Cartography is the study and practice of making maps, and its recorded history runs about 4,500 years — from Babylonian clay tablets showing river valleys and estates, through Ptolemy's coordinate geometry in the second century AD, the great chart-making of the Age of Exploration, the national trigonometric surveys of the eighteenth and nineteenth centuries, and finally to satellite positioning and the continuously updated digital maps of today.
Clay, papyrus and the first surviving maps
Maps predate writing systems in some respects, since a scratched diagram needs no alphabet. The oldest artefacts confidently identified as maps come from Mesopotamia. A clay tablet from Ga-Sur, near modern Kirkuk, dating to roughly 2300 BC, shows a river valley between hills with settlements marked and cardinal directions indicated at the edges.
The Babylonian Imago Mundi, from around the sixth century BC, is the earliest known attempt at a world map. It shows Babylon on the Euphrates, encircled by a ring labelled the "Bitter River", with triangular regions beyond it described in accompanying text. It is a cosmological statement as much as a geographic one, and that combination — measurement fused with worldview — recurs constantly in the centuries that follow.
Egyptian surveyors, meanwhile, developed practical mapping out of necessity. The Nile's annual flood erased field boundaries every year, so property had to be re-measured each season. The Turin Papyrus Map, from about 1150 BC, shows the Wadi Hammamat quarry region with different rock types distinguished by colour, which makes it arguably the first geological map.
Ptolemy's coordinates and their long afterlife
Greek thinkers supplied the theory. By the fourth century BC a spherical Earth was settled among educated Greeks, and around 240 BC Eratosthenes estimated its circumference by comparing the sun's angle at Alexandria and Syene, arriving at a figure within a few per cent of the true value depending on which length is assumed for his unit.
Claudius Ptolemy, working in Alexandria in the second century AD, produced the synthesis that mattered. His Geographia listed coordinates for roughly 8,000 places and set out methods for projecting a curved surface onto a flat sheet. No original maps survive — what came down to later centuries was the data and the instructions, from which maps could be reconstructed.
Ptolemy's authority became a problem as much as an asset. He underestimated Earth's circumference and stretched Asia too far east, and because his work was treated as canonical after its reintroduction to Europe around 1400, those errors persisted for centuries. The mistaken belief that Asia lay a manageable distance west of Europe drew directly on Ptolemaic figures, and shaped the expectations Columbus sailed with in 1492.
Medieval mappae mundi and the Islamic geographers
European medieval mapmaking is often dismissed as a regression, which misreads what those maps were for. The T-O mappae mundi placed Jerusalem at the centre and east at the top, with a T-shaped arrangement of water dividing the three known continents. The Hereford Mappa Mundi of around 1300 is a theological and historical diagram in map form — it was never intended for navigation.
Navigation was served by a separate and highly practical tradition. Portolan charts, appearing from the late thirteenth century, mapped Mediterranean coastlines with remarkable accuracy, covered in radiating rhumb lines for use with a magnetic compass. Their coastal outlines are recognisably modern in a way the mappae mundi never attempted.
Meanwhile, scholars in the Islamic world preserved and extended Ptolemy. Al-Idrisi, working at the Norman court in Palermo, completed the Tabula Rogeriana in 1154 — a world map with an accompanying geographic text that drew on merchant and traveller accounts across Africa and Asia, and which stood as among the most accurate available anywhere for several hundred years. Chinese cartography developed independently, with the Song dynasty Yu Ji Tu of 1136 carved in stone on a precise grid.
Exploration forces the map to change
From the fifteenth century, mapmaking became an information problem. Portuguese voyages down the African coast, then Columbus, then Magellan's circumnavigation completed in 1522, produced coastlines that no existing framework anticipated.
Printing changed the economics. Woodcut and then copperplate engraving allowed maps to be reproduced identically and cheaply, which meant errors propagated widely but so did corrections. The name "America" entered use through Waldseemüller's 1507 world map, honouring Amerigo Vespucci — one printed sheet fixing a continental name permanently.
The Dutch dominated the seventeenth century. Abraham Ortelius published the Theatrum Orbis Terrarum in 1570, generally regarded as the first modern atlas because it gathered uniformly sized maps with consistent presentation and credited its sources. Blaeu and Hondius built substantial businesses on the trade, and accurate charts became state secrets — Portugal and Spain both treated route information as classified.
Mercator's compromise and the projection problem
Gerardus Mercator published his world map in 1569 to solve a specific navigational problem: on a sphere, a course of constant compass bearing is a curve, which made chart plotting awkward. Mercator's projection renders any such line as a straight one, which is exactly what a navigator needs.
The price is area distortion that grows toward the poles. Greenland appears comparable in size to Africa, when Africa is roughly fourteen times larger. Mercator understood this perfectly well — the distortion is the unavoidable cost of the property he wanted.
This is the central truth of the whole field: no flat map can preserve area, shape, distance and direction simultaneously. Every projection sacrifices something, and the honest question is not which projection is correct but which distortion is acceptable for a given purpose.
- Mercator — preserves angles and bearings; badly distorts area at high latitudes
- Gall-Peters — preserves area; distorts shape, stretching equatorial regions vertically
- Robinson — preserves nothing exactly, compromising across all properties for general-purpose appearance
- Winkel Tripel — minimises combined distortion; adopted by the National Geographic Society in 1998
Our explainer on map projections works through the geometry, and the piece on Mercator versus Peters covers the political argument that grew up around this choice in the twentieth century.
Triangulation and the age of national surveys
The eighteenth century turned mapmaking from compilation into measurement. The method was triangulation: measure one baseline with extreme care, then determine every other point by angle alone, propagating a network of triangles across a country.
The Cassini family surveyed France across four generations, producing the first national map of an entire country founded on systematic triangulation. The Ordnance Survey of Great Britain began in 1791, initially for military purposes, and the Great Trigonometrical Survey of India ran from 1802 for decades — the effort that eventually established the height of the peak later named Everest.
Longitude remained the hard problem. Latitude is straightforward from the sun or Polaris, but longitude requires knowing the time at a reference meridian. John Harrison's marine chronometers, developed over the middle decades of the eighteenth century, finally made this practical at sea. Standardising the reference took longer: the International Meridian Conference of 1884 adopted Greenwich, largely because most existing charts already used it.
Aerial photography arrived with the First World War and transformed the work again, allowing terrain to be captured wholesale rather than point by point.
Thematic mapping arrives
The same century produced a second innovation, less discussed but arguably as consequential: maps that showed data rather than terrain. John Snow's 1854 map of cholera deaths in Soho, plotting fatalities against water pumps, is the canonical example, and Charles Minard's 1869 chart of Napoleon's Russian campaign compressed army strength, geography, direction and temperature into a single figure. Once a map could carry a variable instead of a coastline, cartography became an analytical tool rather than only a record of where things are.
Satellites, GPS and the map that never finishes
The modern era began with orbital imagery and satellite positioning. GPS reached full operational capability in the mid-1990s, and the deliberate degradation applied to civilian signals was switched off in 2000, which immediately improved public accuracy roughly tenfold and made consumer navigation viable.
Two consequences followed that would have been unrecognisable to any earlier cartographer. First, maps stopped being editions. A printed atlas was fixed at publication; a digital map is a database queried in real time, updated continuously, and different for every user depending on zoom, layer and location. Second, mapping became participatory. OpenStreetMap, launched in 2004, is built by volunteers and has repeatedly provided the best available data in places commercial providers had not covered — most visibly in the days after the 2010 Haiti earthquake.
What has not changed is the underlying tension. Every map still selects, simplifies and distorts, and every one still encodes decisions about what matters. Contested borders are drawn differently depending on which country a user is in, which is Ptolemy's problem in a modern form. To follow the story further, the cartography timeline lays out the dates in order and the profiles of the cartographers themselves fill in the people behind them.