The core pattern: At 5000 BCE, both China and Europe were multi-polar, with farming dispersed across multiple distinct regional cultures. What happened next diverged sharply. In China, the Yangshao/Longshan expansion progressively absorbed neighboring cultures into a single dominant Yellow River core. In Europe, no analogous convergence occurred: farming populations remained distributed across separated coastal and riverine zones bounded by mountains, seas, and dense forest. By 1 AD, both empires governed ~60 million people, but China's population was concentrated in one overwhelming core while Rome's was dispersed across six centers of roughly equal weight. By 1000 AD, the gap had widened: Song China had ~100M on one linked axis; Europe had ~38.5M scattered across seven regions. This geographic structure maps directly onto state formation outcomes.
I. Convergence vs. Persistent Dispersal
The conventional framing of Chinese geographic unity vs. European fragmentation can mislead if applied too early. At 5000 BCE, China was at least as multi-polar as Europe. K.C. Chang's Map 1.1 (Cambridge History of Ancient China, p. 49) identifies seven or more distinct regional cultures: Yangshao in the Wei valley, Dawenkou in Shandong, Xinle in Manchuria, Daxi in the middle Yangtze, Majiabang and Hemudu in the lower Yangtze, and Dalongtan/Dapenkeng in the far south. Europe at the same date had three to four separated farming zones (Balkans, the Danube/LBK corridor, the Mediterranean coast, and emerging Iberian settlement). If anything, China appeared more dispersed.
The critical difference is what happened between 5000 and 1000 BCE. In China, one zone (Yangshao → Longshan → Erlitou → Shang → Zhou) became overwhelmingly dominant and progressively absorbed or displaced the others. By the Han census of 2 AD, this single core on the Yellow River plain held the vast majority of a 60-million population. The other Neolithic cultures either merged into the expanding core, retreated to peripheries, or contributed ancestry that was subsumed. This is a process of convergence: many centers collapsing into one.
In Europe, no analogous process occurred. The Balkans, Italy, Iberia, Gaul, the Rhine zone, and Britain all persisted as roughly co-equal population centers across every subsequent period. The Mediterranean, the Alps, the Pyrenees, and the English Channel kept them separated. European farming itself hugged coasts and river valleys for millennia; interior Germany was heavily forested until the High Medieval Rodung clearing. The three-field system that transformed northern European agriculture after ~800 AD actually created additional population centers (in the Paris basin, the Rhineland, and southern England) rather than consolidating existing ones. Rome temporarily governed all these regions, but the demographic structure underneath remained multi-polar, and when Rome fell, fragmentation reasserted itself immediately.
By the Han-Rome comparison (~1 AD), the contrast is sharp and precisely measurable. Both empires governed ~60 million people across comparable territory. The Han census heatmap shows density exceeding 200/km² in the Henan/Shandong core with near-zero settlement south of the Yangtze: one overwhelming center. Scheidel's provincial estimates for Rome show six major population centers (Italy ~7M, Greece/Anatolia ~12.5M, Egypt ~5M, Iberia ~6M, Gaul ~6M, N. Africa ~6M), none holding more than a quarter of the total, all separated by water or mountains. By 1000 AD the gap had widened further: Song China had 2.5 times Europe's population, concentrated on a single Grand Canal-linked axis, governed by one fiscal authority extracting 10–17% of GDP.
II. Reference Maps
Data download: The individual site dots on this map are sourced from the spreadsheet below. Every dot corresponds to a named, excavated site with a scholarly citation. Coordinates are geocoded from site names (~1–10 km precision). Periods 5–6 (1 AD, 1000 AD) use geocoded administrative/urban data from CHGIS V6 (Han and Song county seats), Pleiades Gazetteer (Roman-era cities with formal status), and Buringh 2021 (European urban population). See farming-periods-5-6.xlsx for full methodology and selection criteria.
farming-sites.csv ·
farming-sites.xlsx ·
farming-periods-5-6.xlsx (full source documentation)
Major regional cultures in China around 5000 BC. K.C. Chang, Cambridge History of Ancient China (Loewe & Shaughnessy eds., 1999), Map 1.1, p. 49. Seven or more distinct cultures: Yangshao, Dawenkou, Xinle, Daxi, Majiabang, Hemudu, Dalongtan/Dapenkeng. China at this date was at least as multi-polar as Europe. The question is what happened next.
Expansion of farming in Western Eurasia, 9600–4000 BCE. Gronenborn, Horejs, Börner & Ober (2023), CC BY 4.0. Note the two separate routes (Danube inland, Mediterranean coastal) and the vast forested gaps between them.
The spread of farming in East Asia. Phillips Atlas of World History (p. 19). Two cores: millet (amber) in the Yellow River, rice (green) in the Yangtze. Arrows show expansion east to Shandong, south, and eventually to Korea and Japan.
Spread of farming in Europe, 7000–3500 BC. Phillips Atlas (p. 20). Color bands show the chronological wave front, with hatched areas marking persistent hunter-gatherer regions (Scandinavia, Baltic).
Population density by commandery, Han census of 2 AD. Darker = denser. The concentration in the Yellow River/central plain (Henan, Shandong) is overwhelming, with near-zero density south of the Yangtze. Source: Reddit visualization from Han census data.
Population distribution under the Former Han (top left), Tang (bottom left), and Northern Song (right). Mark Elvin, Cultural Atlas. The northward-to-southward shift of the demographic center of gravity across a millennium.
Medieval European population estimates (millions), 500–1450 AD. Josiah Russell (1958), via Fordham Medieval Sourcebook. At 1000 AD: France 6M, Italy 5M, Iberia 7M, Germany 4M, Balkans 5M, Britain 2M. No single region dominates. Caveat: pre-18th century European population estimates are speculative; all figures should be treated as rough orders of magnitude.
III. Sources & Methodology
Farming zones on the interactive map are approximate representations based on the sources below. Circle sizes represent relative population/demographic importance at each period, not precise boundaries. Crop type assignments follow the dominant cereal regime; most zones grew multiple crops. The comparison panel figures for 1 AD and 1000 AD draw on actual census data (China) and scholarly estimates (Europe) as noted. For periods 5 (1 AD) and 6 (1000 AD), individual site dots are drawn from geocoded administrative gazetteers: CHGIS V6 for Chinese county seats, Pleiades for Roman-era cities, and Buringh (2021) for medieval European urban populations. The asymmetry in dot counts between China and Europe at 1 AD (1,142 vs 529) reflects both a genuine difference in administrative granularity and uneven Pleiades description coverage. See farming-periods-5-6.xlsx for full selection criteria.
Bairoch, Paul, Jean Batou, and Pierre Chèvre. La population des villes européennes, 800–1850. Geneva: Droz, 1988.
Bellwood, Peter. First Farmers: The Origins of Agricultural Societies. Oxford: Blackwell, 2005. Chs. 3 (SW Asia), 4 (Europe), 6 (East Asia): site distributions and crop dispersal maps.
Chang, K.C. "China on the Eve of the Historical Period." In The Cambridge History of Ancient China, eds. M. Loewe and E. Shaughnessy. Cambridge: Cambridge University Press, 1999. Map 1.1, p. 49 (regional cultures c. 5000 BCE).
Ge Jianxiong 葛剑雄 et al. Zhongguo yimin shi (Migration History of China). Fuzhou: Fujian People's Publishing House, 1997. [Population distribution data by dynasty.]
Gronenborn, Detlef, Barbara Horejs, Lutz Börner, and Johanna Ober. "Expansion of Farming in Western Eurasia, 9600–4000 BCE." Leibniz-Zentrum für Archäologie / Österreichisches Archäologisches Institut, 2023. CC BY 4.0.
Liu, Li. The Chinese Neolithic: Trajectories to Early States. Cambridge: Cambridge University Press, 2004. Chs. 6–7: settlement distribution maps for Peiligang, Yangshao, and Longshan periods in Henan, Shandong, and the Wei River valley.
Long, Tengwen et al. "The Early History of Wheat in China from ¹⁴C Dating and Bayesian Chronological Modelling." Nature Plants 4 (2018): 272–279. [Wheat appeared in lower Yellow River ~2600 BCE as an exotic good via steppe networks, not as a subsistence crop.]
McEvedy, Colin, and Richard Jones. Atlas of World Population History. London: Penguin, 1978.
Phillips, Charles (gen. ed.). Phillips Atlas of World History. Various contributors. London: Philips, 2005.
Russell, Josiah C. "Late Ancient and Medieval Population." Transactions of the American Philosophical Society 48.3 (1958): 1–152. [European regional estimates 500–1450 AD. Generally the low end of the scholarly range.]
Scheidel, Walter. "Demography." In The Cambridge Economic History of the Greco-Roman World, eds. Scheidel, Morris, and Saller. Cambridge: Cambridge University Press, 2007, pp. 38–86. [Roman provincial population estimates, Table 3.1.]
Elvin, Mark. The Pattern of the Chinese Past. Stanford: Stanford University Press, 1973. / Cultural Atlas of China (various editions).
Yang, Melinda A. et al. "Ancient DNA Indicates Human Population Shifts and Admixture in Northern and Southern China." Science 369 (2020): 282–288.
CHGIS, Version 6. "V6 Time Series County Points." Harvard Yenching Institute and Center for Historical Geography, Fudan University, 2017.
doi:10.7910/DVN/CCV43P. [10,522 county-level administrative seats with begin/end years, WGS84 coordinates.]
Pleiades: A Gazetteer of Past Places, Version 4.1. Institute for the Study of the Ancient World, New York University.
pleiades.stoa.org. [41,480 place resources for the ancient Mediterranean world. CC BY 3.0. Filtered here to 529 settlements with formal urban/administrative status.]
Buringh, Eltjo. "The Population of European Cities from 700 to 2000."
Research Data Journal for the Humanities and Social Sciences 6.1 (2021): 1–18.
doi:10.17026/dans-xzy-u62q. [2,262 European settlements, 700–2000. Updates and expands Bairoch et al. (1988).]