Thinking about how democracy works

Politics and Society

Volumn 24, Issue 1, 1996, Pages 7-18

  Goldberg, Ellis ^a  

^a NONE

Author keywords

[No Author keywords available]

Indexed keywords

EID: 0030306694     PISSN: 00323292     EISSN: None     Source Type: Journal    
DOI: 10.1177/0032329296024001003     Document Type: Article

References (27)

1. Robert Putnam, Making Democracy Work (Princeton, NJ: Princeton University Press, 1993), 86.
2. George Hildebrand, Growth and Structure in the Economy of Modern Italy (Cambridge, MA: Harvard University Press, 1965), 281.
3. Putnam dismisses, far too quickly I believe, the standard account precisely because it fails this test. See fn. 6 on 220 where Putnam argues that existing macroeconomic theories fail because they can explain variation between north and south but not within the north and south taken separately.
4. See David Laitin, "The Civic Culture at Thirty," American Political Science Review 89, no. 1 (March 1995): 168-73, who emphasizes, "Putnam first shows that . . . [economic development] . . . can explain differences between north and south but not among regions within north and south. More importantly his [Putnam's] explanatory variable, civic community (dense horizontal networks of associations) can explain both sets of differences" (172). Or at least I presume the italics, which are Laitin's, should be read emphatically.
5. As I was completing the final revision of this article, and thanks to Barbara Geddes, I was able to see Antonio Mutti's "Paths of Development," APSA-CP Newsletter 6, no. 2 (1995): 6-7. Mutti calls attention to the within-group correlations which he terms "much less strong, even though acceptable and statistically significant" and also raises methodological issues which I address below. The within-group variations are certainly less strong; unfortunately it is not apparent to me that they invariably retain their significance.
6. The original chart to which I refer here is in Putnam, Making Democracy Work, 108, fig. 4.12.
7. Strictly speaking, the data within the north and south on civic community and clericalism appear to be orthogonal. If two data sets are orthogonal or at right angles to each other they are unrelated to each other. Such data sets could still exhibit a very high r as long as the dispersion on one variable around its mean was more or less equal to the dispersion of the other variable around its mean; unfortunately the use of data sets developed from factor scores is more likely to create this condition than data sets composed of raw data.
8. Figure 1 is a chart drawn from data points which I derived from Putnam's chart. It looks similar to but is not identical with fig. 4.12 in Making Democracy Work. Table 1 is the set of derived data points; the "units" are quarter-inch hatch marks and have no relevance other than providing numbers for some simple crunching. Working in this way was initially simply an indulgence of my curiosity and a skepticism about the book. Professor Putnam has since seen a version of this review and very generously offered to share the data from which he did his own calculations. I declined primarily due to the pressure of other obligations, but on reflection I also think it preferable to judge a book by what the author chose to put into print. I also feel Professor Putnam is owed a double round of thanks by the profession and those researchers - and I know of at least two - to whom he has given access to his primary data so that they may critique, confute, or even agree with him.
9. Using the original data, Putnam's r is -.76. Because all that is required to calculate r is to locate data points in two-dimensional space, for which purpose any units in a rectangular coordinate system will do, it was not absolutely necessary to check as long as the hatch marks are adequately spaced. It is reassuring, however, to know that not much error had been introduced by the process.
10. Putnam, Making Democracy Work, 113, fig. 4.14. In the three successive examples the all-Italian correlation coefficient is reported from MDW; I calculated southern coefficient as described above.
11. Ibid., 151, fig. 5.3.
12. Ibid., 151, fig. 5.4.
13. A slightly different way to think about this which may help some readers is to use Table 1 instead of Figure 1 and imagine many similar tables all of which list the names of the regions ranked from number one to number twenty. If, for example, civic community causes or predicts outcomes in an election in 1946 and another one in 1980 then we can expect the regions to have the same rank orderings in both elections. Nonparametric statistics allow us to test the similarity between two sets of ordinal rankings.
14. I do not insist very strongly on this result and the difference in the descriptive statistic is what I am aiming at here; if one thought of the elections as a kind of sample then it is interesting to know that the all-Italian calculation is statistically significant but neither the northern or southern ones are. The relatively high value for all-Italian tau b derives from the same source as the high all-Italian r values discussed earlier: every southern region ranks lower than all the northern ones. It is quite possible that attempting to estimate the rank-orders by eye introduced significant measurement error on my part and had I been able to take advantage of Professor Putnam's very kind offer perhaps the calculation would more strongly support his case. It is, however, cautionary to note that the high all-Italian tau b coexists with exactly the same within-group variation problems I have already noted.
15. I know very little about Italy, but I am puzzled that Putnam's data exhibits some other inexplicable features at the region level. For example, neighboring Liguria and Trentino are quite far apart on scores for clericalism but very similar in civicness. One might have expected a more "natural" or at any rate slightly large social ecology to be at work providing continuity at points of geographical contiguity.
16. For a review of the literature on credit (including the hawalah and suftaja) in the Islamic Mediterranean and some of its implications, see Ellis Goldberg, "Was There an Islamic 'City?' " in Cities in the World System, ed. Resat Kasaba (New York: Greenwood, 1991). The use of credit was known to the Babylonians. See W. F. Leemans, The Old-Babylonian Merchant: His Business and Social Position (Leiden, The Netherlands: E. J. Brill, 1950).
17. For a review of the literature on credit (including the hawalah and suftaja) in the Islamic Mediterranean and some of its implications, see Ellis Goldberg, "Was There an Islamic 'City?' " in Cities in the World System, ed. Resat Kasaba (New York: Greenwood, 1991). The use of credit was known to the Babylonians. See W. F. Leemans, The Old-Babylonian Merchant: His Business and Social Position (Leiden, The Netherlands: E. J. Brill, 1950).
18. See inter alia, Ira Lapidus, A History of Islamic Societies (Cambridge, UK: Cambridge University Press, 1988) and his earlier pathbreaking Muslim Cities in the Later Middle Ages (Cambridge, UK: Cambridge University Press, 1984).
19. See inter alia, Ira Lapidus, A History of Islamic Societies (Cambridge, UK: Cambridge University Press, 1988) and his earlier pathbreaking Muslim Cities in the Later Middle Ages (Cambridge, UK: Cambridge University Press, 1984).
20. See Resat Kasaba, "Democracy and Populism in Turkey," in Rules and Rights in the Middle East, ed. Ellis Goldberg, Resat Kasaba, and Joel Migdal. (Seattle: University of Washington Press, 1993).
21. Lacking census data, I did some back-of-the-envelope calculations in which I assumed that 1 percent of northerners moved south each year and 5 percent of southerners moved north. This predicts depopulation of the south and also suggest that a strictly cultural argument (adding southerners and their values to the north) would only work if a significant degradation of northern civic community is observed over twenty years. Putnam's work does not lend much support to the political culture perspective if culture is supposed to have any cognitive content. On the other hand, to talk about social capital and trust may really be a discussion about the distribution of honesty or commitment in society. See, for example, Amartya Sen, "Rational Fools: A Critique of the Behavioural Foundations of Economic Theory," in Choice, Welfare and Measurement, ed. S. Amartya (Cambridge: MIT Press, 1982) and Michael Taylor, "Good Government," The Journal of Political Philosophy (forthcoming). In the last few years tens of thousands of Egyptians have lost tens of millions of dollars by trusting their hard currency savings to "Islamic" banks in preference to the formal state-regulated system. Unfortunately many of the Islamic banks turned out to be little more than Ponzi schemes and most of the remainder made extremely speculative investments which went sour.
22. Lacking census data, I did some back-of-the-envelope calculations in which I assumed that 1 percent of northerners moved south each year and 5 percent of southerners moved north. This predicts depopulation of the south and also suggest that a strictly cultural argument (adding southerners and their values to the north) would only work if a significant degradation of northern civic community is observed over twenty years. Putnam's work does not lend much support to the political culture perspective if culture is supposed to have any cognitive content. On the other hand, to talk about social capital and trust may really be a discussion about the distribution of honesty or commitment in society. See, for example, Amartya Sen, "Rational Fools: A Critique of the Behavioural Foundations of Economic Theory," in Choice, Welfare and Measurement, ed. S. Amartya (Cambridge: MIT Press, 1982) and Michael Taylor, "Good Government," The Journal of Political Philosophy (forthcoming). In the last few years tens of thousands of Egyptians have lost tens of millions of dollars by trusting their hard currency savings to "Islamic" banks in preference to the formal state-regulated system. Unfortunately many of the Islamic banks turned out to be little more than Ponzi schemes and most of the remainder made extremely speculative investments which went sour.
23. I do not wish to minimize in the slightest that I believe Putnam's enterprise, an attempt to understand persistent patterns of social and economic disadvantage in one country, is an extremely serious one for Americans. A far more controversial attempt recently to bring such heavy statistical guns to bear on the problem of persistent regional and ethnic inequality is Richard J. Herrnstein and Charles Murray, The Bell Curve (New York: Free Press, 1994), which devotes a chapter to "Civility and Citizenship" (253-66). People with higher IQs, they suggest, are likely to be more engaged in behavior which makes for political community. The explanatory strategies of the two books are not both likely to be right, although I suppose it is possible that northern Italians are, on average, both smarter and more cooperative than southerners.
24. I do not wish to minimize in the slightest that I believe Putnam's enterprise, an attempt to understand persistent patterns of social and economic disadvantage in one country, is an extremely serious one for Americans. A far more controversial attempt recently to bring such heavy statistical guns to bear on the problem of persistent regional and ethnic inequality is Richard J. Herrnstein and Charles Murray, The Bell Curve (New York: Free Press, 1994), which devotes a chapter to "Civility and Citizenship" (253-66). People with higher IQs, they suggest, are likely to be more engaged in behavior which makes for political community. The explanatory strategies of the two books are not both likely to be right, although I suppose it is possible that northern Italians are, on average, both smarter and more cooperative than southerners.
25. See Vera Lutz, Italy, A Study in Economic Development (London: Oxford University Press, 1962), 36-7.
26. Hildebrand and Lutz reflected such scholarship but more recent scholars continue to believe such an explanation has merit. See Robert Bergeron, La Basilicate (Rome: École Française de Rome, 1994).
27. For an attempt to write a non-ideological history of another powerful labor party see Michael Shalev, Labour and the Political Economy in Israel (London: Oxford University Press, 1992).