Read here, PDF, here, here and here. Way back in 1997, researchers published a paper that was based on data from 6,000 plus borehole sites from all the continents. The reconstructed temperatures clearly showed a Medieval Period warming that was, and is, unprecedented. The data also makes clear that subsequent warming began well before the growth of human CO2 emissions and this natural rebound would obviously lead to temperatures similar to the Medieval Period.
A year later, the infamous Mann hockey-stick temperature chart was published to wild acclaim by the IPCC and AGW-centric activists. So popular did the Mann chart become, the 6,000+ borehole chart was completely ignored since its data refuted the Mann study. The borehole scientists then decided to re-publish their study with primarily only the blue-side (the typical AGW-favored data cherry-picking) of the chart below. This repackaged borehole study became accepted by the AGW-centric scientists as it seemed to support their cause and the Mann's hockey-stick. (click on image to enlarge)
"The authors searched the large database of terrestrial heat flow measurements compiled by the International Heat Flow Commission of the International Association of Seismology and Physics of the Earth's Interior for measurements suitable for reconstructing an average ground surface temperature history...Based on a total of 6,144 qualifying sets of heat flow measurements obtained from every continent of the globe, they produced a global climate reconstruction, which, they state, is "independent of other proxy interpretations [and] of any preconceptions or biases as to the nature of the actual climate history."...From their reconstruction of "a global climate history from worldwide observations," the authors found strong evidence that the Medieval Warm Period was indeed warmer than it is now."
"Quite suddenly, the same borehole authors - Pollack, Huang, Shen published a new, two-page-long paper in Nature: it appeared in October 1998. The paper contained a rather different graph than the graph from 1997...The new paper was using temperatures and 358 sites only instead of the 6000 sites used in 1997 (94 percent of sites eliminated) and it has erased 19,500 years out of 20,000 years (97.5 percent of the time interval eliminated) from the paper written in 1997 in order not to contradict Mann et al....That's what they call "independence". Moreover, if someone wanted to extend the record as far as possible while avoiding any hints of a warmer period in the past such as the medieval warm period, he would have made the same cut: 500 years ago. What a coincidence."