 percent.” What’s more, Bueno de Mesquita’s forecasts were much more detailed than those of the more traditional analysts. “The real issue is the specificity of the accuracy,” says Feder. “We found that DI (Directorate of National Intelligence) analyses, even when they were right, were vague compared to the model’s forecasts. To use an archery metaphor, if you hit the target, that’s great. But if you hit the bull’s eye—that’s amazing."

David Brin suggests that some kind of political system populated with humans and diverse but imperfectly rational and friendly AIs would evolve in a satisfactory direction for humans.

I don't know whether creating an imperfectly rational general AI is any easier, except that limited perceptual and computational resources obviously imply less than optimal outcomes; still, why shouldn't we hope for optimal given those constraints?  I imagine the question will become more settled before anyone nears unleashing a self-improving superhuman AI.

An imperfectly friendly AI, perfectly rational or not, is a very likely scenario.  Is it sufficient to create diverse singleton value-systems (demographically representative of humans' values) rather than a consensus (over all humans' values) monolithic Friendly?  

What kind of competitive or political system would make fragmented squabbling AIs safer than an attempt to get the monolithic approach right?  Brin seems to have some hope of improving politics regardless of AI participation, but I'm not sure exactly what his dream is or how to get there - perhaps his "disputation arenas" would work if the participants were rational and altruistically honest).