If you want to follow this stuff on the cutting edge, the site to visit is ClimateAudit, which is run by Steve McIntyre, the intrepid outsider who caught Mann with his pants down on the "hockey stick" graph. (Actually ClimateAudit is overwhelmed right now with new visitors because of Climategate, and so McIntyre is placing his new posts at a mirror site, here.)

The problem with McIntyre is that he shares Ludwig von Mises' expectations on the knowledge of the reader. Just as Mises would make one-off comments about Bohm-Bawerk's capital theory or the Weber-Fechner physiological law, so too does McIntyre assume all of his thousands of readers are totally familiar with the history of climate science and just want McIntyre to give them the actual freaking computer code to replicate Jones graph from 1999. (!!)

So anyway, that's why you have me here. Lately my purpose has not been to invent new material but to translate the work of other giants into a version that you, the unwashed masses, can understand...

Now that I've made the opening funnies, let's get down to business. I had been aware of this (vaguely) but it didn't really hit me until a commenter at ClimateAudit said it explicitly. To set the context, PaulM first said:

Steve I really think you need to explain things more clearly for the thousands of new readers who are now reading your blog. Most of them will not understand this. I’ll have a go, please correct:

The green line is tree-ring data on the left. On the right it has been smoothly merged into temperature data. This is Michael Mann’s trick, that he falsely claims is never done. The reason they do it is to hide the fact that otherwise the green tree-ring data curve would go down in the late 20th century rather than up, showing that the tree-ring data is useless at representing temperature. Hence “hide the decline”.

Then to endorse this take (which I endorse as well, for what that's worth), Calvin Ball said:

Paul, that’s my take. And I agree that this is important for the masses to get; the popular interpretation of “hide the decline” is somehow concealing the lack of warming over the past decade. This is completely wrong, but it’s catching on.

Does everybody get what he's saying? When Jones said in the infamous CRU email that he used "Mike [Mann]'s Nature trick...to hide the decline," he was NOT talking about a decline in actual global temperatures that thwarted the claims of global warming.

On the contrary, he was talking about a decline in the proxy for global temperatures based on tree ring observations. So what happens is that from 1960 on, the proxy starts dropping while the actual temperatures go up. (See the divergence in this post.) Ironically then, if actual global temperatures had NOT risen in the last half of the 20th century, then there would be less of a problem to "hide."

Now you might ask, "Huh? Why do they need to hide the decline in a proxy of temperature, when we already know the actual temperature post-1960?"

Good question; I'm glad you asked. The answer is that to make the claim that the warming of the 20th century is unprecedented in the last x years (I don't remember the actual number the alarmists use, but it's big)--meaning it must be due to human activity and not natural variability--we need to know what the global temperature was, say, 800 years ago. And unfortunately all the educated people back then were busy copying the Bible by hand, rather than setting up temperature stations (and being careful to filter out the Monastery Heat Island effect).

So in order to demonstrate that the 20th century warming is unusually rapid, the climate scientists have to construct a proxy of global temperatures going back before the period of modern instrumentation. One of these series is based on tree rings.

Now you see the problem. If the tree-ring proxy diverges sharply from actual recorded temperatures from 1960 onward, then we can't trust it when it tells us that temperatures have been fairly stable before modern economic growth. Hence the claims of the alarmists are undercut.

In conclusion, Jones was not trying to "hide" the lack of global warming since 1998, as many people are probably concluding. So the problematic phrase in his email is "hide," not "the decline."

(I am pretty sure my explanation above is correct, but by all means if someone thinks I have botched a particular point please bring it to my attention.)