1 Much of the impetus for quantification in historical research grew out of the “new data” movement of the 1960’s. See Fogel, Robert, ed., The Dimensions of Quantitative Research in History (Princeton: 1972)Google Scholar; and Lorwin, Val R. and Price, Jacob M., eds., The Dimensions of the Past (New Haven: 1972)Google Scholar. For additional comment and analysis see Zapf, Wolfgang and Flora, Peter, “Differences in Paths of Development: An Analysis for Ten Countries;” and Flora, “Historical Processes of Social Mobilization: Urbanization and Literacy, 1850-1965,” in Eisenstadt, S.N. and Rokkan, Stein, eds., Building States and Nations: Volume I (Beverly Hills: 1973), 161–211 and 213–58Google Scholar respectively.

2 See Flanigan, William H. and Fogelman, Edwin, “Patterns of Political Violence in Comparative Historical Perspective,” Comparative Politics, III (October 1971), 1–20Google Scholar; Graham, Hugh Davis and Gurr, Ted Robert, eds., Violence in America (New York: 1969)Google Scholar; Lodhi, Abdul Qaiyum and Tilly, Charles, “Urbanization, Crime and Collective Violence in 19th-century France,” American Journal of Sociology, LXX1X (September 1973), 296–318CrossRefGoogle Scholar; Snyder, David and Tilly, Charles, “Hardship and Collective Violence in France, 1830 to 1960,” American Sociological Review, XXXVII (October 1972), 520–32CrossRefGoogle Scholar; and Tilly, Richard, “Popular Disorders in Nineteenth Century Germany: A Preliminary Survey,” Journal of Social History, IV (Fall 1970), 1–40CrossRefGoogle Scholar.

3 The analytical utility of 19th-century censal and other data for quantitative historical research is discussed in Wrigley, E.A., ed., Nineteenth Century Society (Cambridge: 1972)CrossRefGoogle Scholar. See also Mitchell, Brian, European Historical Statistics, 1750-1970 (London: 1975)Google Scholar.

4 For an overview of the various theoretical and methodological issues involved in studying the spatial dynamics of crime, see Harries, Keith D., The Geography of Crime and Justice (New York: 1974).Google Scholar

5 A discussion of 19th-century views on crime with additional bibliography can be found in Mannheim, Hermann, Comparative Criminology (Boston, 1965), 532–62Google Scholar; and Lane, Roger, “Crime and the Industrial Revolution: British and American Views,” Journal of Social History, VII (Spring 1974), 287–303CrossRefGoogle Scholar.

6 See Mannheim, , Comparative Criminology, 95–98Google Scholar; and Morris, Terrence, The Criminal Area (London: 1957), 44–52Google Scholar.

7 See Ridker, Ronald G., “Discontent and Economic Growth,” Economic Development and Cultural Change, XI (October 1962), 1–15CrossRefGoogle Scholar; and Nieberg, H.L., Political Violence (New York: 1969), 15–16Google Scholar.

8 See Christiansen, Karl O., “Industrialization and Urbanization in Relation to Crime and Juvenile Delinquency,” International Review of Criminal Policy, XVI (October 1960), 3–8Google Scholar; and Hagedorn, R.B., Miller, J.P., and Labovitz, S., “Industrialization, Urbanization and Deviant Behavior: An Examination of Some Basic Issues,” Pacific Sociological Review, XIV:2 (1971), 177–95CrossRefGoogle Scholar.

9 See Szabo, Denis, Crimes et villes (Paris: 1960)Google Scholar for a study of urban-rural differences and criminal behavior in Belgium and France.

10 An elaboration of these arguments can be found in Olson, Mancur Jr., “Rapid Growth as a Destabilizing Force,” Journal of Economic History, XXIII (December 1963), 529–52CrossRefGoogle Scholar; and Kornhauser, William, The Politics of Mass Society (New York: 1959)Google Scholar.

11 The writings of Marx, Karl and Engels, Friedrich have had a profound influence on this line of thinking. See the classic study by Bonger, William, Criminality and Economic Conditions (Blommington: 1969Google Scholar; reprinted). Many early studies attempted to show that the poor were more criminal prone because of their deprived social condition.

12 See Mannheim, Comparative Criminology, 580-81.

13 An example is Ducéptiaux, Edouard, Mémoire sur le pauperisme dans les Flandres (Brussels: 1850).Google Scholar

14 For a discusssion of this view, see Tobias, J.J., Crime and Industrial Society in the 19th-century (New York: 1967), 150–52.Google Scholar

15 For example, high correlations were reported between property crimes, emigration, and the price of rye in Bavaria between 1835 and 1861. See Mayr, Georg Von, “Statistik der gerichtlichen Polizei im Konigreiche Bayern,” in Beiträge zur Statistik im Königreiche Bayern (Munich: 1867)Google Scholar. A study limited to Saxony suggested a strong dependence of economic crimes on the rise and fall of real wages. See Renger, Ewald, Kriminalität, Preis und Lohn (Leipzig: 1933)Google Scholar. Snyder and Tilly examine some of these relationships in “Hardship and Collective Violence in France.”

16 The study referred to is Lodhi and Tilly, “Urbanization, Crime and Collective Violence in 19th-century France.

17 Tilly and Lodhi may have been influenced by the work of A.M. Guerry, who examined geographical variations in wealth and the incidence of crime in France during the 1830’s. Guerry measured wealth as the amount of direct taxation in each administrative department. Guerry’s results failed to show any significant relationship between crime and the distribution of wealth across France. Although the north was generally wealthier than the south at this time, crime against property was higher in the north. Personal crime, being higher in the south than in the north, showed an inverse pattern. Guerry’s findings also revealed a negative relationship between the incidence of crime and the level of educational attainment (literacy) within the departments. See Guerry, A.M., Essai sur la statistique morale en France (Paris: 1833)Google Scholar. Chen also discusses the role of wealth inequalities and crime in France at a later period. Chen finds a positive relationship. See Chen, Y.Y., Études statistiques sur la criminalité en France de 1895 à 1930 (Paris: 1937), 173–75Google Scholar.

18 See Cutright, Phillips, “Inequality: A Cross-National Analysis,” American Sociological Review, XXXII (August 1967), 562–78CrossRefGoogle Scholar; and Kuznets, Simon, “Economic Growth and Income Inequality,” American Economic Review, XLV (March 1955), 1–28Google Scholar.

19 These arguments are developed by Sofranko, Andrew J. and Bealer, Robert C., Unbalanced Modernization and Domestic Instability: A Comparative Analysis (Beverly Hills: Sage Professional Papers in Comparative Politics, 1972)Google Scholar; and Hechter, Michael, “Towards a Theory of Ethnic Change,” Politics and Society, 11:1 (1971), 2145Google Scholar.

20 For a discussion of these studies, see Nagel, Jack H., “Inequality and Discontent: A Nonlinear Hypothesis,” World Politics, XXVI (July 1974), 453–72CrossRefGoogle Scholar; and Hibbs, Douglas A. Jr., Mass Political Violence (New York: 1973), 196–99Google Scholar.

21 On the economic development of Germany during this period, see Clapham, J.H., The Economic Development of France and Germany, 1815-1914 (Cambridge: 1951), 195–231 and 278–338Google Scholar; Landes, David, The Unbound Prometheus (London: 1969)Google Scholar, chapter 4; and Stolper, Gustav, The German Economy, 1870-1940 (New York: 1940)Google Scholar.

22 Steam power and transportation experienced a tremendous expansion during this period. In 1879, Prussia recorded 29,895 stationary steam engines; and by 1901, this number had increased to 75,958. Between 1870 and 1890, the total length of rail lines in Prussia had been increased from 11,523 to 26,350 kilometers. For additional data, see Gerhard Bry, with Charlotte Boschan, Wages in Germany, 1871-1945 (Princeton: 1960); Hoffmann, Walther G., Grumbach, Franz, and Hess, Helmut, Das Wachstum der deutschen Wirtschaft seit der Mitte de 19. Jahrhunderts (Berlin: 1965)CrossRefGoogle Scholar; and Rosenberg, Hans, Grosse Depression und Bismarckzeit: Wirtschaftsablauf Gesellschaft und Politik in Mitteleuropa (Berlin: 1967)CrossRefGoogle Scholar.

23 The major statistical series is the Statistik des Deutschen Reiches. Scholars may find this series unwieldy although it is the more complete set of data, and they may become inundated by the vastness of its holdings. It is perhaps more practical to rely on the abbreviated, but equally data-rich, statistical yearbooks. In addition to the principal statistical yearbook covering all of Imperial Germany, each separate Land generally published its own statistical compendium. Unless otherwise noted, the primary data source for this study was Jahrbuch fuer die Amtliche Statistik des Preussischen Stoats (Berlin, Koeniglichen Statistische Bureau). Data have been drawn from various annual editions. For additional commentary on German statistical sources, see Sheehan, James J., “Quantification in the Study of Modern German Social and Political History,” in Lorwin, and Price, , eds., Dimensions of the Past, pages 301–31Google Scholar.

24 For a general discussion of the problems associated with the use of crime statistics, see Mannheim, Comparative Criminology, 95-140.

25 Tobias, Crime and Industrial Society, 14-21.

26 Difficulties in cross-national and cross-cultural analyses of crime are discussed in Mannheim, Hermann, Social Aspects of Crime in England Between the Wars (London: 1940), 36–38Google Scholar.

27 Gatrell, V.A.C. and Hadden, T.B., “Criminal Statistics and their Interpretation,” in Wrigley, , ed., Nineteenth Century Society, 337–38Google Scholar.

28 Sorokin, Pitirim A., Social and Cultural Dynamics: Fluctuation of Social Relationships, War and Revolution (New York: 1937), Volume III, 398Google Scholar.

29 A comparison of selected indicators for Prussia and all of Imperial Germany in 1880 will confirm this point:

30 On the use of territorial indicators to measure wealth and social welfare, see Smith, David M., The Geography of Social Well-Being in the United States: An Introduction to Territorial Social Indicators (New York: 1973), 1–23 and 52–78Google Scholar. The measurement of change at the aggregate level poses a problem for crosssectional designs of this sort. After considering several possible alternatives, urban growth has been measured as the first differences in urban population percentages between 1871 and 1885, while steam power change represents a percentage increase between 1878 and 1886. Neither score is independent of the earlier values and this represents a source of bias. See Van Meter, Donald S., “Alternative Methods of Measuring Change: What Differences Does it Make?” Political Methodology, IV (Fall 1974), 125–40Google Scholar.

31 John Knodel and Steven Hochstadt, “Illegitimacy in Imperial Germany: A Study of Urban-Rural Differentials,” forthcoming in a volume dealing with illegitimacy in 19th-century society under the general editorship of Peter Lazlett.

32 Exner, Franz, Kriminalbiologie (Hamburg: 1939)Google Scholar. Gustav Ashaffenburg has also emphasized regional patterns in the consumption of alcohol as an explanatory factor in the German crime rate during this period. See his study, Crime and its Repression, trans. Adalbert Albrecht (Boston: 1913), 40-45.

33 Burchardt, Hans H., Kriminalität in Stadt und Land (Berlin and Leipzig: 1935)Google Scholar.

34 Disturbance data for the Prussian districts are reported in Richard Tilly, “Popular Disorders in Nineteenth Century Germany,” 35-39.

35 Guerry had commented in this finding in the 1830’s. A recent geographical examination of the crime rate in post World War II France indicates that this situation has continued to remain quite stable. See Benjamin, Roger, “Aperçus géographiques sur la criminalité et la delinquance in France,” Revue française de sociologie, III (July-September 1962), 301–15CrossRefGoogle Scholar.

36 The crime component loadings were as follows: crime against property (.94); crime against the person (.83); crimes against the state, religious, and public laws (.90). District scores on this component indicated that the high crime areas were located in the provinces of East and West Prussia, Posnania, and Silesia, with the exception of the district of Liegnitz. This geographical stability in the crime rate began to erode by the end of the century. We will discuss reasons for the change in Part II.

37 See Richard Tilly, “Popular Disorders in Nineteenth Century Germany,” 25.

38 On the use of taxation as a measure of bureaucratic institutionalization, see Hibbs, Mass Political Violence, 98-99.

39 The high degree of intercorrelation among various wealth indicators in Prussia during this period allows us to construct a single wealth component similar to the one constructed for crime. This wealth component summarizes approximately three-fourths of the variation among the following indicators: direct taxes (.96); practicing physicians (.94);local tax revenues (.92); educational expenditures per student (.89); and the number of bank deposit accounts over 1,000 marks (.47). The poorest districts were located in eastern Prussia. However, rather than use this wealth metric in our analyses, we will continue to employ the amount of direct taxes per capita as our wealth indicator. It is simple to interpret and it displayed the highest loading on our wealth component.

40 Despite impressive economic growth over the last quarter of the 19th-century, the distribution of wealth within Prussia remained relatively static up to World War I. Thomas Orsagh has attempted to estimate the geographical distribution of income in Germany during the 1880’s based on regional characteristics of the occupational structure. Using the parameters of Orsagh’s model to estimate the distribution of income for 1882, we found it to be strongly correlated with our wealth measure, which is based on per capita taxation (r = +.73). For the assumptions underlying Orsagh’s model, see his “The Probable Geographical Distribution of German Income, 1882-1963,” Zeitschrift für die gesamte Staatswissenschaft, CXXIV (May 1968), 280-311.

41 The problem of multicollinearity among the predictor variables, as reflected in high correlations (>-70) among migration, wealth, urbanity, and population growth, makes it difficult for us to untangle the separate effects of these variables on crime with any degree of precision and certainty. Despite these high correlations, the overall effects of multicollinearity do not appear to be severe in this case. Several of the regression estimates are substantially larger than their standard errors, and a simple heuristic test of the intercorrelations among the entire set of predictors yields a probability of less than .001 that the matrix is singular (X² = 62.56, with 21 degrees of freedom). This test, which is based upon the chi-square statistic, is computed as follows:

where :

For additional comments on the problems of multicollinearity and an extended discussion of this test, see Richard C. Rockwell, “Assessment of Multicollinearity,” Sociological Methods and Research, III (February 1975), 308-20; and Yoel Haitovsky, “Multicollinearity in Regression Analysis: Comment,” Review of Economics and Statistics, LI (November 1969), 486-89.

42 On adjusting the R² for small samples, see Guilford, J.P., Fundamental Statistics in Psychology and Education (New York: 1965), 400–01Google Scholar. The adjusted value (Ṝ² ) is computed as follows:

where n equals the number of cases, and m equals the number of variables.

43 The assumptions underlying path analysis are treated in Kenneth Land, “Principles of Path Analysis;” and Heise, David R., “Problems in Path Analysis and Causal Inference,” in Borgatta, Edgard F., ed., Sociological Methodology (San Francisco: 1969), pages 3–37 and 38–78Google Scholar respectively. We do not wish to impute any strict causal interpretation to our results. Our goal in this analysis is not to achieve an optimal statistical formulation, but to present a “rationally defensible and substantively interesting interpretation of a set of data.” See Duncan, Otis Dudley, “Partials, Partitions, and Paths,” in Borgatta, Edgar F. and Bohrnstedt, George W., eds., Sociological Methodology (San Francisco: 1970), 38–39Google Scholar.

44 For a discussion of some of the arguments underlying the formulation of our path model, see Hawley, Amos H., Human Ecology: A Theory of Community Structure (New York: 1950), 328–32Google Scholar.

45 See Hibbs, Mass Political Violence, 36-42. Richard Tilly suggests that demographic conditions may have contributed to social stress in 19th-century Germany. There is some evidence that population growth during the first half of the 19th-century badly outstripped employment opportunities, and contributed to a heavy labor surplus in later years. We noted previously a significant correlation (Spearman’s r = +.48) between population growth (1867-1875) and popular disturbances (1850-1875) in Prussia. For additional comments, see Tilly, Richard, “Germany,” in Tilly, Charles, Tilly, Louise, and Tilly, Richard, The Rebellious Century: 1830-1930 (Cambridge: 1975)CrossRefGoogle Scholar.

46 Tobias, Crime and Industrial Society, 162. Juvenile delinquency steadily increased in Prussia over the second half of the 19th-century. Throughout this period, as we will demonstrate in Part II, it was significantly related to the adult crime rate. Data on the number of convictions for all of Imperial Germany during the period 1886 to 1895 indicate that the 12 to 25 year old group was the most criminal prone. The overall crime rate during this period averaged 1143.3 convictions per 100,000. The average rate for males between 18 and 25 years of age was 3309.1 per 100,000. More detailed comparisons can be found in our study, “Socioeconomic Aspects of the Delinquency Rate in 19th-century Germany” (forthcoming).

47 Tobias, Crime and Industrial Society, 167.

48 See Richard Tilly, “Popular Disorders in Nineteenth Century Germany,” 25. Richard Tilly points to Berlin as the “great and important exception.” Berlin was not only one of the most highly urbanized areas of Imperial Germany during the decade of the 1880’s, but it was also plagued with the problems of sporadic strikes and collective violence, in addition to a high crime rate. Between 1870 and 1885, the population of Berlin had increased by over 60 percent. There had been a severe housing shortage in 1872. Berlin had a well-established police force as early as 1848. The rate of violent crime in Berlin reached a level of 1799 convictions per 100,000 in 1885, of which the largest category was composed of violent sexual crimes, followed by assault and battery, and murder or attempted murder. There is some evidence to indicate that a general positive relationship existed between levels of violent crime and collective protest in Berlin during the second half of the 19th-century. See Jeffrey Bergner, “On Collective and Individual Violence: Berlin and Vienna” (unpublished ms., University of Pennsylvania, n.d.).

49 Prussia was not an homogenous entity but one that expanded by conquest and annexation. This raises the issue of culture conflict and its impact upon the crime rate in the eastern border districts, especially in light of the anti-Polish policies pursued by the Prussian administration in this region. Areas of high ethnic diversity also tended to have significantly higher overall crime rates (r = +.77). Ethnic diversity is measured here as the proportion of the inhabitants in each district of non-German mother tongue, principally Poles. Crime and wealth appeared to be negatively related in the eastern districts (with the exception of Berlin and Stralsund), and somewhat positively related throughout the rest of Prussia. One might argue that since the eastern districts were among the poorest, inequalities in wealth would likely combine with other social factors (e.g., ethnicity), and thus tend to have a greater impact on the crime rate. Certain types of crime would be expected to diminish with moderate increases in the level of social welfare. On the other hand, it is also reasonble to assume that with higher levels of wealth come a number of structural changes in society, along with new opportunities that are likely to stimulate an increase in the level of crime. This reasoning would account for the curvilinear relationship observed between wealth and the crime component.

50 See Wiliamson, Jeffrey G., “Regional Inequality and the Process of National Development,” Economic Development and Cultural Change, IV (July 1965), Part II, 1–84CrossRefGoogle Scholar.

51 For additional arguments and evidence, see Hechter, Michael, “Industrialization and National Development in the British Isles,” Journal of Development Studies, VIII (April 1972), 17–74Google Scholar.

52 Hamerow, Theodore S., The Social Foundations of German Unification, 1858-1871 (Princeton, 1969), 45–55Google Scholar.

53 Ibid., 63.

54 See Davies, James C., “Toward a Theory of Revolution,” American Sociological Review, XXVII (February 1965), 5–19Google Scholar.

55 For evidence and additional commentary, see Hamerow, Social Foundations of German Unification, 3-132.