Subject · Overview
Ancient civilisations
What we know, how we know it, and where the map is still blank.
Almost every argument about the oldest civilisation is really an argument about the word. Set the bar at cities, writing and bureaucracy and you get one answer. Set it at organised labour raising something nobody needed for shelter and you get a much older one. Neither is wrong. They are measuring different things, and most public disagreements never say which.
So this page does the boring thing first and says what the terms mean, then shows where the map is genuinely being redrawn.
Ground like this is where most of the argument happens. The shapes are natural. Telling those apart from the ones that are not is the whole job.
Brand imagery · generated, not survey data · Before the Record 2026
The definition does the arguing
The conventional checklist for a civilisation is some combination of cities, food surplus, craft specialisation, social hierarchy, monumental construction and record-keeping. On that checklist, southern Mesopotamia is the usual first entry, with Egypt, the Indus valley and northern China following. [UNVERIFIED: the precise ordering and dates, which vary by source]
Standard usage across archaeology and ancient history
That checklist was written before anyone knew about monumental construction thousands of years earlier, by people with none of the other boxes ticked. See Göbekli Tepe. The checklist has not really been rewritten; the sites have simply been filed as exceptions.
Pre-Pottery Neolithic sites of southwest Asia · UNESCO and excavation literature
What is argued: how much is still missing
Both cases at full strength.
The case that the map is badly incomplete.
Every time a new survey technology has been pointed at a region, the estimated scale of past settlement has gone up, not down. Lidar under jungle canopy has repeatedly turned landscapes assumed to be sparsely occupied into dense networks of structures and roads. Most of the deep past sits under forest, sand, farmland or water, and sea level was far lower for most of human history, so a great deal of the coastline people actually lived on is now offshore. On that reading, the current map is a map of where we have looked.
Attributed. The revisionist reading, examined not endorsed
The case that the map is roughly right.
Absence of evidence is weak evidence, but it is not nothing, and it accumulates. Metallurgy, agriculture and dense settlement leave durable and distinctive traces: pollen signatures, isotope shifts, slag, middens, genetic changes in domesticated species. Those traces are searched for globally and by methods that do not depend on anyone having chosen the right dig site. The new lidar findings, striking as they are, have revised the density and extent of known cultures rather than revealed unknown ones. Nothing so far requires a civilisation the record has entirely missed.
Attributed. The conservative reading, examined not endorsed
What would change our mind
A durable, dateable technological signature, metallurgical or agricultural, from a stratum that predates the earliest known culture capable of producing it, recovered and dated by a group that went looking for something else. That is a high bar deliberately. It is also exactly the bar the existing record cleared.
Where to go next
- Göbekli Tepe: monumental building before the checklist says it was possible.
- The Younger Dryas: the climate reversal sitting underneath the whole argument.
Every claim above carries a confidence label, and a source or an unverified mark. Read the standard.