American Political Science Review Vol.87,No.3 contracts.The moral appeal of democracy is now some territory somewhere.In the limit,complete uncertainty almost universally appreciated,but its economic ad- about what territory an autocrat will control implies roving vantages are scarcely understood. banditry.The advantages of stationary banditry over roving banditry are obviously greatest when there are natural and militarily defensible frontiers.Interestingly,the earliest states in history emerged mainly in what one anthropologist calls "environmentally circumscribed"areas,that is,areas of ara- Notes ble land surrounded by deserts,mountains,or coasts (see Carneiro 1970).The environmental circumscription not only I am grateful to the U.S.Agency for International Develop- provides militarily viable frontiers but also limits the oppor- ment for support of my research on this subject through my tunity for defeated tribes to flee to other areas in which they Center for Institutional Reform and the Informal Sector. could support themselves(as Carneiro points out).This in turn means that the consensual democracy characteristic of 1.There is quantitative evidence from an exhaustive sur vey of ethnographic accounts showing that references to the earliest stages of social evolution is,in these geographical conditions,replaced by autocratic states earlier than in other slaves are virtually absent in the accounts of the very most conditions. primitive peoples but rather common in more advanced agricultural 11.For more examples of other types of reason,see Olson societies (Hobhouse,Wheeler, and Ginsberg 1990. 1930).Slavery is unprofitable in hunting-gathering societies 12.In the interest of brevity,democracy is here defined as (Olson 1967). 2.Small tribes can sometimes form federations and competitive elections,social pluralism,and the absence of thereby increase the number who can obtain collective goods autocracy,rather than in terms of universal suffrage.Al- though how a narrower suffrage turns into a wider suffrage through voluntary action (Olson 1965,62-63).Some of the can be explained by straightforward extensions of the logic of very earliest agricultural societies may have been of this the theory offered here,developing these extensions and character.But when the number of small groups itself be comes very large,the large-number problem is evident again testing them against the historical evidence would not be a small undertaking. and voluntary collective action is infeasible 3.For citations to much of the best literature extending 13.For striking evidence on how the growth of cities was and testing the argument in The Logic of Collective Action,as much greater in medieval and early modern Europe in dem- well as for valuable new analyses,see Hardin 1982 and ocratic or less autocratic regimes,see DeLong and Schleifer Sandler 1992 1992.In effect,the DeLong and Schleifer paper is a test of the advantages of democracy that I put forward. 4.This literature is most constructive and interesting,but to the extent to which it tries to explain government in terms of voluntary transactions,it is not convincing.North,while emphasizing transactions costs and contracts,also uses the notion of the "predatory state" and the logic of collective References action in his account of the state,so his approach must be distinguished from Barzel's Banfield,Edward.1958.The Moral Basis of a Backward Society. 5.For the definition of an encompassing interest and Glencoe,IL:Free Press. evidence of its importance,see Olson 1982.The logical Carniero,Robert L.1970."A Theory of the Origin of the structure of the theory that encompassing interests will be State."Science 169:733-38. concerned with the outcome for society whereas narrow Clague,Christopher,and Gordon Rausser,eds.1992.The groups will not is identical with the logic that shows that small Emergence of Market Economies in Eastern Europe.Cambridge: groups can engage in voluntary collective action when large Basil Blackwell. groups cannot DeLong,J.Bradford,and Andrei Schleifer.1992."Princes and 6.Many of the more remarkable advances in civilization Merchants:European City Growth before the Industrial even in historic times took place in somewhat democratic or Revolution."Harvard University.Mimeo. nondictatorial societies such as ancient Athens,the Roman Hardin,Russell.1982.Collective Action.Baltimore:Johns Hop- Republic,the North Italian city-states,the Netherlands in the kins University Press. seventeenth century,and(at least after 1689)Great Britain Hobhouse,L.T.G.C.Wheeler,and M.Ginsberg.1965.The The explanation for the disproportionate representation of Material Culture and Social Institutions of the Simpler Peoples nonautocratic jurisdictions in human progress is presented London:Routledge K.Paul. later in the article. Kalduhn,Ibn.1967.The Mugaddimah.Trans.Franz Rosenthal. 7.The theory offered here applies to communist autocra Princeton:Princeton University Press. cies as much as to other types,though the theory needs to be Kiser,Edgar,and Yoram Barzel.1991."Origins of Democracy elaborated to take account of the"implicit tax-price discrim in England."Journal of Rationality and Society 3:396 ination"pioneered by Joseph Stalin.This innovation enabled Lake,David A.1992."Powerful Pacifists:Democratic States Stalinist regimes to obtain a larger proportion of social output and War."American Political Science Review 86:24-37. for their own purposes than any other regimes had been able Murrell,Peter,and Mancur Olson.1991."The Devolution of to do.This explained Stalin's success in making the Soviet Centrally Planned Economies."Journal of Comparative Eco- Union a superpower and the great military capacity of many 1 omics15:23965. communist regimes.It also generated a unique dependence of North,Douglass.1981.Growth and Structural Change.New the system on its manageme ent cadre,which ultimately proved York:Norton. fatal.For how the offered theory applies to communist autocra North,Douglass,and Robert Thomas.1973.The Rise of the cies and the societies in transition,see Clague and Rausser West. Cambridge:Cambridge University Press. 1992,pref.,chap.4;Murrell and Olson 1991:Olson 1993. Olson,Mancur.1965.The Logic of Collective Action.Cambridge: 8.Schumpeter's analysis is in his"Crisis of the Tax State," Harvard University Press. written in the highly taxed Austria-Hungarian Empire late in Olson,Mancur.1967."Some Historic Variation in Property World War I;Ibn Kalduhn's is in his classic,The Mugaddimah. Institutions."Princeton University.Mimeo. 9.A mathematical and a geometrical proof of this conclu Olson,Mancur.1982.The Rise and Decline of Nations.New sion and an analysis of many other technical questions raised Haven:Yale University Press. by the present theory is available on request Olson,Mancur.1986."A Theory of the Incentives Facing 10.When war erodes confidence about what the bound- Political Organizations:Neo-corporatism and the Hege- aries of an autocrat's domain will be, an autocrat's time monic State." International Political Science Review 7:165-89 horizon with respect to his possession of any given territory Olson,Mancur.1990."The Logic of Collective Action in shortens- -even if he believes that he will remain in control of Soviet-type Societies."Journal of Soviet Nationalities 1(2):8-33. 575 This content downloaded from 202.120.14.67 on Sun,19 Feb 2017 15:11:57 UTC All use subject to http://about.jstor.org/termsAmerican Political Science Review Vol. 87, No. 3 contracts. The moral appeal of democracy is now almost universally appreciated, but its economic ad- vantages are scarcely understood. Notes I am grateful to the U.S. Agency for International Develop- ment for support of my research on this subject through my Center for Institutional Reform and the Informal Sector. 1. There is quantitative evidence from an exhaustive sur- vey of ethnographic accounts showing that references to slaves are virtually absent in the accounts of the very most primitive peoples but rather common in more advanced agricultural societies (Hobhouse, Wheeler, and Ginsberg 1930). Slavery is unprofitable in hunting-gathering societies (Olson 1967). 2. Small tribes can sometimes form federations and thereby increase the number who can obtain collective goods through voluntary action (Olson 1965, 62-63). Some of the very earliest agricultural societies may have been of this character. But when the number of small groups itself be- comes very large, the large-number problem is evident again and voluntary collective action is infeasible. 3. For citations to much of the best literature extending and testing the argument in The Logic of Collective Action, as well as for valuable new analyses, see Hardin 1982 and Sandler 1992. 4. This literature is most constructive and interesting, but to the extent to which it tries to explain government in terms of voluntary transactions, it is not convincing. North, while emphasizing transactions costs and contracts, also uses the notion of the "predatory state" and the logic of collective action in his account of the state, so his approach must be distinguished from Barzel's. 5. For the definition of an encompassing interest and evidence of its importance, see Olson 1982. The logical structure of the theory that encompassing interests will be concerned with the outcome for society whereas narrow groups will not is identical with the logic that shows that small groups can engage in voluntary collective action when large groups cannot. 6. Many of the more remarkable advances in civilization even in historic times took place in somewhat democratic or nondictatorial societies such as ancient Athens, the Roman Republic, the North Italian city-states, the Netherlands in the seventeenth century, and (at least after 1689) Great Britain. The explanation for the disproportionate representation of nonautocratic jurisdictions in human progress is presented later in the article. 7. The theory offered here applies to communist autocra- cies as much as to other types, though the theory needs to be elaborated to take account of the "implicit tax-price discrim- ination" pioneered by Joseph Stalin. This innovation enabled Stalinist regimes to obtain a larger proportion of social output for their own purposes than any other regimes had been able to do. This explained Stalin's success in making the Soviet Union a superpower and the great military capacity of many communist regimes. It also generated a unique dependence of the system on its management cadre, which ultimately proved fatal. For how the offered theory applies to communist autocra- cies and the societies in transition, see Clague and Rausser 1992, pref., chap. 4; Murrell and Olson 1991; Olson 1993. 8. Schumpeter's analysis is in his "Crisis of the Tax State," written in the highly taxed Austria-Hungarian Empire late in World War I; Ibn Kalduhn's is in his classic, The Mugaddimah. 9. A mathematical and a geometrical proof of this conclu- sion and an analysis of many other technical questions raised by the present theory is available on request. 10. When war erodes confidence about what the bound- aries of an autocrat's domain will be, an autocrat's time horizon with respect to his possession of any given territory shortens-even if he believes that he will remain in control of some territory somewhere. In the limit, complete uncertainty about what territory an autocrat will control implies roving banditry. The advantages of stationary banditry over roving banditry are obviously greatest when there are natural and militarily defensible frontiers. Interestingly, the earliest states in history emerged mainly in what one anthropologist calls "environmentally circumscribed" areas, that is, areas of ara- ble land surrounded by deserts, mountains, or coasts (see Cameiro 1970). The environmental circumscription not only provides militarily viable frontiers but also limits the oppor- tunity for defeated tribes to flee to other areas in which they could support themselves (as Cameiro points out). This in turn means that the consensual democracy characteristic of the earliest stages of social evolution is, in these geographical conditions, replaced by autocratic states earlier than in other conditions. 11. For more examples of other types of reason, see Olson 1990. 12. In the interest of brevity, democracy is here defined as competitive elections, social pluralism, and the absence of autocracy, rather than in terms of universal suffrage. Al- though how a narrower suffrage turns into a wider suffrage can be explained by straightforward extensions of the logic of the theory offered here, developing these extensions and testing them against the historical evidence would not be a small undertaking. 13. For striking evidence on how the growth of cities was much greater in medieval and early modem Europe in dem- ocratic or less autocratic regimes, see DeLong and Schleifer 1992. In effect, the DeLong and Schleifer paper is a test of the advantages of democracy that I put forward. References Banfield, Edward. 1958. The Moral Basis of a Backward Society. Glencoe, IL: Free Press. Carniero, Robert L. 1970. "A Theory of the Origin of the State." Science 169:733-38. Clague, Christopher, and Gordon Rausser, eds. 1992. The Emergence of Market Economies in Eastern Europe. Cambridge: Basil Blackwell. DeLong, J. Bradford, and Andrei Schleifer. 1992. "Princes and Merchants: European City Growth before the Industrial Revolution." Harvard University. Mimeo. Hardin, Russell. 1982. Collective Action. Baltimore: Johns Hop- kins University Press. Hobhouse, L. T., G. C. Wheeler, and M. Ginsberg. 1965. The Material Culture and Social Institutions of the Simpler Peoples London: Routledge & K. Paul. Kalduhn, Ibn. 1967. The Mugaddimah. Trans. Franz Rosenthal. Princeton: Princeton University Press. Kiser, Edgar, and Yoram Barzel. 1991. "Origins of Democracy in England." Journal of Rationality and Society 3:396 Lake, David A. 1992. "Powerful Pacifists: Democratic States and War." American Political Science Review 86:24-37. Murrell, Peter, and Mancur Olson. 1991. "The Devolution of Centrally Planned Economies." Journal of Comparative Eco- nomics 15:239-65. North, Douglass. 1981. Growth and Structural Change. New York: Norton. North, Douglass, and Robert Thomas. 1973. The Rise of the West, Cambridge: Cambridge University Press. Olson, Mancur. 1965. The Logic of Collective Action. Cambridge: Harvard University Press. Olson, Mancur. 1967. "Some Historic Variation in Property Institutions." Princeton University. Mimeo. Olson, Mancur. 1982. The Rise and Decline of Nations. New Haven: Yale University Press. Olson, Mancur. 1986. "A Theory of the Incentives Facing Political Organizations: Neo-corporatism and the Hege- monic State." International Political Science Review 7:165-89. Olson, Mancur. 1990. "The Logic of Collective Action in Soviet-type Societies." Journal of Soviet Nationalities 1(2): 8-33. 575 This content downloaded from 202.120.14.67 on Sun, 19 Feb 2017 15:11:57 UTC All use subject to http://about.jstor.org/terms