This week PunditFact played that game with the following tweet from conservative pundit Phil Kerpen:
Amazing fact: Senate has already voted on more amendments in 2015 than Reid allowed ALL YEAR last year.PunditFact found that the new Republican-controlled Senate has already voted on more amendments in 2015 than Reid allowed in the Democrat-controlled Senate for all of 2014: "On the numbers, that is right."
— Phil Kerpen (@kerpen) January 23, 2015
But PunditFact went on to find fault with Kerpen for leaving out needed context:
On the numbers, that is right. But experts cautioned us that the claim falls more in the interesting factoid category than a sign of a different or more cooperative Senate leadership.We'll spell out the obvious problem with PunditFact's rating: Kerpen's tweet doesn't say anything about different or more cooperative Senate leadership. If Kerpen's not making that argument (we found no evidence he was), then it makes no sense at all to charge him with leaving out information. In effect, PunditFact is amending Kerpen's tweet, giving it context that doesn't exist in the original. Kerpen's statement doesn't need clarification or additional information to qualify as simply "True."
The statement is accurate but needs clarification and additional information. That meets our definition of Mostly True.
PunditFact's rating offers us a perfect opportunity to point out that if Kerpen's statement isn't simply "True" then there's probably no political claim anywhere that's immune to the type of objection PunditFact used to justify its "Mostly True" rating of Kerpen. A politician could claim the sky is blue and the fact checker could reply that yes, the sky is blue but no thanks to the policies of that politician's party! There are endless ways to rationalize withholding a "True" rating.
This rating convinces us that it would be productive to look at the breakdown between "True" and "Mostly True" ratings to look for a partisan bias. Since there's always context missing from political claims, drawing that line between "True" and "Mostly True" may prove no more objective than the line between "False" and "Pants on Fire."