Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
c How Cities Grew in the Western World: A Systems Approach LYNN HOLLEN LEES AND PAUL M. HOHENBERG If there is a single defining characteristic of urban life, it is dependence. Not only are citizens interdependent, but the truly isolated city is both unviable and pointless. Unable to sustain itself, it would have no outlets for the fruits of specialization and complex organization. In the words of B. J. L. Berry, "Cities are systems within systems of cities" (1964, p. 1471. During the last millennium, as settlement in Europe became pre- dominantly urban, city systems of great complexity evolved as population grew. As conduits for goods, people, and information, they helped to distribute resources throughout vast territories. In fact, the very stuff of urban life is exchange, and the exchange process has generated extensive infrastructures to aid the distribution of commodities, ideas, and popu- lation. The complexity of urban systems poses a challenge to planners and urban scholars. How can the interactions of market towns, capitals, ports, resorts, and manufacturing cities as well as their functions be captured by an analytical model? Because of the varied quality of urban processes, no single schema will suffice. Drawing on the disciplines of economics, geography, and history, we propose that a dual model of interurban re- lationships captures many important distinctions and offers a tool through which changes over time and variations among city types can be explained. It is important to note that urban growth has been neither automatic nor uniform but has exhibited consistent patterns linked to structure and func- tion. Many of the dynamic influences on city size demographic change, migration, transportation, technology, and communications- have af 71
72 LYNN [IOLLE1V LEES AND PAUL M. HOHENBERG fected urban fortunes through processes referred to as the central place system and the network system (Hohenberg and Lees, 19851. A DUAL MODEL OF URBAN SYSTEMS: ECONOMIC EXCHANGE AND SPATIAL CONFIGURATIONS The notion of a town as a central place is well established (Berry and Pred, 1961; Christaller, 1933, Losch, 1954; Tinbergen, 19681. As a center for the exchange of goods and services, a town serves the needs of a farming area, for example, by offering an outlet for surplus production and by housing artisans as well as purveyors of cultural and political products. The central place concept can also be applied to a set of settle- ments. Because people need complex goods and services less frequently but are willing to travel longer distances for them, such items are offered only in larger settlements. Differing types and intensities of demand create a hierarchy of settlements in which larger towns serve territories also organized by smaller central places. Through a regular nesting and inter- dependence of centers, urbanization creates integrated regions that are more than the sum of their parts. Roads and railroads make visible the multiple linkages that tie a large array of settlements together. Nations and kingdoms can be visualized as groups of regions, beginning with the hinterlands of small markets and progressing through the territories di- rected by administrative centers to the major economic and political cap- itals, which in some countries coincide and in others do not. Medieval Leicester is a good example of a regional marketplace (Brown, 1970; Hoskins, 1955; Simmons, 19741. Located approximately in the center of the English county of Leicestershire, it served as the county's commercial and cultural capital. Fairs, schools, and charities regularly brought outsiders into the town, and city dwellers themselves provided goods and services to a wide area. Besides producing staples, the city's workers made substantial amounts of woolen cloth for sale in the region. Smaller market towns, roughly equidistant from Leicester, offered simpler goods and services to the surrounding villages, but the county's highest order central place was the only locus of political authority, as well as the site of courts and tax collections. Royal administrators working from London contacted officials in Leicester, who spread information through- out the region. Leicester also had external ties that cannot be explained through a central place model. Note the city's location at the place where a Roman road forded the river Soar. Leicester was one link in a chain of cities that stretched to London and were finally connected to central Italy. The principal activity of Leicester merchants in the later Middle Ages was the
HOW CITIES CREW I1V THE WESTERN WORLD 73 sale of raw wool to distant cloth makers, and the amounts of wool traded were far greater than what would have been necessary to buy indispensable imports or to pay the taxes and rents due the Earl of Leicester and the English king. Leicester therefore linked its region to a larger commercial network, which stretched from English east coast ports to the Low Coun- tries and from London to manufacturing towns throughout Europe. The example of Leicester shows how a city's trade may involve an exchange of goods with both local and distant destinations. Neither size nor position in a central place system is sufficient to predict the long- distance linkages of a settlement. Indeed, even small villages may engage in the specialized production of a traded staple, which an urban system then helps to distribute. The network of exchange thus generated is still hierarchic, but a regular geometric ordering of centers is replaced by a series of nodes, junctions, and outposts arranged according to a core- periphery logic (Figure 3-lb). Rather than a contiguous group of regions, often centrally administered, such urban systems comprise loosely linked cities that anchor trade routes. We call this type of linkage the network system (Hohenberg and Lees, 19851. Both Braudel (1979) and Waller- stein's (1974) visions of a world economy are consonant with this model. In a network system, maritime empires and urban leagues rival one an- other. Their wealth and influence depend not on the control of the territory but on access to long-distance transportation, luxury goods, information, and human capital. Few places in early Europe illustrate this alternative urban system better than Venice, the Queen of the Adriatic. Venice began on an unpromising site as a small gathering of boatmen who produced fish and salt (Lane, 1973; Luzzatto, 19614. Building on its ties to Constantinople, the city became the trading gateway between a rapidly developing European periphery and a thriving eastern economy that stretched through Arab and Byzantine lands to China. In time, mer- cantile wealth and naval power reinforced one another to undermine the dominance of the Middle East and to create a Venetian maritime empire. Despite meager resources and a limited population, Venice was able to preserve its autonomy, if not its influence, until the early nineteenth century. In spatial terms the central place system is territorial and geometric. Distances between cities of a given size or rank are more or less uniform, as is the relation of city size to the dimensions of the territory served (Figure 3-la). Topographic irregularities of course modify this mosaic, but for the most part the spacing of the cities in central place systems is regular. By contrast the network system consists of linkages dictated by profit and facilitated by water transport, which blunt the effects of distance. Even in the transfer of ideas, a case in which bulk is not an issue, distance
74 LYNN HOLLEN LEES AND PAUL M. HOHENBERG O Major center O City in question /' X X / O Lesser city ~X X Small town Coastline - - a. Central Place System core periphery - hinterland ,/lx ~ 1: friend b. Network System to j to\-I\x X ~ X /' umland ~ FIGURE 3-1 Patterns of interaction for cities in two different urban systems: central place system (3-la) and network system (3-lb). From Hohenberg and Lees (1985~. Reprinted by permission. proves a greater barrier to exchange in the central place system than in the network system. In the latter, principal centers may well be clustered, as in northern Italy and the Low Countries during the Middle Ages or along the East Coast in contemporary America; or they may surround an inland sea, each serving as a gateway to a region of the interior, as Braudel (1966) so vividly illustrated for the case of the Mediterranean. Indeed, the clustering of major centers in wealthy areas, compared to their wider spacing in many agriculturally rich but less economically active territories, is a phenomenon that the central place model cannot explain. Because
HOW CITIES GREW IN THE WESTERN WORLD 75 this model implies that each large city commands a region in which internal exchange is more intense than external trade, it deals poorly with the active, long-distance exchanges of people and goods directed through highly urbanized areas. An urban system does not require geographic compactness or symmetry. Instead, transportation linkages can overcome the barrier of distance, and infrastructures can be created to support asymmetrical flows of commod- ities and population. The huge port areas of Rotterdam and Hamburg or the railway yards in Essen were built not for the economic exchanges of a central place region but for an international trading system. THE DYNAMICS OF GROWTH AND CHANGE Long waves of population growth provided the human material for city building. Although statistical data for medieval and early modern times are often incomplete and unreliable, the combination of tax records, house- hold listings, demographic registers, and censuses yields plausible esti- mates for the sizes of major cities and for overall patterns of growth and decline (Bairoch, 1985; de Vries, 1984; Russell, 19721. Demographers estimate that the European population rose steadily from around the year 1000 to 1350, after which it declined. Increases predominated from 1450 to 1600, and accelerated growth resumed well after 1700. Tabulations for the period 1600-1700 are disputed, but regional variations probably add up to a slight decline in overall population. Changes in the sizes of Eu- ropean regional capitals illustrate both the timing and the magnitude of growth (Figure 3-24. The demographic engine produced in Europe three periods when increasing numbers of people had to be accommodated by social and economic systems and by urban infrastructures. The first and third witnessed large jumps in the urban share of population; the second was marked by shifts within a more static urban sector. Historical patterns of urbanization reflect the interaction of geography and economic development with the availability of human capital. The first, post-Roman wave of city creation rested on the secure base of rising agricultural productivity. At a time of relatively low population density and weak political centralization, the most common urban type was the market town in which the predominantly rural population could sell its surplus and buy locally manufactured goods. Central place systems of cities fed by short-distance and step-by-step migration grew up in most parts of western and southern Europe, as well as England. Long-distance trade also encouraged urban growth, primarily along waterways and sea- coasts. Between the years 1000 and 1350, thousands of settlements; ac- quired the social, economic, and legal characteristics of cities, becoming
76 LYNN HOLLEN LEES AND PA UL M. HOHENBERG Population (log scale) 2,000,000 1,000,000 1 00,000 ,~': ,, ~ Milan /Lyons / '/ / ~__ / r Cologne 10,000 ~, , Date 1000 1200 1400 1600 1800 1950 Figure 3-2 Growth of provincial capitals, 100~1950. Data from Chandler and Fox (19741. From Hohenberg and Lees (19851. Reprinted by permission. important forces in an essentially rural, sometimes feudal environment. City creation and the rapid buildup of central place urban networks were the order of the day. Castle towns and settlements with newly granted market rights multiplied all the way from Ireland through eastern Europe. The next phase of urban development, between 1450 and 1650, was markedly different. Population growth remained relatively slow, hindered by plague, war, and famine. Economic competition undermined the pros- perity of Mediterranean towns; empires in the Americas brought both costs and benefits to cities along the Atlantic coast (Hohenberg and Lees, 19851. The result was a redistribution of the urban population among regions and types of towns (de Vries, 19841. Few new cities were founded, and overall rates of urban growth slowed. Revival from the demographic losses of the fourteenth century was concentrated among the larger market towns and regional capitals. Hundreds of medieval towns, particularly in southern
HOW CITIES GREW IN THE WESTEW WORLD 77 France and central Europe, disappeared or became villages once more. But capital cities and ports grew explosively. Short-term declines in the urban populations of Italy, Spain, and the Low Countries were counter- balanced by the rising urbanization of England and Scandinavia. In the sixteenth century, a northern core of towns displaced the older southern cities, and rivalries among Antwerp, Amsterdam, and London replaced earlier Italian competitions for influence (Braudel, 19791. During this period when Europe asserted its leadership over an enlarged world economy, network system linkages brought more growth than did local trade. Nevertheless, the fastest growth of city populations at this time occurred in the capitals and court cities of dynastic states, the cap- stones of central place systems (Table 3-11. The political centralization of kingdoms, with the associated power to levy increased taxes and rents, brought growth to the top levels of urban hierarchies at the expense of smaller towns. In several cases for example, France, Spain, and southern Italy centralization and exploitation led to primacy in urban size distri- butions as oversized capitals (Paris, Madrid, Naples) dominated an oth- erwise static urban ensemble. These early metropolises drew resources from the rest of the kingdom without stimulating economic growth else- where (Ringrose, 19831. London, which was also England's principal seaport and manufacturing town, is one of the few exceptions to the hypothesis that, at this time, great size and rapid expansion in a political capital were a sign of poor rather than good economic health in the urban system as a whole. Two dynamics of growth, each characteristic of a different urban sys- tem, can be observed at this time. Network system centers grew on the basis of specialized activities, often that of international trade. As a result, they tended to stimulate development at sources of supply, for example, in colonies or other distant places (Wallerstein, 19741. In urban distri- butions, therefore, network systems encouraged the development of new cities in peripheral territories as well as the development of specialized places, such as mining towns, resorts, and ports. Growth rates were in- versely related to size and tended to level off among the larger centers. But in the older central place systems of western and southern Europe, where new towns rarely arose, increasing centralization resulted in primacy with growth rates biased toward higher order urban places (Hohenberg and Lees, 19851. Urban growth after 1750, during the long phase of population increase and industrialization that continues today, has also been channeled through both network and central place systems. Most strikingly, new cities de- veloped in large numbers for the first time since the Middle Ages as settlers were attracted to the sites of industrial production. In particular,
78 _` Cal Ct U. o o ._ As o _' to to or to to o U) ._ ._ U. sit A i_ 1 m ~ i 8 ;^ t_ . _ =8 Q~ o of C: o o ~ ~ ox ~ Go o ~ o oo ~ ~ ~ ~ ~ ~ ~ - o -)~______ "D ~ ce ~ C~ ~ ' ~ ~ E 3 3 ~ U: ~ ~ o o ~ oo ~ ~ ~ o o oo ~ ~ ~ ~ C~ C~ ~U~_______ Ct ,_~ ~ zCL-co E E ~ o ~ ~ o 0 0 ~ o 0 0 1- C-l C~ - O O ~ r~ ~ ~ ~ _ _ _ _ _ Ct ,,_ ~ CC-= C O tdo ~- C O O O ~ ~ ') O O w) 0= ~ ~ e ~ O oo `_ C~ ~o .= . ~ _` _' X S~ C: .. C: O
HOW CITIES GREW IN THE WESTERN WORLD 79 when large coal deposits were opened up and heavy industry gathered near the pitheads, explosive urbanization of formerly rural areas resulted. New linkages among producers, exporters, and consumers were created to handle the increasing volume of production and trade. The Ruhr, with its rapidly expanding pattern of canals, railroads, roads, and towns, is the classic example, but England's Black Country and the Belgian Borinage should also be kept in mind (Reulecke, 1984; Steinberg, 19671. In the territory around originally small settlements, urbanization disregarded cen- tral place patterns of market centers and instead engulfed and absorbed villages and small towns alike. Many of these new "cities" lacked the urban services and functions common to central places. Financial and other services came from outside, and the internal cultural and adminis- trative links were forged later, if at all. Such industrial conurbations developed in ways quite foreign to the central place dynamic but readily accounted for in the network system model. Specialized activity in an isolated location implies strong ties with more distant locations because industrial settlements remained dependent for all essentials: food, finance, markets, and even people. The steel furnaces of the Ruhr were fed by iron from Lorraine and Asturias; yet they also drew on Polish workers, Russian wheat, and Hungarian cattle. Meanwhile, the region's larger banks moved to Berlin. Proximity could not preserve the influence of the older Rhenish financial centers over the Ruhr's industry (Reulecke, 1984; Steinberg, 19671. Although network linkages fostered the early stages of industrial ur- banization, a resurgence of the economic power of the larger, older central places can be seen during the late nineteenth and early twentieth centuries when technological processes made it both more feasible and more at- tractive to concentrate industrial activities in existing cities. Railways linked the bigger towns, and electric power transmission severed the lo- cational pull of the pithead. Ready access to labor and markets soon outweighed the lure of cheap land and raw materials. In Europe after 1880 the result was an acceleration in the centralizing tendency of central place systems. Growth tended to be concentrated in and on the outskirts of major capitals and conurbations, and rapid migration from rural areas to major cities accentuated the imbalance. Industrial activities congregated in ports and national centers. But the expansion of service employment ranging from retail trade and construction to intellectual, artistic, and scientific pursuits alongside finance and administration accounted for much of the growth. The largest cities by far were those that combined a multiplicity of functions. If a list of the 40 largest cities in Europe in 1750 is compared with similar lists for 1850 and 1950, the continued dominance of political
80 LYNN HOLLEN LEES AND PA UL M. HOHEI!iBERG capitals is striking. A few cities from the Mediterranean area Venice, Palermo, and Granada-disappear from the lists, and a handful of in- dustrial cities, such as Manchester, Birmingham, and Katowice, become urban giants. For the most part, however, the top ranks of the European urban system continued to be occupied by political capitals, which com- bined service and production functions with their administrative powers (see Table 3-21. This numerical predominance of political capitals highlights the most important centripetal influence on urban systems, the increasing role of information as opposed to raw labor and materials. Over time, economies of agglomeration outweighed economies of scale and vertical integration, so crucial to heavy industrial complexes. The proximity of government offices, suppliers, competitors, and financiers has drawn entrepreneurs into the major towns. The concentration of people and activities in large and very large cit- ies the growth of the threatening and imposing metropolis could pro- ceed only so far, however, before cost and congestion imposed a reversal in the trend. In the twentieth century, rising land prices in combination with the decentralizing potential of the internal combustion engine and the mass ownership of automobiles have permitted the dispersion of routine production tasks. In the longer run, chemical factories shifted away from Paris (Gaillard, 1977), and docks in central London were abandoned for sites downriver. Nevertheless, the long-run choices of entrepreneurs for the siting of companies are only partially explained by the direct costs of production. The extraordinary flow of new English industry into the Lon- don region, despite its higher wages, rents, and land prices, illustrates the pull of the metropolis for both managers and workers (Hall, 19661. And although the computer permits easy dispersion of personnel, the attraction of both Silicon Valley and Boston's Route 128 for new entrants in the industry shows economies of agglomeration at work in the United States. IMPLICATIONS FOR THE DEVELOPMENT OF INFRASTRUCTURES Urban sytems are far more than physical structures. To be sure, their roads, airports, and telephone lines have a physical existence. But re- member that cities with six-digit populations long antedated railroads and turnpikes. Cities were magnets for people and ideas before modern tech- nology developed elaborate infrastructures to link them. Feet brought new citizens from mountain villages to towns in the plain, and ships linked Greek coastal city-states to counterparts throughout the Mediterranean. Studies of interurban systems need to include the intangible links be- tween central places, ties that shape the formation of physical structures.
HOW CITIES GREW IN THE WESTERN WORLD TABLE 3-2 Leading Cities of the European Hierarchy in 1750, 1850, and 1950 (population in thousands) 81 1750 1850 1950 Rank City Pop. City Pop. City Pop. 1 London 676 London 2,320 London 8,860 2 Paris 560 Paris 1,310 Paris 5,900 3 Naples 324 St. Petersburg 502 Moscow 5,100 4 Amsterdam 219 Berlin 446 Ruhr 4,900 5 Lisbon 213 Vienna 426 Berlin 3,707 6 Vienna 169 Liverpool 422 Leningrad 2,700 7 Moscow 161 Naples 416 Manchester 2,382 8 Venice 158 Manchester 412 Birmingham 2,196 9 Rome 157 Moscow 373 Vienna 1,755 10 St. Petersburg 138 Glasgow 346 Rome 1,655 11 Dublin 125 Birmingham 294 Hamburg 1,580 12 Palermo 124 Dublin 263 Madrid 1,527 13 Madrid 123 Madrid 263 Budapest 1,500 14 Milan 123 Lisbon 257 Barcelona 1,425 15 Lyons 115 Lyons 254 Milan 1,400 16 Berlin 113 Amsterdam 225 Glasgow 1,320 17 Hamburg 90 Brussels 208 Liverpool 1,260 18 Marseille 88 Edinburgh 194 Naples 1,210 19 Rouen 88 Hamburg 193 Leeds 1,164 20 Copenhagen 79 Marseille 193 Copenhagen 1,150 21 Florence 74 Milan 193 Athens 1,140 22 Genoa 72 Leeds 184 Bucharest 1,100 23 Granada 70 Palermo 182 Katowice 977 24 Barcelona 70 Rome 170 Brussels 964 25 Seville 68 Barcelona 167 Amsterdam 940 26 Bologna 66 Warsaw 163 Prague 938 27 Bordeaux 64 Budapest 156 Stockholm 889 28 Turin 60 Bristol 150 Lisbon 885 29 Valencia 60 Sheffield 143 Munich 870 30 Cadiz 60 Bordeaux 142 Newcastle 830 31 Stockholm 60 Venice 141 Rotterdam 803 32 Dresden 60 Turin 138 Warsaw 803 33 Prague 58 Copenhagen 135 Kiev 800 34 Brussels 55 Munich 125 Kharkov 730 35 Edinburgh 55 Prague 117 Sheffield 730 36 Lille 54 Breslau 114 Turin 725 37 Cork 53 Wolverhampton 112 Cologne 692 38 Breslau 52 Newcastle 111 Frankfurt 680 39 Koenigsberg 52 Valencia 110 Genoa 676 40 Leiden 50 Ghent 108 Lodz 675 SOURCE: Chandler and Fox (1974).
82 LYNN HOLLEN LEES AND PA UL M. HOHENBERG For example, one basic urban service is the provision of information. Through resident experts, schools, newspapers, and other institutions, cities provide information to their hinterlands, and critical linkages are forged by the movement of people and paper along regular channels of communication. Moreover, innovations are diffused by means of urban elites and institutions. New ideas and inventions speed across national boundaries through the channels of the network system and then are adopted by people and institutions connected in regional urban hierarchies. Power plants, telephone exchanges, and printing presses have tended to appear early in capital or very large cities and then to diffuse downward through central place systems. Although these technologies generated a tangible infrastructure, they only gave form to preexisting ties. In the twentieth century, the growth of a "wired society" permits the decentralization of many management and service activities; yet paradoxically, these activities are just the functions that cluster in metropolitan areas. Large-scale man- ufacturing plants, which thrive on cheap land and easy transportation, are much more likely to decentralize than management or finance, which use the amenities of the city to attract highly skilled workers and use its roads, mass transit, schools, and information services to link employees in a high-density environment. Managers have realized that creativity is fos- tered by the propinquities of the city. The ability of urban populations to innovate is also shaped by the systemic, political ties of towns outside their borders. For centuries, pat- terns of urban governance have varied with the type of interurban con- nection. Whereas central places are economically somewhat insular, deriving their principal activities from local service functions, they have been politically dependent on an outside authority. Central places, after all, administer the area around them. Except for national capitals, this function is delegated from the top of the urban hierarchy. By contrast, the network system forges close economic ties among cities often geographically distant and functionally dissimilar with no necessary political ties. Ideally, the network city is politically autonomous because governmental central- ization inhibits the fluidity of those interurban ties on which network cities thrive. The most economically successful trading cities of the past places such as Venice, Hamburg, and Amsterdam remained on the peripheries of European nation-states during their periods of dominance (Rokkan, 19751. This contrast between the political dependence of central place towns and the relative independence of network centers has shaped the financing of municipal government as well as investment in major improvements. Central places, other than national capitals, have limited resources of their own. They act as conduits for revenues and expenditures controlled by
HOW CITIES GREW IN THE WESTERN WORLD 83 the territorial administration. Such cities must compete, often unsuccess- fully, for a share of national resources. Capitals, whose buildings sym- bolize both a state and a particular regime, are best situated to profit from this competition. There, the desire for glory or the need to maintain public order in the face of crisis often results in physical improvements. This is an odd but continuing story. Compare the levels of governmental spending on streets, water supplies, and public buildings in nineteenth-century Paris or seventeenth-century Madrid with those of smaller towns in France or Spain (Evenson, 1969; Ringrose, 19831. In contrast, network cities, al- though typically oligarchic rather than democratic entities, have had greater control of their own resources as well as a lively sense of the value of the investment in both collective and individual ventures. Local funds built Amsterdam's canals, modernized Liverpool's port, and added parks to central Milan (Dalmosso, 1971; Konvitz, 1985; Vigier, 19701. Whereas formerly independent German trading cities have maintained traditions of large-scale public investment in services and municipal buildings, the level of such expenditures in France and Spain, countries where even network cities have been kept under relatively strong central control, has been low. Urban resources and the infrastructures they generate vary with the nature of a city's exchange relations. Indeed, the adoption of new tech- nologies depends heavily on networks of communication and finance, which are generated by a city's regional and international position. The analysis of infrastructures therefore requires examination of urban systems as they have evolved through time and as they persist today. REFERENCES Bairoch, P. 1985. De Jericho a Mexico. Villes et economic dans l'histoire. Paris: Gallimard. Berry, B. J. L. 1964. Cities as systems within systems of cities. Papers and Proceedings of the Regional Science Association 13: 147- 163. Berry, B. J. L., and A. Pred. 1961. Central Place Studies: A Bibliography of Theory and Application. Philadelphia: University of Pennsylvania Regional Science Research Insti- tute. Braudel, F. 1966. La Mediterranee et le monde mediterraneen a ltepoque de Philippe II. Deuxieme edition. Paris: A. Colin. Braudel, F. 1979. Civilisation Materielle, Economic et Capitalisme. 3 vols. Paris: A. Colin. Brown, A. E., ed. 1970. The Growth of Leicester. Leicester, England: Leicester University Press. Chandler, T., and G. Fox. 1974. 3000 Years of Urban Growth. New York: Academic Press. Christaller, W. 1933. Central Places in Southern Germany, trans. C. W. Baskin. Reprint. Englewood Cliffs, N.J.: Prentice-Hall, 1966. Dalmosso, E. 1971. Milan, Capitale Economique de l'Italie. Ph.D. dissertation. Universite d'Aix-Marseille.
84 LYNN HOLLEN LEES AND PAUL M. HOHENBERG de Vries, J. 1984. European Urbanization, 1500-1800. Cambridge, Mass.: Harvard Uni- versity Press. Evenson, N. 1969. Paris: A Century of Change, 1878-1978. New Haven, Conn.: Yale University Press. Gaillard, J. 1977. Paris, la ville, 1852-1870. Paris: Champion. Hall, P. 1966. The World Cities. New York: McGraw-Hill. Hohenberg, P. M., and L. H. Lees. 1985. The Making of Urban Europe, 1000-1950. Cambridge, Mass.: Harvard University Press. Hoskins, W. G. 1955. An Elizabethan provincial town: Leicester. Pp. 33-67 in Studies in Social History: A Tribute to G. M. Trevelyan, J. H. Plumb, ed. London: Longmans, Green. Konvitz, J. W. 1985. The Urban Millennium: The City-Building Process from the Early Middle Ages to the Present. Carbondale: Southern Illinois University Press. Lane, F. C. 1973. Venice: A Maritime Republic. Baltimore, Md.: Johns Hopkins University Press. Losch, A. 1954. The Economics of Location. New Haven, Conn.: Yale University Press. Luzzatto, G. 1961. Storia Economica de Venezia. Venice: Centro Internazionale delle Arti e del Costume. Reulecke, J. 1984. The Ruhr: Centralization versus decentralization in a region of cities. Pp. 403-430 in The Metropolis, 1890-1940, A. Sutcliffe, ed. Chicago: University of Chicago Press. Ringrose, D. R. 1983. Madrid and the Spanish Economy, 1560-1850. Berkeley: University of California Press. Rokkan, S. 197S. Dimensions of state formation and nation-building: A possible paradigm for research on variations within Europe. Pp. 601-638 in the Formation of National States in West;ern Europe, C. Tilly, ed. Princeton, N.J.: Princeton University Press. Russell, J. C. 1972. Medieval Regions and Their Cities. Bloomington: Indiana University Press. Simmons, J. 1974. Leicester Past and Present. Vol. 1, Ancient Borough to 1860. London: Methuen. Steinberg, H. G. 1967. Sozialraumliche Entwicklung und Gliederung des Ruhrgebietes. Bad Godesberg: Bundesanstalt fur Landeskunde und Raumforschung. Tinbergen, J. 1968. The hierarchy model of size distribution of centers. Papers and Pro- ceedings of the Regional Science Association 20:65-68. Vigier, F. 1970. Change and Apathy: Liverpool and Manchester During the Industrial Revolution. Cambridge, Mass.: MIT Press. Wallerstein, I. M. 1974. The Modern World System. New York: Academic Press.