The Graduate Student’s Library: On Owning Books You Cannot Carry
There is a particular kind of grief familiar to anyone who has moved between graduate programmes. It arrives at the moment you stand in front of a bookshelf holding a copy of Being and Time that you annotated across two years, and you calculate what it will cost to ship it four hundred miles, and you consider — briefly, shamefully — leaving it behind.
I want to take that moment more seriously than it usually gets taken, because I think it exposes something about how philosophers relate to their working libraries that we rarely examine.
The book as instrument
We tend to speak about books instrumentally. A book contains arguments; the arguments are what matter; the physical object is a delivery mechanism, and an inefficient one at that. On this view the digital copy is strictly superior — searchable, weightless, infinitely reproducible. The attachment to paper is sentiment, and sentiment is not an epistemic category.
I have made this argument myself, usually while trying to justify buying another book.
But the instrumental account struggles with something obvious. My annotated copy of a text is not interchangeable with a clean copy of the same text. The marginalia constitute a record of my engagement with the argument over time — including, crucially, the parts where I was wrong. The pencil note from 2019 that I crossed out in 2022 preserves a version of my thinking that no digital revision history captures, because I never would have thought to save it.
Nelson Goodman’s distinction between autographic and allographic works is useful here, though not in the way he intended. A musical score is allographic: any accurate copy is the work. A painting is autographic: the physical object is the work. Books are allographic in their content and autographic in their use. The text is reproducible. My relationship with the text is not.
The practical problem this creates
Which brings us back to the moving van.
A working philosophy library of any seriousness runs to several hundred volumes. Mine, at the point I left my second institution, was around eight hundred — modest by the standards of anyone with a permanent post, substantial by the standards of someone renting a room.
Eight hundred books weigh roughly four hundred kilograms. They occupy something in the region of three cubic metres. They cannot be flown, are expensive to ship, and are the single largest logistical obstacle in a move that also involves an entire life.
The standard graduate solution is triage. You keep the annotated ones and the out-of-print ones, you sell the duplicates, you tell yourself that the collected works you bought in a moment of optimism can be reacquired if the need arises. This is sensible and it is also a small ongoing loss, because the need does arise, and reacquisition means buying a clean copy that carries none of your history with the text.
The other solution, which I came to late and wish I had come to earlier, is simply to decouple the library from the move. Not everything needs to arrive at the new flat in the first week. A substantial fraction of a working library is reference material consulted twice a year, and there is no epistemic reason it must sit within arm’s reach.
Most university cities have arrangements for this. In London, where the gap between what a postgraduate can afford in floor space and what a postgraduate accumulates in books is at its most absurd, services offering student storage london exist more or less specifically for the mismatch — term-time collection, summer holding, the assumption that you will move again within a year. The economics work out favourably against either shipping repeatedly or renting a larger room to accommodate a shelf.
I mention this not as advice — the APA blog is not the place for logistics — but because the availability of the option changes the shape of the philosophical problem. If the choice is between owning a book and carrying it, sentiment loses to physics. If the choice is between having a book at hand and having it retrievable in a week, the calculation is entirely different.
What we lose when we triage
There is a version of this essay that ends with a defence of the physical book against digital reproduction, and I want to resist writing it, because I do not think that is the interesting problem.
The interesting problem is that graduate mobility imposes a selection pressure on personal libraries that has nothing to do with intellectual merit. What survives a move is what is portable, not what is important. Over three or four moves — which is a normal early career — that pressure compounds. The library that arrives at your first permanent post is not the library you built. It is the subset that was convenient to transport.
I do not think this is catastrophic. Books are replaceable, mostly. But I notice that the things I lost were disproportionately the large ones: the collected editions, the heavy critical apparatus, the primary sources in languages I read slowly. The things that survived were paperbacks.
There is something worth thinking about in a discipline whose material culture is quietly shaped by what fits in a car.
I want to take that moment more seriously than it usually gets taken, because I think it exposes something about how philosophers relate to their working libraries that we rarely examine.
The book as instrument
We tend to speak about books instrumentally. A book contains arguments; the arguments are what matter; the physical object is a delivery mechanism, and an inefficient one at that. On this view the digital copy is strictly superior — searchable, weightless, infinitely reproducible. The attachment to paper is sentiment, and sentiment is not an epistemic category.
I have made this argument myself, usually while trying to justify buying another book.
But the instrumental account struggles with something obvious. My annotated copy of a text is not interchangeable with a clean copy of the same text. The marginalia constitute a record of my engagement with the argument over time — including, crucially, the parts where I was wrong. The pencil note from 2019 that I crossed out in 2022 preserves a version of my thinking that no digital revision history captures, because I never would have thought to save it.
Nelson Goodman’s distinction between autographic and allographic works is useful here, though not in the way he intended. A musical score is allographic: any accurate copy is the work. A painting is autographic: the physical object is the work. Books are allographic in their content and autographic in their use. The text is reproducible. My relationship with the text is not.
The practical problem this creates
Which brings us back to the moving van.
A working philosophy library of any seriousness runs to several hundred volumes. Mine, at the point I left my second institution, was around eight hundred — modest by the standards of anyone with a permanent post, substantial by the standards of someone renting a room.
Eight hundred books weigh roughly four hundred kilograms. They occupy something in the region of three cubic metres. They cannot be flown, are expensive to ship, and are the single largest logistical obstacle in a move that also involves an entire life.
The standard graduate solution is triage. You keep the annotated ones and the out-of-print ones, you sell the duplicates, you tell yourself that the collected works you bought in a moment of optimism can be reacquired if the need arises. This is sensible and it is also a small ongoing loss, because the need does arise, and reacquisition means buying a clean copy that carries none of your history with the text.
The other solution, which I came to late and wish I had come to earlier, is simply to decouple the library from the move. Not everything needs to arrive at the new flat in the first week. A substantial fraction of a working library is reference material consulted twice a year, and there is no epistemic reason it must sit within arm’s reach.
Most university cities have arrangements for this. In London, where the gap between what a postgraduate can afford in floor space and what a postgraduate accumulates in books is at its most absurd, services offering student storage london exist more or less specifically for the mismatch — term-time collection, summer holding, the assumption that you will move again within a year. The economics work out favourably against either shipping repeatedly or renting a larger room to accommodate a shelf.
I mention this not as advice — the APA blog is not the place for logistics — but because the availability of the option changes the shape of the philosophical problem. If the choice is between owning a book and carrying it, sentiment loses to physics. If the choice is between having a book at hand and having it retrievable in a week, the calculation is entirely different.
What we lose when we triage
There is a version of this essay that ends with a defence of the physical book against digital reproduction, and I want to resist writing it, because I do not think that is the interesting problem.
The interesting problem is that graduate mobility imposes a selection pressure on personal libraries that has nothing to do with intellectual merit. What survives a move is what is portable, not what is important. Over three or four moves — which is a normal early career — that pressure compounds. The library that arrives at your first permanent post is not the library you built. It is the subset that was convenient to transport.
I do not think this is catastrophic. Books are replaceable, mostly. But I notice that the things I lost were disproportionately the large ones: the collected editions, the heavy critical apparatus, the primary sources in languages I read slowly. The things that survived were paperbacks.
There is something worth thinking about in a discipline whose material culture is quietly shaped by what fits in a car.