Chapter 2. Reading a timeline

Every timeline carries exactly two pieces of information, and they have to be told apart to read it correctly. The first is ORDER: what is written on the left happened before what is written on the right. The second is DISTANCE: the further apart two landmarks are on the line, the more time has gone by between them. A timeline that respected only the first would be a plain list; it is the second that makes it a timeline.

The French official syllabus asks that pupils be able to place the great periods studied on a timeline, the end of its sentence naming school levels that this book does not take up here. Placing means making both moves together: putting one fact after another, and putting it at the right distance. The second move is the one that gets missed, because it requires comparing durations and not dates.

This chapter breaks down the two moves, gives the way to build a timeline from a list of dates, and names the four reading errors that come back most often. It also explains why the timelines in this book are tables, and why that is not a degraded version of a drawn timeline.

An honest timeline of human history is almost impossible to draw on a page. The reason is arithmetical. If the oldest traces of our species are taken as the starting point, something in the order of three hundred thousand years has gone by. The last five thousand five hundred years, those for which written texts are available, represent less than two per cent of that.

On a line of twenty centimetres drawn to scale, the whole of written history would therefore fit into the last four millimetres. Everything this book calls Antiquity, the Middle Ages, Modern Times and the Contemporary Era would be indistinguishable in it. The timeline would be correct, and it would be of no use.

The timelines met everywhere solve the problem by breaking the scale: Prehistory occupies a short segment, and the four other periods spread out. That is a legitimate decision, on one condition, which is almost always forgotten: it has to be SAID. A broken-scale timeline that does not signal its break teaches the reader something false, namely that Prehistory is a brief period.

This book makes the opposite choice and owns it. Its timelines are tables, and a table claims no scale: it gives the order and it gives the durations in plain figures, in a column. A reader who wants to picture the proportion finds the number written down, instead of deducing it from a distance whose faithfulness is unknown.

A drawn timeline passes on its information through position: the place of a point on the line IS the data. That works for whoever looks at it. It does not work for whoever listens to it, because reading software restores a page line by line, and a line-by-line reading has no surface. Position is lost, and with it all the information.

The body that writes the standards of the web, in its documentation on complex images, describes what has to be supplied alongside a graphic: a short description to identify it, and a long description that restores the essential information. It then gives the shape that this long description takes, and that is worth remembering. The document is written in English, and its wording says that the long description is presented as headings, text and a table.

In other words, the form this book uses is not a makeshift: it is the one this body gives as an example for restoring a graphic. The document cited is a tutorial and not a normative text, and that has to be said; but it is indeed the one followed by those who make a figure accessible. A book that writes its timelines directly as headings, text and tables produces from the start the content that the drawn version would have had to supply in addition to itself.

A wide table raises a practical question: on a telephone, it does not fit across the screen. The accessibility rule that forbids scrolling a page in both directions EXEMPTS data tables by name, provided that the table scrolls inside its own area rather than carrying the whole page with it. That is what this book does: the page never moves sideways, the table alone does, and it stays reachable from the keyboard.

There is a price, and it is honest to say so. A table does not give the immediate impression of proportion that a drawing gives at a glance. This book makes up for that by writing the durations out in plain figures and by repeating them: the proportion is then read instead of being felt. A reader who wants the glance can build their own timeline from the tables, and the procedure above is made for that.

This small timeline is deliberately built on a subject already dealt with in the previous chapter, so that the reader can check every line without learning anything new. It illustrates three things at once.

First, the duration column brings out what a drawn line would have hidden: the first interval is worth about one thousand and fifty years, the last one hundred. A timeline drawn to a constant scale would have crushed the last four landmarks one on top of the other. Next, the first landmark carries two years instead of one, because the proposal and its adoption are seven years apart. A change rarely fits into a single date, and an honest timeline lets that gap be seen instead of arbitrarily choosing one of the two.

Finally, two of the five landmarks are not events but non-events: 1900 and 2000 are years in which a rule applied, without anyone noticing while living through them. A timeline can carry landmarks of this kind, and they are often the most instructive.