His CO2 estimates would represent 4 doublings of CO2 so they are clearly saying the senstivity to CO2 is on the very low end of the estimate - probably under 1.5C (rather than between 1.5C to 6.5C) for a doubling of CO2.
The article says that sensitivity is between 1.5 and 5.5 C. The Cambrian in this figure gets to 7.5 - 7.8 C higher than present, in the 4 doubling range. If the sensitivity is 2 C, that's 8 degrees C higher.
Now, the article says the "average" sensitivity was 3 C - maybe there's a period when the sensitivity is considerably higher. As the continents move around, climate is substantially affected -- the proposed explanation, from Crowley and Berner, of the brief (as can be nicely seen in ancient_geezer's plot) Ordovician glaciation is that Gondwanaland wandered directly over the South Pole.
He probably needed to downplay the result (while still publishing his research) so he could still get invited to all the great global warming parties (if you know what I mean).
Oh, I'm sure ol' Doc Berner still gets his invites. Especially those wine-and-cheese National Academy of Sciences parties.
Thanks for alerting me to that plot. It really demonstrates the Holocene stability "envelope" nicely, and everybody should remember that the much-maligned Hockey Stick doesn't show up until 1000 years BP.