The First Lines: When Humans First Drew the World

Chapter 1: The First Lines: When Humans First Drew the World

In the autumn of 2011, deep inside a sea-facing cave on South Africa's southern Cape coast, an archaeologist knelt in red sand and brushed away forty thousand years of dust. Christopher Henshilwood had spent most of his career studying the Blombos Cave site, but on this morning his brush uncovered something that would not stay within the boundaries of his discipline. It was a small lump of silcrete, hardly larger than a matchbox, into which a human hand had once pressed a deliberate, geometric crosshatch: a diamond filled with parallel lines, framed by three straight lines running off its edge. Henshilwood and his team later dated the fragment to roughly 73,000 years ago. It was, in the language of the press release, "the world's oldest known drawing." But that phrasing, accurate as far as it went, missed the more interesting claim hiding inside the stone. The marks were not random. They were not doodles. They were the earliest surviving evidence that a human being had looked at a piece of the world and then, with a chiseled piece of ochre, chosen to *represent* it on a different surface. They were the first lines.

That is the question this chapter is built around, and it is older than any cave: *What is a map, and when did human beings first begin making them?* The answer, it turns out, is not a date. The answer is a journey — a slow, several-hundred-thousand-year walk from the moment our ancestors first scratched a line in dirt to mark where an antelope lay wounded, to the moment they bent over clay tablets and pressed the shape of an entire river valley into permanence. The map is older than writing by tens of thousands of years. It is older than the wheel, older than agriculture, older than the Pyramids. It is, in a real sense, the second technology — the first being language itself — that our species invented in order to *think* about the world rather than simply react to it.

What follows is a guided tour of that invention's earliest chapters. We will stand in the Blombos Cave with the ochre in Henshilwood's palm. We will examine a baboon fibula from a Swazi mountain. We will lean close to the lime-plastered walls of a Neolithic town in central Anatolia. We will crouch beside an alpine boulder in Lombardy where Bronze Age shepherds carved a network of fields and rivers into stone. And we will hold, as best we can without touching, a fragment of mammoth tusk from a Ukrainian village that has been called, with only mild exaggeration, the first atlas ever made. Along the way, I will propose a simple framework — three stages of mapmaking — that lets us sort these objects into a meaningful order rather than a chronological scrapbook. Then, near the end of the chapter, I will ask you to do something. I will ask you to put this book down, take a piece of paper, and draw the route from your front door to a place you know well — using only landmarks, never street names. In doing so, you will reproduce, almost exactly, the cognitive act that produced the world's first maps.

Let us begin, then, with a small piece of stone and a large question.

1.1 The Oldest Mark: Blombos, Lebombo, and the Problem of "First"

One of the deepest traps in writing the history of cartography is the word "first." Every textbook wants one. Every museum wants a vitrine with a single object and a label that says THE FIRST. The reality, as Henshilwood's ochre fragment reminds us, is messier. A discovery at one site in one decade often gets demoted when another site turns up an older example a decade later. The "first map" of the 19th century is not the "first map" of the 21st. What does not change, however, is the *category* of the question. We are not really asking for a calendar date. We are asking for a threshold — the moment when *marking* became *representing*. A footprint in mud is a mark. A footprint deliberately drawn on a rock, intended to stand for the absent foot, is a representation. Somewhere in our deep past, our ancestors crossed that threshold. The Blombos fragment sits on the line.

Let us look at it closely. The Blombos ochre — formally known as Blombos Cave engraved ochre specimen AUC-4380 — is a small flake of ferruginous siltstone that originally came from a larger piece of red ochre. The engraving, executed with a pointed bone or lithic tool, consists of three sub-parallel lines crossed at roughly a right angle by another set of lines, producing a kind of lattice. A second phase of engraving then overlays this with a row of smaller, more delicate hatchings. In 2002, when the specimen was first announced, the reaction was skeptical. Could the pattern be natural? Could it be the result of the stone being used as a working surface, accumulating scratches from butchering or hide processing? Henshilwood's team systematically ruled this out by examining wear patterns, by replicating the marks experimentally on similar stone, and by demonstrating that the lines were applied in multiple distinct phases of intentional activity. The crosshatched pattern, they argued, was *designed*. It was not merely a record of use. It was a graphic.

Why does that matter for cartography? Because the line between "abstract design" and "spatial representation" turns out to be fuzzier than we like to think. The Blombos artist may have been drawing a map of a landscape feature. He or she may have been drawing a pattern that referred to something we cannot now identify — a counting aid, a kinship diagram, a memory aid for a story. But the *cognitive operation* is the same in both cases. To commit an image to a portable surface, with the understanding that the image will outlive the moment of its making and serve as a reference for someone else (or even for one's later self), is to perform the essential trick of cartography. Once you can do this, you can, in principle, draw a map. The Blombos fragment does not show us a map. It shows us the *birth of the capacity* that will eventually produce maps.

Now let us travel forward in time but stay on the African continent, because the earliest surviving candidate for what most cartographers would actually recognize as a map comes from a different corner of southern Africa. The Lebombo bone was discovered in the early 1970s in a Border Cave-like context in the Lebombo Mountains of what is now Eswatini (formerly Swaziland). It is a fibula — the long shinbone of a baboon — and on its surface, someone carved a series of 29 parallel notches along one edge. The artifact was originally interpreted as a lunar calendar or a tally stick, the kind of object you find in ethnomathematical studies of pastoral societies. But in 2014, a French team led by Francesco d'Errico and Lucinda Backwell proposed a different reading. They pointed out that the notches are arranged in distinct subgroups that mirror the proportions of a real, geographically-coherent landscape. When the bone is rotated and oriented, the 29 notches could correspond to a roughly 30-kilometer stretch of the Pongola River, with each major notch representing a bend or confluence. If this interpretation is correct — and the argument is far from settled — the Lebombo bone would be the oldest known map-like representation of a real landscape, dating to perhaps 35,000 BCE.

Why might we call it "map-like" rather than just a "map"? Because the maker took an observed feature of the world (a river) and reduced it to a system of marks on a portable object (a bone), in such a way that the marks preserve certain *spatial relationships* of the original. The notches are not arbitrary. They vary in spacing in a way that seems to track variation in the river's bends. That is the cardinal sin and the cardinal virtue of a map: it is a lie that tells the truth. It compresses a three-dimensional landscape into a one-dimensional sequence of marks on a flat bone. It sacrifices everything except spatial relationship. And once you have made that s