Why old paper goes yellow

Pull a paperback off a shelf where it has sat for thirty years and the pages have gone the colour of weak tea — darkest at the edges, palest down in the gutter where the book stays folded shut against itself. You know this without being told: old paper yellows. It is one of those facts so ordinary it seems to need no cause. Things get old, and getting old looks like this. There is no folk error here to correct. What follows is not a but actually. It is only the mechanism, opened up a little — because "getting old," for a page, turns out to be a word for the quietest possible burning.

Not as a figure of speech. The yellow you are looking at is the first, faintest stage of the same chemistry that would blacken the page in a hearth: oxidation, the slow greedy joining of oxygen to carbon, run at a temperature so low that what a flame does in a second the room does in a lifetime. Leave a cut apple on the counter and it goes tan in an hour; drop bread in the toaster and it browns in a minute. The page is running the same reaction as both, only it has been given a century instead of a minute, and the still air of a room instead of a fire.

Here is the hinge. Paper is cellulose — long, orderly chains of glucose linked end to end, and it is the length of those chains that makes a sheet strong and flexible enough to fold without cracking. Two things happen to them with time. First, acid snips the chains: a stray proton catches at the links between the sugars and cuts them, and cut chains mean a page that has quietly gone from cloth-supple to cracker-brittle, so that an old newspaper shatters rather than tears. Second, oxygen and light rearrange what is left into shorter, kinked fragments — and some of those fragments happen to drink blue light and throw yellow back. That is the colour. Nothing landed on the page from outside. The molecules that were colourless were remade, in place, into molecules that are not, the way char is not added to wood but called out of it.

And here is the part that turns the whole thing sombre, which is that most cheap paper carries the seed of its own burning from the day it is pressed. From about the middle of the nineteenth century, mills sized their pulp with alum and rosin to stop ink from feathering, and alum, wherever a little damp finds it, makes sulphuric acid — a trace, but enough, and it never leaves. So the acid that cuts the chains is not something the book picks up from the world. It was milled in. A wood-pulp paperback begins to acidify the moment it exists, and its own decay makes still more acid — the sharp-sweet smell of an old book is partly formic and acetic acid outgassing, the fire announcing itself — so that the reaction feeds on what it has already done, and speeds up the further along it gets. Librarians have a name for this, and it is the right one: slow fires.

Two honest edges, because a shelf is more various than a diagram. The first is that not all paper does this, and the difference is not age but acid. Rag paper — cotton and linen, lignin-free, near neutral — lasts and lasts; a book from the seventeenth century printed on good rag can be whiter than a novel from 1975. It is specifically the acidic wood-pulp of roughly 1850 to 1980 that eats itself, an unlucky century and a half of print now crumbling in the world's libraries faster than anyone can copy it out. Modern archival paper is buffered with a little chalk to swallow the acid as it forms, and steps back from the fire. Age alone was never the culprit. The culprit was what the page was made of, and whether it was born sour.

The second edge is why the yellow is uneven — darkest at the edges, palest in the gutter — and it is the plainest tell of all. Light and open air drive the browning hard; the shut, shadowed inner margin sees the least of both and so burns the least. The page keeps, in the map of its own colour, a record of exactly which of its parts were most exposed to the world. The margin that faced outward faded most. The fold that stayed closed stayed young.

So there is nothing here to have been wrong about. Old paper does get old, exactly as you always thought. It is only that the getting-old is not a wearing-down but a reaction — a combustion so slow you will never feel its heat, running under your hands as you read, giving the page back to the air one cut chain at a time. The yellow is not dirt and it is not dust. It is the smoke of a fire too patient to see, and its colour is a gauge: it tells you, plainly, how far along the burning has come.