The theory of democracy revisited by Giovanni Sartori, , Chatham House Publishers edition, in English. In Defense of Democracy as a Way of Life: A Reply to Talisse's Pluralist Objection .Shane J. Ralston - - Transactions of the Charles S. Peirce Society The theory of democracy revisited / Giovanni Sartori. Later Title. Journal of forensic odonto-stomatology (Online). Former Title. International journal of forensic.
|Language:||English, Spanish, French|
|Genre:||Children & Youth|
|Distribution:||Free* [*Registration needed]|
Get this from a library! The theory of democracy revisited. [Giovanni Sartori]. The theory of democracy revisited. by: Sartori, Giovanni, cn urn:acs6: theoryofdemocrac00sart:pdf:a2dbdc PDF | This article reviews Giovanni Sartori's contribution to and the theories of democracy as they . The Theory of Democracy Revisited is.
Survey on Eastern Europe special supplement. March 1— Google Scholar Eisenstadt, S. Google Scholar Feshbach, Murray. Joint Economic Committee. Soviet Economy in the s: Problems and Prospects. Part 2. DC: Government Printing Office. December — The Political Economy of the Soviet Union. July 26 and September 91— Google Scholar Flanigan, W. Gillespie and B.
Beverly Hills: Sage. Google Scholar Flora, Peter. I, edited by S. Eisenstadt and S. Beverly Hills: Sage Publications. Google Scholar Freedom Review. Google Scholar Garton Ash, Timothy. Google Scholar Gellner, Ernest.
Google Scholar Gibson, James L. Miller, W. Reisinger and V. Boulder, CO: Westview Press. Google Scholar Gladdish, Ken. Google Scholar Glassman, Ronald. China in Transition: Communism, Capitalism and Democracy. Westport, CN: Praeger. Google Scholar Gorbachev, Mikhail. Google Scholar Service Daily Report. Google Scholar Guroff, G. Google Scholar He Baogang. Google Scholar Horowitz, Donald. Ethnic Groups in Conflict.
Google Scholar Huntington, Samuel. Political Order in Changing Societies. Norman: University of Oklahoma Press. Google Scholar Inglehart, Ronald. Culture Shift in Advanced Industrial Society.
Princeton: Princeton University Press. Google Scholar Inkeles, Alex, ed. New Brunswick: Transaction Publishers. Google Scholar Johnston, Michael. Google Scholar Karl, Terry Lynn. Google Scholar Kazancigil, Ali.
Google Scholar Klitgaard, Robert. Central Europe after Forty Years. Google Scholar Kohli, Atui. Google Scholar Kuznets, Simon. The European Dictatorships. London: Methuen. Google Scholar Lewis, Bernard. Google Scholar Lijphart, Arend. New Haven: Yale University Preess. Google Scholar Linz, Juan J. Baltimore: Johns Hopkins University Press. The Breakdown of Democratic Regimes. Baltimore: Johns Hopkins.
New York: Holmes and Meier. Google Scholar Lipset, Seymour Martin. The First New Nation expanded paperback edition. New York: Norton. Consensus and Conflict: Essays in Political Sociology. New York: Routledge. Seattle: University of Washington Press.
New York: E. Lipset and S. New York: The Free Press. New York: Free Press. Google Scholar Macdonald, Donald S. Papers published from the June Google Scholar Malia, Martin Google Scholar Mann, Michael.
Madrid: Juan March Institute. Center for Advanced Study in the Social Sciences. Google Scholar Marx, Karl. Capital vol. Moscow: Foreign Languages Publishing House.
Google Scholar Merkl, Peter H. Google Scholar Moore, Barrington.
Boston: Beacon Press. Google Scholar Muller, E. Google Scholar Olsen, Marvin E.
Google Scholar Olson, Mancur Jr. Google Scholar Nathan, A. Google Scholar Pye, Lucian W. Pye and S. Google Scholar Pye, Lucian W.. The Spirit of Chinese Politics. Cambridge: MIT Press.
Asian Power and Politics. The Cultural Dimensions of Authority. Cambridge: Harvard University Press. Google Scholar Riggs, Fred. The Duverger laws posit that electoral systems affect that is, reduce or multiply the number of parties, and yet Duverger never indicates how parties are to be counted.
The related point is that a causal relationship is verifiable only if the effect is clearly specified, whereas the effect of the first law party dualism is accordion-like, and the effect of the second law equally suffers from excessive imprecision. When Douglas Rae took up the matter in his influential The Political Consequences of Electoral Laws, he did not build cumulatively on Duverger but switched instead to a different track. While Duverger looked at concrete party systems even though he did not identify them correctly , the quantitative literature that goes from Rae to Rein Taagepera and Matthew Shugart simply leaves us with arbitrary cutting points along a continuum.
And what do these ratios or proportions entail in terms of systemic properties? This is what we are not told. Not only are we left with a pure and simple counting of parties of dubious empirical validity but what is it that we have? And laws can hardly be formulated unless we are clear-headed as to how they relate to causal analysis, to condition analysis, to probability and deter- minism, and, conclusively, to how they are confirmed or disconfirmed.
We mean, I assume, a generalization endowed with explanatory power that detects a regularity that allows predictions. So a law is required to assert more than a regularity and cannot consist of a mere generalization. And the point to bear in mind is that the explanatory power of a law is just as crucial as its predictive power.
Another preliminary point bears on the conditions under which a law applies or does not apply. Hundreds of critics point to hundreds of occasions in which the electoral system fails to produce the predicted effects. But a law, any law, can hold only when it applies and cannot be disconfirmed by cases to which it does not apply. Water boils at o C at the sea level, not at the top of Everest; bodies fall at the same acceleration, regardless of shape and size, in a void.
Likewise, the laws that specify the effects of electoral systems apply only to party systems, that is to say, to the stage at which a loose collection of notables gives way to a structured ensemble of parties. First of all, what do electoral systems affect? At first blush, the direct effect of an electoral system appears to be on the voter.
However, the electoral system also has a direct effect on the number of parties the format of the party system since it establishes how votes are translated into parliamentary seats. Other effects are instead derivative. We shall come to them in due course.
The effect of an electoral system on the voter can be constraining manipulative or unconstraining. If it is unconstraining, then an electoral system has no effect—and that is that. The effect of an electoral system on the party system can either be reductive it reduces or compresses the number of parties or not—and in the latter case we have again a non-effect.
A pure system of proportional representation PR is ineffective in both respects: It does not have a manipulative impact either on the voter or on the format of the party system. Regardless of the mathematics of Giovanni Sartori 93 proportionality, a small district for example, a district that elects two to five members brings about impure proportionality.
The rule of thumb here is that the smaller the district, the lesser the proportionality. Con- versely, the larger the district, the greater the proportionality. Israel and the Netherlands thus rank among the pure PR systems in that they elect, respectively, and MPs in a single, nationwide constituency. The bottom line then is that when electoral systems are effective, they reduce the number of parties. This brings us back to the question of how parties are to be counted.
Duverger, I have already noted, had no answer for this question, and his assessment was not only impressionistic but also utterly erratic. Clearly, parties cannot be counted at their face value. In my work, I have used the notion of relevant party, where relevance is a systemic assessment based on two criteria, namely, the coalition potential or blackmail potential of any given party.
Unfortunately, they miss relevance as I define it since they tell us just about nothing about whether and in what manner a party affects the party system as a whole. Hence, whenever a two-party format is established, a plurality system exerts a brakelike influence. Rule 2: A plurality system will produce, in the long run, a two-party format not the eternalization of the same parties, however under two conditions: first, a party system structured by nationwide parties and, second, if the electorate which is refractory to whatever pressure of the electoral system happens to be dispersed in below-plurality proportions throughout the constituencies.
Rule 4: Finally, PR systems also obtain reductive effects—though to a lesser extent—in proportion to their nonproportionality; and particularly whenever they are applied in small-sized constituencies, establish a threshold of representation, or attribute a premium. Under 94 The Party Effects of Electoral Systems these conditions, PR too will eliminate the smaller parties whose electorate is dispersed throughout the constituencies; but even a highly impure PR will not eliminate small parties that thrive in concentrated above-quota strongholds.
Let me also explain why my rules largely hinge upon the distribution of electorates, and specifically upon knowing whether or not incoercible above-plurality or, as the case may be, above- quota minorities happen to be geographically concentrated or dispersed. This is so because my rules apply only when the stage of local fragmen- tation of politics is over. At this point we are simply left, therefore, with the allegiances that remain unaffected by the advent of a structured party system.
One may ask why the number of relevant parties the format of party systems is so important? This is because the format explains and predicts the mechanics, that is, the systemic characteristics of distinctive types of party systems. In my analysis, I sort out three major systemic patterns: 1 two-party mechanics, that is, bipolar single-party alternation in govern- ment; 2 moderate multipartism, that is, bipolar shifts between coalition governments; 3 polarized multipartism, that is, systems characterized by multipolar competition, center-located coalitions with peripheral turnover, and antisystem parties.
So the question now is: Will the format be followed by the expected, corresponding mechanics functional properties? Given structural consolidation as a necessary condition and polarization as the intervening and, to some extent, dependent variable, I hypothesize as follows: Hypothesis 1: When the single-member plurality formula produces a two-party format Rule 1 and 2 , the format will, in turn, produce a two- party mechanics if and only if the polarization of the polity is low.
With high polarization, the two-party mechanics breaks down.
However, since a two-party mechanics implies centripetal competition, it tends to lessen systemic polarization. This format will, in turn, engender the mechanics of moderate multipartism if and only if the polity does not display high polarization. Hypothesis 3: Relatively pure or pure PR systems easily allow for a five-to-seven-party format. Even so, under conditions of medium-low polarization, the coalitional mechanics of moderate multipartism are not impeded.
Under conditions of high polarization, however, the format will display the mechanical characteristics of polarized multipartism, thereby including a multipolar competition that eventually heightens systemic polarization.
Thus far I account for just one causal factor, the electoral system. Another causal factor at play may be the party system itself. That is to say, a strongly structured and well-entrenched party system performs as a channelling force of its own.
But the substitution of an independent variable with another does not pose any problem to my laws, which can easily incorporate an alternative causal factor. The New Case: From PR back to Plurality My argument and my laws implicitly assume a movement from majoritarian to proportional systems. For this has been, historically, the unfailing direction of change. PR was first introduced in , and no democracy switched back from proportional to plurality elections for the next 70 years.
France did exactly that in Thus the eventfulness of this unprecedented event went unnoticed. But when Italy attempted a similar comeback from PR to plurality in and in , the attempt failed miserably and indeed backfired. So here we have a problem that has yet to be tackled. Why has France succeeded and Italy failed?
The first reason is that the two countries have followed different paths. France astutely adopted a double-ballot plurality system reinforced by increasingly heightened thresholds of exclusion today of Italy has done nothing of the sort. It adopted a single-shot plurality system20 with an insufficient threshold of exclusion, and has thus far not had a directly elected president. Yet the success of the French comeback has obscured the intrinsic difficulties that confront the dismantling of an entrenched pattern of party fragmentation.
Note that in Italy a single-ballot plurality system has not only failed to reduce the number of parties; it has actually backfired in producing more parties, a still higher level of fragmentation. Why did they multiply? This is the question that goes to the heart of the problem. The answers lies, I suggest, in the originally Downsian notion of blackmail party. The reason for this is simple. With a winner-take-all system, victory or defeat may be decided by one or two percentage points.
Thus even very small parties—as long as they have a core of strongly identified voters—may display crucial blackmail leverage. True, with plurality elections a small party cannot win, but it can easily endanger the winning chances of the major parties. Paradoxically, in the case at hand, it is not PR but the winner-take- all system that multiplies parties by extending and facilitating their relevance.
Clearly, then, returns from PR to plurality require a new engineering not covered by current know-how. The new problem is how to cope with a new kind of blackmail potential.
Note that this new problem is entirely created—since all other circumstances remain equal—by the electoral system. It will have to be remedied, therefore, by electoral counter- measures.
Here the firm point is that substituting PR with a single- ballot plurality system will not, in all likelihood, reduce the number of parties. Indeed, this remedy is likely to boomerang. A third warning is that electoral alliances —the French apparentement—should be prohibited whenever an electoral system has thresholds or premiums. For, clearly, alliances circumvent thresholds of exclusion and defy the aggregative intent of majority premiums.
In the case of Israel, for example, I would recommend a PR system that provides majority premiums for the first two parties in this proportion: 20 percent to the first, 15 percent to the second. While this system is especially intended to reinforce the second-place finishers, the minimal size of the constituencies does crush the smaller parties even though Chile permits electoral coalitions. Misunderstood Electoral Systems Up to this point I have dealt with the effects of electoral systems.
But what about the causal factor itself? Is everything clear at this end of the argument? Not really. Moreover, the understanding of electoral systems leaves much to be desired. For a number of voting methods are both misclassified and misunderstood. Electoral systems are fundamentally divided into majoritarian and proportional and thus defined by reciprocal exclusion: All majoritarian systems are not proportional and, conversely, all proportional systems are not majoritarian.
So far, so good. But PR systems have been devised by mathematicians and a respectable mathematician must seek perfect proportionality. Thus mathematicians have ignored ordinal proportionality and have confined PR to equal quotas or quotients , that is, to systems that allocate seats to equal shares of the voting returns.
Assume, however, that we encounter—as was the case in Japan until —four-member constituencies average that elect the first four most voted candidates. What kind of system is that? For Lijphart and others, it is a variety of the limited vote the voter has fewer votes than there are seats in which each voter has only one vote. First of all, it makes no sense at all to wonder whether the Japanese system may be considered a plurality system, for it is certainly not.
Secondly, in all standard PR systems voters have fewer votes than there are seats, and with PR, it is normal for voters to have just one nontransferable vote. Thirdly, under what criterion is Japan best considered semiproportional?
This is a purely impressionistic assessment. PR systems have always been considered more-or-less pure or impure. The answer is no, and therefore this notion has no classificatory value. In my opinion, 98 The Party Effects of Electoral Systems the Japanese system was quite simply an ordinal proportional system characterized by personalized voting in lieu of list voting and by small constituencies and thus of the impure, least proportional variety.
So why does this straightforward understanding of the case escape us? The reason is the mathematical bias that establishes that proportionality can be achieved only via equal quotas, whereas proportionality can also be achieved, I submit, by having candidates elected in multimember constituencies on the basis of the highest portion proportion of the returns. Nor is it a foregone conclusion that ordinal proportionality is necessarily more imperfect impure than quota-based proportionality.
For this matter the degree of correspondence of votes to seats is decided far more by the constituency size than by algorithms. Moving on, double-ballot systems also called two-round systems represent both a neglected and highly misunderstood area of electoral systems. An expert of the stature of Richard Rose assimilates and indeed subordinates the double ballot to the alternative vote. The difference between the Australian and French forms of alternative vote is limited, but of practical importance.
Both systems. Both ask voters to state more than one preference. But the Australian system leaves it to the voter to decide his preference, ordering candidates all at once in a single ballot. By contrast the French system also gives an initiative to the candidates and parties after the results of the first ballot are in.
Firstly, it is the double ballot that has many variants, not the alternative vote. If anything, then, the double ballot should be the genus of which the alternative vote is a species. Anyway, the double ballot is not necessarily a single-member majoritarian-plurality system; it can also be an ordinal proportional method of electing candidates in multimember if small constituencies.