Here's a statement he recently made regarding a National Research Council committee he served on (I have a hard time keeping all these councils and academies and the like straight :)
"Our panel was asked to settle some of the puzzling inconsistencies that have helped spur the controversy over global warming in the first place. Many in the scientific community were troubled by an apparent incongruity between two different sets of temperature data. Surface-temperature measurements indicate that the Earth has warmed. But data collected by satellites and balloon-borne instruments since 1979 indicate little, if any, warming of the upper atmosphere.
After reviewing the data, we found that despite these differences in temperature data, the Earth has indeed warmed by about 1 degree Fahrenheit in the past century. In fact, surface temperatures in the past two decades have risen at a rate substantially greater than average for the past 100 years. These are important findings, to be sure. But we also need to be aware that although the Earth's surface temperature has increased substantially in the past 20 years, that increase may not necessarily be representative of any long-term climate change caused by greenhouse gas emissions.
The differences between the surface and upper-air data are probably due to a combination of factors. For example, depletion of ozone in the stratosphere may have cooled the atmosphere without affecting the surface temperature. And the debris that entered the atmosphere when Mt. Pinatubo erupted in 1991 also produced cooling, which was probably more pronounced in the satellite measurements than in those recorded at the Earth's surface. When these factors are taken into account in atmospheric models, the simulated trends agree more closely with the temperature observations, although some discrepancies remain.
So we know that despite differences in the two sets of temperature data, the Earth's surface is in fact warming. Our panel did not address whether greenhouse gases have led to the temperature increases of the past two decades."
What he's not saying here is that this is a huge retreat from his earlier position, that surface warming was not real, merely an artifact of the data, and that his satellite data PROVED it wasn't real.
He is also accepting the probability that some of the hypotheses put forward to explain the discrepency, such as the several year cooling effect of the Mt. Pinatubo eruption (important because the satellite data only goes back about tweny years, with half the dataset impacted by the eruption) .
So he's been forced to make a partial, not total, retreat from his earlier position.
Someone - a scientist, but not climatologist - recently told me Christy's balloon data that supposedly contradicts the Science paper above has itself been shot down but I've not checked it out ... I wouldn't be terribly surprised, though, because his first published methodology for indirectly computing temperatures in the atmosphere using his satellite data included a very embarassing error.