While the IPCC is big on the idea that the warmth of the late 20th and early 21st century in the Northern Hemisphere is unprecedented in recent centuries, apparently that finding does not apply universally over longer timescales.
According to the Summary of Policymakers from the IPCC Fourth Assessmnet Report:
Average Northern Hemisphere temperatures during the second half of the 20th century were very likely higher than during any other 50-year period in the last 500 years and likely the highest in at least the past 1,300 years.
This is basically a verbal description of the "hockeystick"-like temperature progression of the past millennia or so.
How that representation came to be and just how scientifically accurate it is a story unto itself, and one which continues to be assessed and reassessed over at the Climate Audit website. An interesting discussion has been taking place there as to yet another methodological flaw in the mathematics involved in multiproxy reconstructions. And another oft-discussed issue there is the very selective use of only particular proxy temperature records which are combined to produce the now-too-familiar hockeystick shape.
One proxy record that most definitely was not included in the assembly of the hockeystick is a just-published proxy reconstruction of sea surface temperatures (SSTs) from the East China Sea.