D2What a Reading List Is For

A reading list looks like an administrative document and functions as a definition. By naming some texts and not others it states what the subject consists of, in what order it should be met, and what a competent person in it is assumed to have read. Arguments about lists are rarely about the individual books; they are about that definition.

Three lists in one document

Most lists silently combine three incompatible purposes. There is the set text, which will actually be worked through and assessed. There is the recommended reading, which is a description of where the subject goes next and which most readers will sample rather than complete. And there is the reference shelf, consulted at need and never read straight through.

A list that does not distinguish these is a bad instrument, because it is read as an obligation and is impossible to meet. The most useful single improvement to any reading list is to mark which category each item is in and to say plainly how much of it is expected. A reader who knows they are meant to read one chapter reads the chapter. A reader who believes they are meant to read all of it reads none of it.

Length as a signal

Long lists are often defended as generous, and they usually function as the opposite. A list of sixty items with no ordering supplies no guidance at all; the reader must construct the priorities the compiler declined to state, which is the harder half of the work. A short list with reasons is far more generous, even though it looks meaner, because the reasons transfer the compiler's judgement rather than only their conclusions.

The one-line annotation is the most efficient element of the whole document. Naming what a text is good for, what is dated in it and what it argues against tells a reader whether it is theirs to read now, and takes a sentence.

Availability is part of the argument

A list is a claim about what should be read, and a list of unobtainable books is therefore a hollow claim. Whether an item is in the collection, whether it can be borrowed, whether an accessible version exists, and whether it exists in a format that can be enlarged or heard are all part of whether it is genuinely on the list. Compilers who check availability are doing pedagogical work, not clerical work.

The same applies to edition. Where pagination matters for discussion, the list should say which edition is assumed, since reflowable electronic editions have no stable pages and a class working from four different editions cannot easily arrive at the same passage.

Revising a list

Lists accumulate. Items stay because removing them requires a decision and keeping them does not, so the document slowly becomes a record of what previous compilers thought rather than a statement of what this subject now requires. Periodic deletion is the difficult and necessary part of maintaining one, and it is best done by asking what each item is doing rather than whether it is good.

The honest test for any list is simple: could a reader who followed it exactly arrive where the compiler intended? If not, the list is not describing the route the compiler actually took, and the missing pieces are usually the ones the compiler absorbed so early that they forgot they were reading at all.