Plagues and Peoples by William H. McNeill Page B

civilized conditions. Until very recently cities were unable to maintain their numbers without a substantial inflow of migrants from surrounding countrysides. Urban health hazards were simply too great, for, in addition to infectious person-to-person diseases transmitted as childhood diseases usually are—by breathing in droplets of infectious matter sneezed or coughed into the atmosphere—ancient cities suffered from an intensified circulation of diseases transmitted through contaminated water supplies, plus a full array of insect-borne infections. Any breakdown of transportation bringing food from afar threatened famine, and local crop failures were often difficult to compensate for. In view of all this it is not surprising that cities could not maintain themselves demographically, but had to depend on migrants from the countryside to replenish the losses arising from famine, epidemic, and endemic diseases.
    A civilized pattern of life therefore required rural cultivators not only to produce more food than they themselves consumed in order to feed urban dwellers, but also to produce a surplus of children whose migration into town was needed to sustain urban numbers. Rural reproductive surpluses had also to be capable of bearing losses resulting from macroparasitism, i.e., from war and raiding, and from the famine such activities nearly always provoked. Only occasionally and for limited periods of time was anything like a stable balance attained between rural birth rates and occupational niches available in urban contexts for the surplus from the countryside. Open and accessible frontiers—so important for European history in the past four centuries—were also unusual, though when land was available, surplus rural population could and did migrate to the frontier and thereby enlarge the agricultural base of the society instead of trying the risky (though to a few, spectacularly rewarding) path of migration into town.
    Until after 1650, when population statistics begin to assume a degree of reliability, it seems impossible even to guess at the magnitudes involved in this pattern of population flow. Nevertheless, such patterns clearly asserted themselves from the time cities first formed. The striking way, for example, in which Sumerian-speakers gave way to Semitic-speakers in ancient Mesopotamia during the third millennium B.C . 39 is probably a direct consequence of this kind of population movement. Speakers of Semitic tongues presumably migrated into Sumerian cities in such numbers that they swamped speakers of the older language. Sumerian lingered on as a language of learning and priestcraft, but for everyday purposes, the Semitic Akkadian took over. This linguistic shift might have resulted from a spurt of urban growth, or more likely from an unusually heavy die-off of established urban populations because of disease, war or famine, although which of these factors or combination of factors may have been at work in ancient Sumer is not known.
    A nineteenth-century parallel may be useful. From the 1830s and especially after 1850, rapid urban growth togetherwith the ravages of a new disease, cholera, disrupted cultural patterns of long standing in the Hapsburg monarchy. 40 Peasant migrants into the towns of Bohemia and Hungary had long been accustomed to learn German, and in a few generations their descendants became German in sentiment as well as in language. This process began to falter in the nineteenth century. When the number of Slav- and Magyar-speaking migrants living in the cities of the monarchy passed a certain point, newcomers no longer had to learn German for everyday life. Presently nationalist ideals took root and made a German identity seem unpatriotic. The result was that Prague became a Czech- and Budapest a Magyar-speaking city within half a century.
    Early civilizations that were linguistically more uniform obviously did not register the process of migration into town by linguistic change as ancient