I may be an ignorant hack, but I know that if you have several sets of proxy data you can't mix and match parts of the sets to get the results you want. Using that technique you can get pretty much any result you want simply by selecting the parts of each set that fit your preconceived and desired result. In any given location and for any given proxy, local conditions play a strong part, often stronger than overall climate. Narrow tree rings and low carbon uptake may indicate cold weather, but it may also indicate a drought or other localized conditions (such as volcanic activity) that inhibit growth. Coral may grow more slowly when it's below its preferred zone, but it also grows more slowly when it's above its preferred zone, when water is stained, or when particular parasites or toxins are present. Ice cores are very poorly understood; the complicated and not necessarily linear cycle from snow to firn to ice is poorly understood, not to mention the effect of extreme pressure on trapped gas bubbles over time. Attempts to correlate ice cores to known historical conditions have been largely unsuccessful as well.
So now you have three sets of proxy data, but none of these can be anything like continuous or complete enough to be representative. Instead we have limited numbers of samples from scattered locations. Many of these samples must be thrown out - tree rings get labeled as drought years, ice cores get labeled as "disturbed", coral growth gets labeled as atypical. With enough samples, one could begin to build climate histories from each set and correlate them against each other. But if allowed to select from among the sets and the samples, such a construct becomes literally meaningless because different selection sets and samples will yield wildly different climate histories. (Else there would be no need to switch from one set to another, as the proxies would match.)
As far as politics goes, when Hansen was screaming that he was being censured he was giving two and three interviews a day warning of dire catastrophe and telling us what we had to do to avoid catastrophe. He is still not only telling us what he thinks the climate is doing (and as we've seen, manipulating data to support his belief) but constantly telling us what we have to do to avoid catastrophe. You can argue that he is right and is only doing this for our best interests, because he's so smart and we're so dumb, but you can't argue that he's not trying to control society without looking like an idiot. The entire CAGW movement spends much more time telling us what we HAVE to do than it does correctly predicting anything. (Oddly enough, what we have to do to combat catastrophic anthropogenic global warming is exactly what we were told to do to combat catastrophic anthropogenic global cooling.) Having bought into a man-made ice age, I'm going to be a bit more discerning this time around.
So far climate prediction theory is a half-step above "Give me your daughter or I'll make the moon disappear next Thursday night." It's a great predictor of things that have already happened, not so great for predicting things yet to occur - although it's pretty good at claiming to have predicted them afterward.