Just about everything I know about provenance - which isn't a lot - made it into The Discipline of Organizing Section 3.5.4, so I was pleased this morning to read a story in the 4 May 2013 NY Times about how the Metropolitan Museum of Art has decided to return some 10th century Khmer sandstone sculptures to Cambodia after determining that they were looted during the 1970s when Cambodia was in chaos (http://nyti.ms/137Owvm). I'll put this case study in my course syllabus because it nicely frames the dilemma that incomplete provenance poses for museums today. The Met has a staggering collection of artifacts from everywhere, but I especially remember galleries with Eqyption, Greek, and Roman artifacts - much of which was acquired when the conventions and practices for collecting were much looser than they are today, and indeed, the Met has returned some things to Egypt and Italy in recent years. But at the same time, one can argue that if the Met and other museums hadn't collected these resources, they wouldn't exist today, because Eqypt and Italy weren't capable of caring for them.
One party who comes out looking bad in the story is Sotheby's, which in 2011 tried to auction a sculpture that undeniably comes from the same site as the ones the Met is repatriating; a Sotheby's statement argues that "the Met's voluntary agreement does not shed any light on the key issues in our case". The case Sotheby mentions is one in which the US government is suing to obtain the sculpture on behalf of the Cambodian government.