Trust But Verify and the Maes-Garreau Tendency

"Trust but verify" is a good way to lead one's life. Ideally, we'd never take anyone's word for anything, and have the time and means to dig up supporting evidence for any position or statement that we encounter. But who has the time for that? We have to organize our busy lives around blocks of selective ignorance, portions of human knowledge and culture wherein we choose to take statements at face value, or follow the consensus viewpoint without doing the necessary groundwork to validate it. Books can and have been written on how to best go about this: acquiring and processing information costs time, and time is the most valuable resource most us of possess.

There exist a growing number of people propagating various forms of the viewpoint that we middle-aged folk in developed countries may (or might, or certainly will) live to see the development and widespread availability of radical life extension therapies. Which is to say medical technologies capable of greatly extending healthy human life span, probably introduced in stages, each stage effective enough to grant additional healthy years in which to await the next breakthrough. You might think of Ray Kurzweil and Aubrey de Grey, both of whom have written good books to encapsulate their messages, and so forth.

Some people take the view of radical life extension within our lifetimes at face value, whilst others dismiss it out of hand. Both of these are rational approaches to selective ignorance in the face of all science-based predictions. It usually doesn't much matter what your opinion is on one article of science or another, and taking the time to validate science-based statements usually adds no economic value to your immediate future. It required several years of following research and investigating the background for me to feel comfortable reaching my own conclusions on the matter of engineered longevity, for example. Clearly some science-based predictions are enormously valuable and transformative, but you would lose a lifetime wading through the swamp of uselessness and irrelevance to find the few gemstones hidden therein.

As a further incentive to avoid swamp-wading, it is generally well known that futurist predictions of any sort have a horrible track record. Ignoring all futurism isn't a bad attention management strategy for someone who is largely removed from any activity (such as issuing insurance) that depends on being right in predicting trends and events. You might be familiar with the Maes-Garreau Law, which notes one of the incentives operating on futurists:

The Maes-Garreau Law is the statement that "most favorable predictions about future technology will fall within the Maes-Garreau Point", defined as "the latest possible date a prediction can come true and still remain in the lifetime of the person making it".

If you want to be a popular futurist, telling people what they want to hear is a good start. "You're not going to be alive to see this, but..." isn't a compelling opening line in any pitch. You'll also be more convincing if your yourself have good reason to believe in your message. Needless to say, these two items have no necessary relationship to a good prediction, accuracy in materials used to support the prediction, or whether what is predicted actually comes to pass. These incentives do not make cranks of all futurists - but they are something one has to be aware of. Equally, we have to be aware of our own desire to hear what we want to hear. That is especially true in the case of predictions for future biotechnology and enhanced human longevity; we'd all like to find out that the mighty white-coated scientists will in fact rescue us from aging to death. But the laws of physics, the progression of human societies, and advance of technological prowess don't care about what we want to hear, nor what the futurists say.

I put value on what Kurzweil and de Grey have to say about the potential future of increased human longevity - the future we'll have to work to bring into being - because I have performed the due diligence, the background reading, the digging into the science. I'll criticize the pieces of the message I don't like so much (the timescale and supplements in the case of Kurzweil, WILT in the case of de Grey), but generally I'm on board with their vision of the future because the science and other evidence looks solid.

But few people in the world feel strongly enough about this topic to do what I have done. I certainly don't feel strongly enough about many other allegedly important topics in life to have done a tenth as much work to validate what I choose to believe in those cases. How should one best organize selective ignorance in fields one does care about, or that are generally acknowledged to be important? What if you feel - correctly, in my humble opinion - that engineered longevity is very important, but you cannot devote the time to validate the visions of Kurzweil, de Grey, or other advocates of longevity science?

The short answer is trust networks: find and listen to people like me who have taken the time to dig into the background and form our own opinions. Figuring out whether ten or twenty people who discuss de Grey's view of engineered human longevity are collectively on the level is not too challenging, and doesn't require a great deal of time. We humans are good at forming accurate opinions as to whether specific individuals are idiots or trustworthy, full of it or talking sense. Fundamentally, this establishment of a trust network is one of the primary purposes of advocacy in any field of endeavor. The greater the number and diversity of advocates to have taken the time to go digging and come back to say "this is the real deal," the more likely it is that that they are right. It's easy, and probably good sense, to write off any one person's views. If twenty very different people are saying much the same thing, having independently come to the same viewpoint - well, that is worth spending more time on.

One of the things I think we need to see happen before the next decade is out is the establishment of more high-profile longevity advocates who discuss advancing science in the Kurzweil or de Grey vein: nanotechnology, repairing the molecular damage of aging, and so on. Two, or three, or five such people is too few.


Very thoughtful article, thank you; especially the first part.

Still, I do have some issues with your thesis in the second part: "If twenty very different people are saying much the same thing, having independently come to the same viewpoint." The numbers don't really matter, as long as they are, just to put a minimum on it, in the >20 field. But there are two issues here, which I think make 20 just as doubtful as one.

The first is that thing about 'independence'. The whole point about a network is that it does not consist of independent units. If they were, they wouldn't be parts of the network. And even if they aren't in the network officially, they still will get their input from others who are networked, because that's the way to get the information. Science is based on dependency, which means that one scientist's presuppositions are based on the results published by others. And so on. So, no, 'independence' is wishful thinking.

Secondly, it's implicit in the nature of things that those who conduct survey-type research into longevity topics do so because they are interested in it; and that invariably means that they would very much like to live forever themselves. (Either that or they are out to prove it all as a nonsense; which will also predispose them toward favoring certain data that produce the results they want to see.) Twenty or more people are just as likely as a single individual therefore, even as a group of quasi-independent entities, to end up with a collective result that's more the product of wishful thinking than anything else.

The notion that a crowd is somehow wiser than a single person, maybe because collectively its members 'see' more, is profoundly flawed. It ignores the lack of the members' observational independence. Networks can lay no more claim to trustworthiness or truth-detection-capability than individuals. It would perhaps be comforting if it were different, but the evidence is overwhelmingly against it.

Posted by: Till Noever at March 28th, 2010 2:56 PM
Comment Submission

Post a comment; thoughtful, considered opinions are valued. New comments can be edited for a few minutes following submission. Comments incorporating ad hominem attacks, advertising, and other forms of inappropriate behavior are likely to be deleted.

Note that there is a comment feed for those who like to keep up with conversations.