by ashurbanipal » Thu 20 Oct 2005, 12:58:22
I'll take a shot at the second question. Upfront, my position is that Campbell is more credible than Lynch. Here's why:
1) As background, it should be noted that the two dates mean something in completely different domains. When Lynch says that Peak will occur in 2050 (if that's what he says--I know he thinks it's a long way off and I seem to recall 2050 is his pick), he's really understood as meaning that peak will happen a long time in the future. If he's still alive then, and he's roughly correct, he probably won't be saying that 2050 is THE year. When Campbell says 2007 or 2010, he's understood as meaning that exact year. So there's a difference in the semantic nature of each prediction.
This is important because by picking a far-off date, Lynch leaves himself quite a bit of wiggle-room. After all, if peak occurs in 2055, who would blame him for being 5 years off? Who would blame him, or say that he's flip-flopping, if he revised his prediction by 5 years right now? On the other hand, Campbell doesn't seem to get that kind of slack. If he switches from 2007 to 2010, people start howling about his credibility.
Seen in the same light as Lynch's prediction, though, Campbell's prediction is that peak is going to happen soon. Not later, and maybe not exactly when Campbell says. Just soon.
It may be said that predicting a far away date exactly is more difficult than predicting a close-in date, but I don't know that this is true. The factors that make the one more difficult make the other more difficult as well. Sure, there could be massive political upheaval between now and 2050. There could be massive political upheaval between now and 2010 as well. There could be a bunch of new data coming to light in 20 years. There could be a bunch of new data coming to light tomorrow. Over the long term, uncertainties are statistically more likely to creep in. But that's a double-edged sword--they don't necessarily creep in at a greater rate for any given period, but Lynch knows that the law of averages is on his side. In other words, knowing that everyone else thinks he's at a disadvantage, he's able to use that to his advantage. Not that I think he actually sits around and thinks about it that way.
2) To me, therefore, the question of who is more credible boils down to whether their methods and the available evidence support an earlier or later date for peak. Nevermind if they support exactly 2010 or exactly 2050. I believe that the available evidence supports an earlier date.
3) Finally, if you read his papers, Lynch hasn't been any better on short or long-term predictions than Campbell. Neither has anyone else that I've read.
That said, I've no doubt that there are lots of double standards out there.