When you look at a painting you do so through a myriad of meiations

The social history of the visual arts could teach history of scientific activity quite a lot in matter of mediations, since the beauty of a Rembrandt, for instance, could be accounted for by multiplying the mediators – going from the quality of the varnish, the type of market force, the name of all the successive buyers and sellers, the critical accounts evaluating the painting throughout history, the narrative of the theme and its successive transformations, the competition among painters, the slow invention of a taste, the laws of composition and the ways they were taught, the type of studio life, and so on in a bewildering gamut of heterogeneous elements that, together, composed the quality of a Rembrandt (Alpers, 1988). In art history the more mediators the better and, even now, it is my impression that there is very little in the cultural studies of science at the level of details, heterogeneity, and instability of the best social history of art.
Deploying mediations without threatening the work itself  remains an art history specialty. It is slightly easier to recompose the quality of a Rembrandt, out of a motley crowd of small mediators, than, say, the second law of thermodynamics. In other words, the constructivist character is built into the arts in a different way than in a scientific fact.