![Sample our Politics & International Relations journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5947/TOC-Subject-Sample-2021-Politics-International-Relations.jpg"})
ABSTRACT
Using a new measure of “comprehensive democracy,” our analysis traces the global democratic trend over the last 116 years, from 1900 until 2016, looking in particular at the centennial trend’s cultural zoning. As it turns out, democracy has been proceeding and continues to differentiate the world’s nations in a strongly culture-bound manner: high levels of democracy remain a distinctive feature of nations in which emancipative values have grown strong over the generations. By the same token, backsliding and autocratization are limited to cultures with under-developed emancipative values. In line with this finding, public support for democracy neither favours democratization, nor does it prevent autocratization in disjunction from emancipative values. On the contrary, public support for democracy shows such pro-democratic effects if – and only if – it co-exists in close association with emancipative values. The reason is that – in disconnect from emancipative values – support for democracy frequently reverts its meaning, indicating the exact opposite of what intuition suggests: namely, support for autocracy. In conclusion, the prospects for democracy are bleak where emancipative values remain weak.
Acknowledgements
Part of our study has been funded by the Russian Academic Excellence Project ‘5-100’, for which we are grateful. We also thank the anonymous reviewers of Democratization as well as its editor, Aurel Croissant, for invaluably helpful comments. Special thanks go to Carl Berning, Anna Luerhmann and various scholars of the V-Dem Institute at the University of Gothenberg for detailed feedback on an earlier version of this article. Remaining shortcomings are all in our own responsibility.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
1 Huntington, The Third Wave; Berg-Schlosser, “Long Waves and Conjunctures of Democratization,” 41–54; Markoff and White, “Global Wave of Democratization,” 55–73.
2 Diamond, The Spirit of Democracy, Ch. 1; Haerpfer et al., “Introduction,” 1–9.
3 Levitsky and Ziblatt, How Democracies Die, 72–96.
4 Alexander and Welzel, “The Myth of Deconsolidation.”
5 Held, Models of Democracy.
6 Coppedge et al., “V-Dem [Country-Year] Dataset v7” (online at www.vdem-net.org).
7 Welzel, Freedom Rising, 66–73.
8 Center for Systemic Peace, Polity IV Annual Time Series (online at www.syetmicpeace.org); Freedom House, Freedom in the World (online at www.freedomhouse.org).
9 Coppedge et al., “The V-Dem Codebook v7” (online at www.vdem-net.org).
10 Dahl, Polyarchy; Sen, Development as Freedom.
11 Teorell et al., “Measuring Electoral Democracy.”
12 V-Dem also provides measures for a “deliberative” and “egalitarian” component of democracy. We do not include these two components into our measure of “comprehensive democracy” because they extend the definition of democracy beyond institutional rules. We limit the definition of democracy to institutional rules and leave out the societal pre-conditions in which these rules are supposedly grounded. Otherwise, we could not examine the influence of the latter on the former, which is the whole point of our study.
13 Coppedge et al., “Measuring High Level Democratic Principles”; Luehrmann, Lindberg, and Tannenberg, “Regimes in the World.”
14 Vanhanen, Democratization; Alexander and Welzel, “Measuring Effective Democracy,” 271–89.
15 Goertz, Social Science Concepts, 45. For a discussion of the pros and cons of multiplicative relative to additive combinations, see Response RII-1 in the “Review Response” part of the OA.
16 Cf. Markoff and White, “Global Wave of Democratization”; Møller and Skaaning, “The Third Wave,” 97–109; Mechkova, Luehrmann and Lindberg, “How Deep is he Democratic Recession?”
17 Huntington, The Clash of Civilizations, 1–50.
18 Braudel, A History of Civilizations; McNeill, The Rise of the West; Quigley, The Evolution of Civilizations.
19 Inglehart and Welzel, Modernization, Cultural Change and Democracy, 15–25.
20 Ferguson, Civilization; Morris, Why the West Rules - For Now.
21 Ibid.
22 Welzel, Freedom Rising, 126–35.
23 Ibid. For a more detailed discussion of the explanatory power of culture zones compared to nations and sub-national regions, see Note 1 in the “Explanatory Notes” part of the OA.
24 Welzel, Freedom Rising, 25–33.
25 Inglehart and Welzel, Modernization, Cultural Change and Democracy, 137.
26 McNeill, The Rise of the West.
27 Mechkova, Luehrmann and Lindberg, “How Deep is he Democratic Recession?” find similar evidence, although our comprehensive measure of democracy exhibits a more modest democratic trend, while our population-weighted perspective shows a more modest recent recession of democracy.
28 Alexander and Welzel, “Measuring Effective Democracy,” 273. For a more detailed discussion of these considerations, see Note 2 in the “Explanatory Notes” part of the OA.
29 This constancy is interrupted by a temporary increase in the inter-war period, due to the rise of fascism.
30 For a more detailed discussion of this point see Response RII-3 in the “Review Response” section of the OA.
31 Doorenspleet, Rethinking the Value of Democracy, 1–25.
32 Using Cingranelli and Richards’ “physical integrity” index as a measure of humanitarian norms and the World Bank’s “control of corruption” index as a measure of public goods provision, we find that full democracies outperform deficient democracies by a large margin, whereas the latter hardly outperform autocracies, whether straight or mixed. These results are available upon request.
33 Welzel, Freedom Rising, 25–33.
34 Levitsky and Ziblatt, How Democracies Die.
35 Welzel, Freedom Rising, 37–56.
36 Welzel and Inglehart, “Political Culture, Mass Beliefs and Value Change,” forthcoming. The measure summarizes country-year data on average life expectancies, per capita incomes, schooling years (primary, secondary and tertiary combined) and inverse female fertility rates in a single factor score variable, which we standardize into a scale range from 0 to 1, with decimal fractions for intermediate positions.
37 The observation period ends here due to data limitations. For a discussion of the slight temporal fluctuation in the linkage between democracy and resources, see Note 3 in the “Explanatory Notes” part of the OA.
38 Welzel, Freedom Rising, 66–73.
39 Welzel standardizes the four sub-indices into a scale range from minimum 0 for the least emancipatory position to maximum 1 for the most emancipatory one, with decimal fractions indicating intermediate positions. He then averages the four sub-indices into the overall index of emancipative values and calculates mean scores for each national population and survey wave.
40 Scholars debate the validity of the emancipative values index. Alemán and Woods (“Value Orientations from the World Values Survey”) emphasize that the twelve single items show variable loadings across countries in individual-level factor analyses. The authors conclude from this observation that scores on the emancipative values index are inequivalent between countries. Welzel and Inglehart (“Misconceptions of Measurement Equivalence”) disprove this conclusion by demonstrating that cross-country variability in single item loadings does not affect the overall index’s linkages to other variables of interest. The authors explain that this phenomenon is due to “compositional substitutability”: national cultures differ and accordingly create variation as concerns the domains of emancipation that are most salient in a country – hence, different loadings in different countries. But these variations in single item loadings are equivalent substitutes to each other, for which reason the overall index’s functioning remains intact. Sokolov (“The Index of Emancipative Values”) merely repeats the point by Alemán and Woods and does not adequately engage with “compositional substitutability.” Response RI-1 in the “Review Response” part of the OA and Note 4 in the “Explanatory Notes” part of the OA discuss this point further. In any case, the bottom line is that compositional substitutability makes the overall index of emancipative values a valid measure of cross-cultural differences, visible in this index’s strong linkages to multiple other phenomena of a developmental nature.
41 Welzel and Inglehart, “Political Culture, Mass Beliefs and Value Change,” forthcoming.
42 Note 5 in the “Explanatory Notes” part of the OA explicates the estimation and its underlying assumptions.
43 Welzel and Inglehart, “Political Culture, Mass Beliefs and Value Change,” forthcoming, demonstrate this point in response to Dahlum and Knutsen, “What Counts as Evidence?” 1–7.
44 The section “Panel Regressions” in the “Supplementary Findings” part of the OA provides a detailed documentation of these findings.
45 Foa and Mounk, “The Signs of Deconsolidation”; Foa and Mounk, “The Dangers of Deconsolidation.”
46 Ibid.
47 Welzel and Kirsch, “Democracy Misunderstood,” 1–29.
48 Response RI-3 in the “Review Response” part of the OA discusses in more detail our observations that people in many countries (a) mis-understand democracy in autocratic terms, (b) mistake the absence of democracy for its presence and that (c) these confusions are real rather than reflecting “false preferences.”
49 Ibid.
50 Kruse, Ravlik and Welzel, “Democracy Confused,” 1–30.
51 This measure is based on a WVS-question asking people how good an idea they think it is to “having a democratic system.” People answer this question in a four-fold response format, which I recode along increasing strength of support: 0 (“very bad idea”), 0.33 (“bad idea”), 0.66 (“good idea”), 1.00 (“very good idea”).
52 Welzel, Inglehart and Kruse, “Pitfalls in the Study of Democratization”; Welzel and Inglehart, “Political Culture, Mass Beliefs and Value Change,” forthcoming. See also the updated evidence in Appendix-Figure 9 in the “Graphical Illustrations” section of the OA.
53 The cross-lagged autoregressive path models in the “Panel Regressions” section of the OA show a significant and positive effect of prior emancipative values on subsequent measures of comprehensive democracy but no reverse effect of prior measures of comprehensive democracy on subsequent emancipative values, controlling for both variables’ temporal autocorrelation. These results hold for different time-period averages. Panel regressions, shown in the same section of the OA, also support this evidence, using different estimators (System-GMM, Difference-GMM, fixed effects, random-coefficient models), different lag lengths and dozens of relevant controls.
54 Levitsky and Ziblatt, How Democracies Die.
55 Foa and Mounk, “The Signs of Deconsolidation”; Foa and Mounk, “The Dangers of Deconsolidation.”
56 Norris and Inglehart, “Trump and the Populist Authoritarian Parties.”
57 Ibid.
58 Alexander and Welzel, “The Myth of Deconsolidation.”
59 Fording and Schram, “Hard White.”
60 Foa and Mounk, “The Signs of Deconsolidation”; Foa and Mounk, “The Dangers of Deconsolidation.”
61 Alexander and Welzel, “The Myth of Deconsolidation.”; Norris, “Is Western Democracy Backsliding?”; Voeten, “Are People Really Turning away from Democracy?”
62 Welzel, Freedom Rising, 91.
63 Among the few places where the WVS data show no increase of emancipative values from older to younger cohorts are the Arab oil monarchies, including Bahrain, Kuwait and Qatar.
Additional information
Funding
Notes on contributors
Lennart Brunkert
Lennart Joe Brunkert is a Doctoral Research Associate in the DFG project “The Cool Water Effect: Why Human Civilization Turned towards Emancipation in Cold-Wet Regions”. In his PhD project he analyzes legitimation strategies of non-democratic regimes from multiple perspectives with a focus on quantitative measurement of regime performance. Brunkert studied European Studies (BSc) and Public Economics, Law and Politics (MA) at the universities of Sønderborg and Lüneburg. His research interests include the comparative analysis of autocratic regimes, civil society in non-democratic settings, political culture and quantitative research methods.
Stefan Kruse
Stefan Kruse is a Postdoctoral Research Associate in the DFG project “The Cool Water Effect: Why Human Civilization Turned towards Emancipation in Cold-Wet Regions”. Kruse obtained his PhD, with distinction, at the Centre for the Study of Democracy at Leuphana University Lüneburg on the institutional conditions of effective development cooperation. He studied Economics and Social Sciences (BSc) and Public Economics, Law and Politics (MA) at the universities of Lüneburg and Barcelona. During 2011 and 2015 he worked as a junior researcher at the Hamburg Institute of International Economics (HWWI). His research interests include political culture and development research as well as methods of comparative political science. He has published in peer-reviewed academic journals including the British Journal of Political Science.
Christian Welzel
Christian Welzel, Member of the German Academy of Sciences (Leopoldina), is the Political Culture Research Professor at Leuphana University in Lueneburg, Germany, and Chief Foreign Director at the National Research University – Higher School of Economics in Moscow, Russia. He is also the former President and current Vice-President of the World Values Survey. Welzel’s research focuses on human empowerment, emancipative values, cultural change, and democratization. Author of roughly 150 scholarly publications, he published in 2013 his award-winning monograph Freedom Rising: Human Empowerment and the Quest for Emancipation (CUP). His most recent books include Democratization (with Christian Haerpfer, Ronald Inglehart and Patrick Bernhagen, 2nd fully revised and updated edition at OUP 2018) and The Civic Culture Transformed (with Russell J. Dalton, at CUP 2014).