THE COLLECTIVE AUTHOR
In the formation of collaborations, there are practical questions that press upon us. How is an individual scholar to be evaluated for hiring and promotion? How can error be detected when every member of a team is not in a position to judge the final publication? But these and related questions are not mine here, at least in the first instance. I am after something different, I want to explore what it means, quite liter ally, for a collaboration to know something about the world, and I want to ask this question of the largest, most intricately technical scientific collectives ever established-the detector teams surrounding colliding beam accelerators at the end of the twentieth century. With collaborations mounting to over a thousand participants, it hardly takes algebraic topology to reckon that quite soon only a handful of these teams will embrace the careers of nearly all of the seven thou sand experimental particle physicists expected to be employed at the end of the century. But to ask in what sense a collaboration can know, argue, or show something, it may be useful to consider a Kantian analogy.