In Feist, the Supreme Court dealt with what it viewed as the fact versus compilation-of-fact dichotomy: facts aren't protectible, but compilations can be. But why should compilations of fact be protectible? Why should the aggregation of 100 unprotectible elements magically be transformed into something different that the parts, Justice O'Connor wondered. How, in other words, is the sum of the parts intrinsically different from each individual part, no matter how many individual parts there are?
The Court answered that question by examining the nature of each side of the equation. As regards facts, they weren't protectible because no one was their author; they are, we are told, objectively existent, a priori. Compilations of facts, on the author hand, can be protectible if there is an original selection, coordination, or arrangement: one can be the author of a compilation because it isn't a priori. This left open the question of what an original selection, coordination, or arrangement is, but given the alphabetical arrangement in Feist, it is no surprise the Court focused on selection as the key element of originality, the exercise of judgment in what to select (the 100 best vegetarian restaurants in Houston, for example).
Claims of copyright in historical works has suffered from courts' reflexive treatment of historical "facts" as facts like addresses in white pages. Historical facts exist a priori, are discovered, and can't be authored, ergo, they can be copied at will. But what is a historical fact? There are events that are documented or otherwise not disputed: that the U.S. Senate's confirmation hearings of John Roberts to replace Chief Justice Rehnquist began today. But those types of facts are what Macaulay called the "dross of history," things historians don't see as their raison d'etre, favoring instead narration as the real conveyor of meaning, a view espoused in the 1970s by American bad-boy historian Hayden White.
True enough in the past some historians referred to "facts, facts, facts - which carry within themselves their lesson and their philosophy" (address of Henri Houssaye at 1900 First International Congress of Historians in Paris), but others, like Frederick Turner 10 years later believed that facts were part of the times and could only be seen as "true" with considerable distance. Carl Becker, in 1958, believed that it is "almost impossible" to distinguish facts from theory, and three years later, English bad-boy historian E.H. Carr wrote that the "facts of history never come to us 'pure' since they do not and cannot exist in a pure form: they are always refracted through the mind of the recorder." Still others, like Paul Conkin and Michael Stanford regarded facts as something like a communal judgment, something a majority of historians agreed upon.
In the 1990s, some historians like Keith Jenkins, viewed facts as "only a description of things that happened, and which, therefore, cannot have an intrinsic meaning." Although there was a phase in which the influence of the "linguistic turn," Roland Barthes and post-structuralists threatened to undermine the legitimacy of the historical exercise altogether, the field seems fragmented today, with no consensus at all, a fact that doesn't seem to faze courts in copyright cases.