elite settlements

7/27/2019 Elite Settlements

1/14

Elite SettlementsAuthor(s): Michael G. Burton and John HigleySource: American Sociological Review, Vol. 52, No. 3 (Jun., 1987), pp. 295-307Published by: American Sociological AssociationStable URL: http://www.jstor.org/stable/2095351 .

Accessed: 27/10/2013 13:47

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of

content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms

of scholarship. For more information about JSTOR, please contact [email protected].

.

American Sociological Association is collaborating with JSTOR to digitize, preserve and extend access to

American Sociological Review.

http://www.jstor.org

This content downloaded from 148.206.159.132 on Sun, 27 Oct 2013 13:47:06 PMAll use subject to JSTOR Terms and Conditions
http://www.jstor.org/action/showPublisher?publisherCode=asahttp://www.jstor.org/stable/2095351?origin=JSTOR-pdfhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/stable/2095351?origin=JSTOR-pdfhttp://www.jstor.org/action/showPublisher?publisherCode=asa


2/14

ELITESETTLEMENTS*MICHAEL G. BURTON JOHN HIGLEYLoyolaCollege nMaryland Universityf Texas t Austin

We examine the elite settlement as a major, yet largely overlooked, form of politicalchange. Elite settlements consist of broad compromises among previously warring elitefactions, resulting in political stability and thusproviding a necessary precondition forthe sustained practice of representative democracy. To identify the commonfeatures ofelite settlements, we draw uponfour historic cases: England in 1688-89, Sweden in1809, Colombia in 1957-58, and Venezuela in 1958. We conclude by arguing for theextraction of elite settlements fiom their current embeddedness in such concepts asbourgeois revolutions and democratic transitions, and we advocate greater attention tothe elite paradigm in efforts to explain macropolitical outcomes.

ELITE SETTLEMENTSUnderstanding how nations become politicallystable and democratic is a central goal ofpolitical sociology and a matter of obviousconcern to policymakers around the world.Current scholarly opinion on the subject differsmarkedly from the once-dominant moderniza-tion perspective, which saw stable democracy asa happy by-product of social, economic, andcultural development. Many scholars now viewthe establishment and maintenance of demo-cratic institutions as decidedly political acts.These scholars divide broadly into two camps:adherents to the class or Marxian paradigm andthose working within the elite or managerialparadigm (Alford and Friedland 1985). Thecurrently more influential class paradigm, de-spite many intricate nuances, essentially inter-prets stable democracy as established andmaintained by the bourgeoisie, primarily for thebourgeoisie. This paradigm has customarilyexplained the exercise of bourgeois powerthrough the democratic state in terms of basiceconomic processes. Responding to the limita-tions of such economic determinism, however, anumber of Marxian scholars now stress therelative autonomy of political actors, typicallyaggregated as "the state."This latter trendhas broughtMarxiantheorists

closer to the elite paradigm's core contentionabout the independence and centrality of explic-itly political actors, or elites. But how does theelite paradigm explain the origins of stabledemocracy? The answer is not at all clear.Although a good deal of valuable work on therelationship between elites and democracy hasbeen done (for useful overviews see Putnam1976, pp. 129-32; Peeler 1985, pp. 4-41),hardly anyone has located this work explicitlywithin the elite paradigm. Consequently, theparadigm has not been elaborated in thisdirection and, not surprisingly, research on therole of elites in the origins of democracy has notbeen especially cumulative.Working explicitly within the elite paradigm,we attack this problem by drawing on pertinentliteratureand on an examination of the originsof stable democracy in several countries todevelop the concept of elite settlements as oneroute to stable democracy. Elite settlements arerelatively rare events in which warring nationalelite factions suddenly and deliberately reorga-nize their relations by negotiating compromiseson their most basic disagreements. Elite settle-ments have two main consequences: they createpatternsof open but peaceful competition, basedon the "norm of restrained partisanship"(Manley 1965; Di Palma 1973), among allmajorelite factions; and they transform unstablepolitical regimes, in which irregularseizures ofgovernment executive power by force arefrequent or widely expected occurrences, intostable regimes, in which forcible power seizuresno longer occur and are not widely expected.These changes in elite behavior and regimeoperation pave the way for, though they do notguarantee, the emergence of democratic politics.In many ways, elite settlements are asconsequential as social revolutions, yet theyhave not been systematically studied as adiscrete class of events. There is some relevantliterature, however, that stresses the importance

* Direct all correspondenceto Michael G. Burton,Departmentof Sociology, Loyola College in Maryland,4501 N. Charles St., Baltimore,MD 21210.This is a revision of a paperpresentedat the AmericanSociological Association Meetings, New York City,August 30-September 3, 1986. Work on this paper wasfacilitated by a National Endowmentfor the HumanitiesSummer Stipend(No. FT-26396-85) awarded o Burtonin 1985, and by a Loyola College Summer Researchgrant to Burton in 1986. We thank Joan K. Burton,Moshe M. Czudnowski, G. William Domhoff, JackGoldstone, Richard Gunther,G. Lowell Field, ClarenceStone, and three anonymous reviewers for their com-ments on earlierversions of this paper.

AmericanSociological Review, 1987, Vol. 52 (June:295-307) 295

http://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsp


3/14

296 AMERICAN SOCIOLOGICAL REVIEWof elite unification or accommodation in transi-tions to democracy (e.g., Rustow 1970; Levine1978; Wilde 1978; Linz 1978; Karl 1981;Huntington 1984; Peeler 1985). We intend tobuild on this literature by examining elitesettlements as one especially importantmode ofelite unification, thereby shifting attention fromthe establishment of democratic institutions tothe empirically distinct, causally prior, circum-stances and actions of elites. First, we want tolocate the concept of an elite settlement within abroader set of concepts and assumptions abouthow elite structures vary and with whatconsequences for major political outcomes.Second, we want to draw on four especiallydramatic and seminal elite settlements-En-gland in 1688-89, Sweden in 1809, andColombia and Venezuela in the late 1950s-tospecify their common features. Third, we wantto discuss certain implications that the focus onelite settlements has for theories of politicalchange.ELITES, POLITICAL STABILITY,AND DEMOCRACYThe concept of elite settlements comprises anextension and modification of classical elitetheory as developed by Mosca (1939) andPareto(1935). At the heart of the theory was thecontention of elite variability; that is, that elitestructureand behavior vary significantly amongsocieties and within them over time; that thesevariations occur independently of social, eco-nomic, and cultural forces; and that elitevariations have importantdeterminate effects forthe character of political regimes. As Mosca putit (1939, p. 51), "The varying structure ofruling classes has a preponderantimportance indeterminingthe political type, and also the levelof civilization, of the different peoples." Pareto(1935, esp. pars. 2274-77) was similarlyconcerned with specifying variations amongelites according to the mix of nonlogical"sentiments"that ostensibly guide their thinkingand behavior, and he tied such variations todifferent kinds of political regimes. But neithertheorist got far in developing the contention ofelite variability, and certainlyneither focused onelite settlements as one of the most importantinstances of it. The variability contention,therefore, constitutes the point of departure forexamining elite settlements, but in most otherrespects one must start anew.First, to what does the elite concept itselfrefer?In line with prevailing definitions of elites(Burton and Higley 1987), we avoid assump-tions about talents, moral qualities, degrees ofconsciousness and cohesion, or other properties.Elites are simply people who are able, throughtheir positions in powerful organizations, to

affect national political outcomes individually,regularly, and seriously. Elites thus constitute anation's top leadership in all sectors-political,governmental, business, trade union, military,media, religious, and intellectual-includingboth "establishment" and "counterelite" fac-tions. A national elite can be said to encompass"all those persons capable, if they wish, ofmaking substantial political trouble for highofficials (i.e., other elite persons who happen tobe incumbents of authoritative positions) with-out being promptly repressed" (Field and Higley1973, p. 8).Although the subject of elite variation meritsmuch more careful examination than it hasreceived, there is loose scholarly agreement thatnational elite structurestake three basic forms inthe modem world. By "structure"we mean theamalgam of attitudes, values, and interpersonalrelations among factions making up the elite.One type of elite structure, variously labeled"divided" (Beck and Malloy 1964), "competi-tive" (Putnam 1976), or "disunified" (Field andHigley 1985), is characterizedby ruthless, oftenviolent, interelite conflicts. Elite factions deeplydistrust each other, interpersonal relations donot extend across factional lines, and factions donot cooperate to contain societal divisions or toavoid political crises. A second type, termed"totalitarian" (Dahrendorf 1969), "mono-cratic" (Fleron 1969), or "ideologically uni-fied" (Field and Higley 1985), is characterizedby the outward appearance of nearly completeunity in that all elite factions publicly professthe same ideology andpublicly supportthe samemajor policies. Moreover, all or nearly all elitepersons are members of the same party ormovement and their interpersonal relations aresharply centralized in this party or movement.The third type of elite structure, called "plural-istic" (Fleron 1969), "competitive-coalescent"(Putnam 1976), or "consensually unified"(Field and Higley 1985), displays substantial,but much less than monolithic, unity. Elitefactions regularly take opposing ideological andpolicy stances in public, but they consistentlyrefrain from pushing their disagreements to thepoint of violent conflict. Although they inveighagainst each other on policy questions, theyapparently share a tacit commitment to abide bycommon codes of political conduct centering onthe norm of restrainedpartisanship, and there isan extensive web of interpersonal relationshipsthat encompasses all factions and providessatisfactory access to key decision makers(Higley and Moore 1981).There is also loose scholarly agreement thateach of these major forms of elite structure isclosely associated with a distinctive type ofpolitical regime. Thus, divided or disunifiedelites operate unstable regimes in which coups,



4/14

ELITE SETTLEMENTS 297uprisings, revolutions, and other forcible sei-zures of government power occur frequently orare widely expected. Although representativedemocratic politics may be practiced intermit-tently in such unstable regimes, it usually breaksdown in the face of a political crisis. Totalitarianor ideologically unified elites, on the otherhand, operate stable, politically unrepresentativeregimes in which overt coups or other forciblepower seizures do not occur, and publicconflicts of interest and opinion are consistentlyrepressed in favor of some official ideology.Though institutional functioning may be for-mally democratic, elections and other represen-tative processes are not seriously competitive ordeterminative of government personnel andpolicies. Finally, pluralistic or consensuallyunified elites operatestable, politically represen-tative regimes in which the incumbency of topgovernment positions passes peacefully amongdifferent persons and factions according torepresentative principles and processes, mostnotably periodic, competitive, and bindingelections. However, the precise degree ofpolitical representationdiffers according to theextent of regional, ethnic, religious, or othersubnational conflicts, the (non)existence ofexternal threats, and the extent of economicprosperity or other facilitative conditions. Thustwo of the three elite types-the ideologicallyunified and the consensually unified-are asso-ciated with stable regimes. But only regimesoperatedby a consensually unified elite involveimportant degrees of sustained representativedemocratic politics.Insofar as these associations between elitetype and regime type appear to be widespreadboth in history and the contemporary world(Field and Higley 1980, 1985), one can say thata consensually unified elite is a precondition for,but not a guarantee of, stable democraticpolitics. The origins of this type of national eliteare therefore highly relevant to assessing thelikelihood of democratic transitions. Without abasic change to a consensually unified elite therecan be no lasting transition from an unstableand, at best, only intermittently democraticregime, or from a stable but politically unrepre-sentative regime, to a stable regime in which thesustained practice of representative democracyis a real possibility. In what circumstances,then, do consensually unified elites originate?

Except where a country has been defeated ininternational war (e.g., Germany, Italy, andAustria in World War II), no ideologicallyunified elite has ever been transformed into aconsensually unified elite. The effects ofinternationalwarfareaside, ideologically unifiedelites and the stable, unrepresentative regimesthey operate appear to be reliably self-perpetuating (e.g., Bialer 1980). So the origins

of consensually unified elites and stable,representative regimes must be sought in othercircumstances. Colonial experience is the mostobvious and most widely discussed of these(Rustow 1970; Huntington 1984). Consensuallyunified elites have most frequently originated inthe habituation of major elite factions to openbut peaceful competition while their society isstill a colony or territorial dependency. Byoperating representative political institutionsunder some form of "home rule," or by keepinga large and complex independence movementintact politically, or both, national elites in asignificant minority of former colonies emergedas consensually unified upon attaining indepen-dence. An early example is the elites of Hollandand certain other Dutch provinces when theyemerged from Spanish domination toward theend of the sixteenth century. Otherexamples arethe United States, Canada, New Zealand,Australia, Ireland, India, and Malaysia, allformer colonies of Britain, as well as, from theFrench colonial empire, Tunisia, the IvoryCoast, and possibly Senegal. More or lessimmediately after these countries became inde-pendent, the existence of consensually unifiedelites operating stable political regimes permit-ted importantdegrees of sustained representativedemocratic politics.A second, less frequent, origin of this elitetype appears to involve the gradual attenuationof radical, antisystem stances among one ormore major factions in a disunified nationalelite. Over a period of two or three decades,more specifically, radical elite factions discoverthere is nothing approaching majority supportfor their programs and no real chance of takingpower forcibly. As in the cases of theonce-radical Social Democratic elite factions inDenmark and Norway earlier in this century,and of the until recently intransigent Communistand Socialist elite factions in Italy, France,and Japan, doctrines and programs that hamperthe mobilization of electoral majorities areprogressively abandoned and replaced by prom-ises to defend existing political institutions andto abide by existing rules of the political game.With this moderation of radical left factions,right-wing factions "relax," feeling increasinglycertain that their basic interests are no longerthreatened. In such cases, the national elitegradually becomes consensually unified so that,where representative democratic politics was atbest a precarious tradition, it becomes a secureone.Neither of these two origins of elite consensusand unity, and thus routes to democraticpolitics, appears likely or even possible in manycontemporary countries, however. The disman-tling of colonial empires after World War II,which in most instances left the former colonies



5/14

298 AMERICAN SOCIOLOGICAL REVIEWin the hands of disunified elites, has essentiallyended prospects for the colonial-experienceorigin. And the other origin, involving a gradualattenuation of elite radicalism, apparently de-pends on a society achieving a level oforganizational and economic complexity suffi-cient to discredit radical egalitarian alternativesin the minds of a majority of voters-obviouslynot a current or foreseeable condition in mostnon-Western countries. So unless elite consen-sus and unity originate in still another way,political stability and sustained democraticpolitics will not extend much beyond theirpresent locations in the world. Much, therefore,depends on elite settlements, the third origin ofconsensually unified elites, in societies withdisunified elites and unstable regimes.COMMON FEATURES OFELITE SETTLEMENTSWe think elite settlements have occurredin four,and perhaps in half a dozen or more, countriesduring the modem period. A number ofsettlements were also attempted but failed, forexample, the efforts of Italian, Portuguese, andSpanish elites to constructbroad coalitions or totake turns in government office during the latterpart of the nineteenth century, and similarefforts by Uruguayanand Colombian elites earlyin this century. Such failed elite settlementsdeserve closer scrutiny, but here we focus onfour clearly successful settlements: England in1688-89, Sweden in 1809, Colombia in1957-58, and Venezuela in 1958. We examinedthe relevant historical and social scientificliterature on these countries in considerabledetail, and what we learned constitutes the basisof the following discussion.' A less detailedexamination of the relevant literature on Mexicoin the late 1920s and early 1930s (e.g., Purcelland Purcell 1980; Tardanico 1982), Costa Ricain 1948 (e.g., Montealegre 1983; Peeler 1985),Austria before 1955 (e.g., Steiner 1972; Stief-bold 1974), and Spain in the late 1970s (e.g.,Maravall 1982; Gunther 1985; Gunther et al.1986) turned up indications of possible elitesettlements that warrant furtherinvestigation.Let us first consider the gross historicalcircumstances that motivate disunified elites toenter into settlements. These appearto be of two

basic kinds. One is the recent elite experience ofcostly, but also essentially inconclusive, con-flict. Precisely because no single faction hasbeen a clear winner, and all factions have morenearly been losers, elites are disposed tocompromise if at all possible. The recentexperience of civil war, which entailed consid-erable elite fratricide but had no clear victor, inEngland during the 1640s and in Colombia after1948 exemplifies this kind of circumstance.Bloodied but not wholly bowed, English Toriesand Whigs and Colombian Conservatives andLiberals had, for the moment at least, nostomach for more fighting (Schwoerer 1981;Wilde 1978). Moreover, the unleashing ofleveling social revolutionary tendencies in bothcivil wars made the two national elites keenlyaware that continued conflict could ultimatelycost them their tenure. Whereas no clear civilwar preceded the elite settlements in Swedenand Venezuela, national elites in both countrieshad experienced several decades of intense butinconclusive struggles for factional ascendancy,struggles that were accompanied by indicationsof the potential for leveling tendencies to takecontrol: A peasant uprising and march onStockholm during the ongoing struggle betweenthe two major elite factions, the Hats and theCaps, in 1743; and, in Venezuela, mass protestsagainst the dictatorship of Marcos Perez Jime-nez during 1956-57, combined with an increas-ingly mobilized working class and peasantry.The second kind of circumstance thatdisposeselites to seek a settlement is the occurrence of amajor crisis, which provokes elite action. Suchcrises appear usually to center on the incumbenthead of state and are the culmination of his orher policy failures, power abuses, and demon-strated personal weaknesses, made manifest bya particular action or event that brings elitediscontent to a boil. In England the crisis wasthe news that King James II would have aCatholic heir, a development that climaxedbitter resistance on the partof the predominantlyProtestant elite to James's aggressively pro-Catholic policies and that occurred in thecontext of growing elite alarm about the possibil-ity of an alliance between the Catholic kings ofEngland and France. In Sweden the crisisinvolved the loss of Finland to Russia in 1808,impending Russian and Danish-French inva-sions of Sweden proper, and economic disarray,all of which were viewed by elites as outcomesof King Gustav IV Adolf's ill-consideredpolicies and personal failings (Brown 1895).The crises in Colombia and Venezuela weresharpeconomic downturns punctuatedby effortsof the military dictators Rojas Pinilla and PerezJimenez, respectively, to extend their tenures.In Colombia the crisis surrounding Rojas wasmade particularly intense by continuing civil

' To conserve space, we will provide citations forinterpretivestatementsonly. Our principal sources forfactual statementsare: for England, Goldstone (1986),Hill (1972), Horowitz (1977), Jones (1972), Plumb(1967), Schwoerer(1981), Trevelyan (1938); for Swe-den, Brown (1895), Elstob (1979), Scott (1977); forColombia,Dix (1980), Hartlyn 1984), Wilde (1978); forVenezuela, Blank (1973), Burggraaff (1972), Karl(1981), Levine (1978), Martz(1966).



6/14

ELITE SETTLEMENTS 299strife. In each of the four cases, in short, a crisispartlybroughtabout and made intolerableby theincumbent head of state's blunders and ambi-tions motivated elites not only to remove himand to exclude the clique closely associated withhim, but, more important, to transform thesystem that produced him.Once such circumstances motivate elites toseek a settlement, the ensuing process hasseveral common features. One is speed. Itappears that elite settlements are accomplishedquickly or not at all. Triggered by a seriouspolitical crisis that threatens renewed elitewarfare, settlements involve intensive efforts tofind a way out. Fear of the consequences of notdoing so loosens the fixed positions andprinciples of various factions and disposes themto consider concessions that in other circum-stances they would not countenance. In none ofthe cases under discussion did a settlement takemuch longer than a year. The coming togetherof Tory and Whig factions in England began inearnest during the first half of 1688 with aconspiracy among key Tory and Whig leadersand the Dutch stadtholder, Prince William ofOrange, to unseat King James II. The keycomponents of the settlement were agreed to bythe major factions less than a year later, inFebruary 1689. The Swedish settlement waseven more rapid, involving a similar eliteconspiracy against the king during the winter of1808-9, followed by the drafting and accep-tance of a new constitution during May andJune 1809. In Colombia the overthrow of RojasPinilla was orchestratedby a coalition of Liberaland Conservative party leaders between July1956 and the following May. The constitutionalcomponents of the Colombian settlement werenegotiated by the same coalition from July toOctober 1957 and overwhelmingly approved ina plebiscite two months later. The Venezuelansettlement got under way with a meeting of theheads of the three major parties and twobusiness leaders in New York City in December1957; the settlement agreements were finalizedexactly a year later.This is not to suggest that an elite settlementbecomes complete and secure in such a shorttime; rather,creationof its essential componentsis accomplished rapidly. One must distinguishbetween the initial, basic settlement and itssubsequent consolidation. In each of the casesbeing considered, for example, the new rules ofthe political game embodied in the settlementfaced dangerous challengers: the Jacobites, whowanted to return the Stuarts to the Englishthrone, the attempt to organize a royalistcountercoup in Sweden in 1810, coup attemptsby supporters of the ousted Rojas Pinilla inColombia, a leftist guerrilla insurgency inVenezuela during the early 1960s. Extending

over several years, possibly a generation, thesudden and deliberate elite cooperation thatmakes a settlement possible in the first placemust be sustained to thwart such challenges if asettlement is to be fully consolidated.A second feature of the settlement process isface-to-face, partiallysecret, negotiations amongparamount leaders of the major elite factions.Through a combination of skill, desperation,and accident, impasses are broken and crucialcompromises are struck. Such meetings mustnumber in the scores, even hundreds, in eachcase, for a settlement not only requires negoti-ating compromises between major factions butalso within them. Indeed, in the cases underdiscussion one gets the sense of an almostcontinuous round of secret meetings and consul-

tations.The settlement process in England originatedin 1687 with secret meetings among a smallgroup of key Tory and Whig leaders and aDutch adviser to Prince William. These meet-ings gave rise to William's invasion in Novem-ber 1688, and they produced the broad outlinesof the new political system he helped establish.Further meetings among the principal actorsgenerated the decision to hold a specialparliamentary convention to address the issuesstill in conflict: Who would be king, or queen?What would be the line of succession? Whatwould be the relative powers of the monarchyand parliament? What would be the nation'sreligious posture? This three-week convention,itself a flurry of secret meetings, produced theDeclaration of Rights, the formal expression ofthe elite settlement, which William and Maryaccepted at their coronation of February 13,1689. In Sweden, two weeks of intensive, secretdeliberations among a 15-man committee, plusits pivotal secretary Hans Jarta, produced theconcessions and draft constitution that were thenratified in three more weeks of discussion by thefour Estates of nobles, clergy, merchants, andfree farmers. One of the earliest importantmeetings in the Colombian settlement occurredin Spain in July 1956 between just two people:Laureano Gomez, the exiled former presidentand still leader of a major Conservative partyfaction, and Alberto Lleras, a former presidentand leader of the Liberal party. The two metagain in Spain in July 1957, signing the Pact ofSitges, which set the frameworkfor a successionof talks within and between party factions fromJuly to October 1957. The result was theNational Front platform for constitutional re-form, which was overwhelmingly approved inthe December 1957 plebiscite. The broadoutlines of the Venezuelan settlement wereshaped in the New York City meeting inDecember 1957 among three party heads andtwo business leaders, and the written expres-



7/14

300 AMERICAN SOCIOLOGICAL REVIEWsions of the settlement-the Pact of Punto Fijoand the Statement of Principles and MinimumProgram of Government-were fashioned inmeetings at the home of a party leader, RafaelCaldera of the Christian Social Party (COPEI),during the fall of 1958.Such formal, written agreements constituteanother common feature of elite settlements.Written agreements commit elite factions pub-licly to the concessions and guarantees theyhave made. In all four cases, the mostimmediately visible manifestation of an elitesettlement was one or more constitutionaldocuments: A Declaration of Rights and then aBill of Rights in England; a new constitution assuch in Sweden; a fourteen-pointNational FrontAgreement, which was incorporated into theconstitution in Colombia; the Pact of Punto Fijoand the Statement of Principles and MinimumProgram of Government, which were incorpo-rated into the Venezuelan constitution in 1961.But formal agreements and constitutions bythemselves hardly suffice to produce the com-mon elite acceptance of a new code of politicalconduct, which is the most fundamental andlasting consequence of an elite settlement.Behind such agreements there must be a greatdeal of forbearance and conciliatory behavioramong the most central elite actors. By theirnature, historical records contain few indicationsof these subtle retreats from intransigence andenmity, but important examples can be dis-cerned in all four settlements.English legal experts at the time generallyagreed that the Declaration of Rights, acceded toverbally by William when he and Mary werecrowned, was not legally binding. As king,William could have ignored the restrictionsimposed on his authority. Yet he honored themand acquiesced to further restrictions in late1689. Very importantly, he also distributedoffices among his supporters to achieve abalance between Tories and Whigs (Jones 1972,pp. 31-32). Continuing to act in the spiritof thesettlement, William accepted additional restric-tions during his reign: annual parliamentarysessions became the norm even though notrequired by law, and the House of Commonsgradually assumed a significant role in foreignpolicy, though this was the crown's prerogative.Similarly in Sweden, the interim king regent,Karl XIII, uncle of the deposed Gustav IVAdolf, refused to support efforts to organize aroyalist countercoup in 1810, thereby givingleaders of the 1809 settlement vital time toconsolidate the new regime. And the crownprince, Bernadotte, recruited from France tobecome Sweden's new king, agreed to delay hisascendance to the throne for a full eight years soas to ensure a gradual and peaceful transitionfrom the old order to the new. In Colombia, the

pressing question of whether the Liberal-Conservative coalition-which had agreed to afifty-fifty split of all government offices forsixteen years- should have a Conservative or aLiberal as its first presidential candidate wasresolved through informal agreements amongthe factions just ten days before the 1958election. In Venezuela, almost three yearspassed before the terms of the elite settlementwere given legal status in the constitution. Butthough not legally bound to do so, RomuloBetancourt, the new president, immediatelyevidenced his commitment to power sharing byappointing members of the two majoroppositionparties to his cabinet, and he moved in otherways to create a climate favorable to thoseparties.

Another notable feature of the settlementprocess is the predominance of experiencedpolitical leaders; "new men" play only periph-eral roles. In England, the instigators of PrinceWilliam's invasion, the members of the parlia-mentaryrights committees, and William and hisadvisers were all veterans of many previousconflicts. In Sweden, Hans Jarta and themembers of the 15-man constitutional commit-tee had been politically active for severaldecades. In Colombia, the leading negotiatorsofthe settlement, Laureano Gomez and AlbertoLleras, were both former presidents; they andmost other principal actors had been involved inthe failed effort to form a Liberal-Conservativecoalition a decade earlier. In Venezuela, thecentral negotiators were the Democratic Action,Christian Social, and Democratic RepublicanUnion party leaders, each with at least 20 yearsof political experience.Political experience probably facilitated thesettlements in several ways. Through pastexperience these leaders had developed powerand prestige, which were essential resources asthey pushed their plans forward. Also, theseindividuals had deep knowledge of the issuesand of how their political systems worked; theyknew what had to be done and how to go aboutdoing it. In addition, past experience allowedfor political learning (Levine 1978, p. 103).After suffering from previous conflicts, eliteswho were once intransigent could more clearlysee the value of compromise.In addition to these common processualfeatures of speed, face-to-face negotiations,formal agreements, informal forbearance, andexperienced leadership, do elite settlementsshare some more clearly structuralfeature? It isprobably significant that at the time English,Swedish, Colombian, and Venezuelan elitesachieved settlements, all four countries were at arelatively low level of socioeconomic develop-ment. England in the late seventeenth centuryand Sweden in the early nineteenthcentury were



8/14

ELITE SETTLEMENTS 301predominantly rural, agrarian societies. Al-though Colombia and Venezuela were substan-tially more urbanized in the 1950s than wereEngland and Sweden at the time of theirsettlements, neither of the Latin Americancountries was highly industrialized. Togetherwith some special circumstances of Colombianand Venezuelan elites, mentioned below, thissuggests that all four national elites enjoyedconsiderable autonomy from mass followingsand pressures. Elite factions and their leaderscould compromise on questions of principlewithout strongpressuresto standfirm. Membersof traditional oligarchies rather than leaders oflarge and complex mass organizations andmovements, the four elites were comparativelyfree to make the concessions and deals that elitesettlements require.Outwardly at least, Colombian and Venezue-lan elites seem to have possessed less autonomythan the elites of preindustrial England andSweden. Most key Colombian and Venezuelanactors led organized political parties, and theywere presumably constrained by calculations ofelectoral costs, party splits and the like. Yet,under the dictatorial regimes of both countriesduring the 1950s, those parties were hardlyvital, full-bodied mass organizations. Indeed,several party leaders were in exile at the time,and it is probably of no small consequence thatsome of the meetings that produced theColombian and Venezuelan settlements tookplace abroad-in Spain, in Puerto Rico, in NewYork City. In short, the absence of full-scaleindustrialization in Colombia and Venezueladuring the 1950s, combined with the partiallyrepressed situation of parties and other massorganizations in those countries, implies that,like English and Swedish elites, elites inColombia and Venezuela also enjoyed substan-tial autonomy.This is not to suggest, however, that eliteautonomy is ever total, that elites fashion theirsettlements without regard for nonelite reac-tions. We have already mentioned elite fears ofleveling sentiments as a prod to quick action.Beyond this, mobilized nonelites frequentlyserve as resources, whether in bringing down agovernment or in defending a particularpositionas elites jockey toward compromises they canlive with. Indeed, elites in each of the casesunder discussion demonstrated substantial con-cern with public opinion. Even in their day, theEnglish elites who plotted William's invasionand subsequently worked out the rules underwhich he would be king felt compelled to mounta huge public relations campaign announcingand defending their actions. Their principalopponent, the royalist faction around KingJames II, responded with its own campaign forpublic support. Public discussion of the settle-

ment process was further informed by leaksfrom secret meetings about who took whatposition. Similar patterns appear in the othercases. This public aspect of elite settlements isalso seen in the promulgation of eminentlypublic documents, especially constitutions, inall four cases.In short, although settlements are primarilythe result of private negotiations among substan-tially autonomous elites, they have an importantpublic, or nonelite, aspect. The significance ofthis aspect probably has grown with theexpansion of information about elite activitiesdisseminated by modem news media and withthe development of opinion polling.Nonelite involvement presents a tricky prob-lem for elites who would fashion a settlement.On the one hand, it is essential that compromis-ing moderates be able to mobilize widespread,probablyoverwhelming, nonelite support againstintransigent elite persons and groups. On theother hand, these compromisers run the risk oflosing nonelite support it they are perceived asselling out their followers.Taken with the other features we have noted,the need for substantial elite autonomy helps usunderstand why elite settlements are so rare inmodem history and in the contemporary world.The historical circumstances, short-term pro-cesses, and elite autonomy that apparentlyprovoke and facilitate such settlements seldomoccur together. This is why disunified elites andunstable political regimes are such persistentfeatures of today's developing countries nomatter how much change occurs in other aspectsof their social structures or in their economicand international circumstances.ELITE SETTLEMENTS ANDDEMOCRATIC TRANSITIONSTo further clarify what elite settlements entail,and to suggest their theoretical implications, wesituate our analysis within the larger discussionof transitionsto democracy in the four countrieswe have considered and elsewhere. Possibly thebest-known sociological analysis of the condi-tions that facilitated England's eventual democ-ratization is that of Barrington Moore (1966).Moore focuses on the Civil War of the 1640s,giving primacy to the presence of a strong,relatively independent, landed upper class benton capitalist expansion, chafing under the fettersof the old order and royal prerogatives, andfacing a peasantry whose power and cohesionwere being undermined by the enclosures.Moore contends that "through breaking thepower of the king, the Civil War swept away themain barrier to the enclosing landlord andsimultaneously prepared England for rule by a'committee of landlords,' a reasonably accurate



9/14

302 AMERICAN SOCIOLOGICAL REVIEWif unflattering designation of Parliament in theeighteenth century" (p. 19). The resulting de-struction of the peasantry, he argues, sparedEngland both the reactionary and the socialrevolutionary responses to modernization thatappeared in other nations such as Germany,Japan, Russia, and China.Much in line with Moore, Theda Skocpol(1979, pp. 140-44) treatsEngland as a negativecase in her analysis of social revolutions inFrance, Russia, and China. She views En-gland's "political revolution" as spanning theperiod from 1640 to 1689, but asserts that "mostof the relevant action took place between 1640and 1660" (p. 141). According to Skocpol, theCivil War established the dominance of thelanded upper class over the monarchy; when, inthe 1680s, James II ignored this fact, he wasremoved "with very little fuss" (p. 144). LikeMoore, Skocpol concludes that the strength ofthe landed upper class relative to the monarchyand weakness of the peasantry combinedthereafter to immunize England against socialrevolution.From our perspective, Moore and Skocpol arecertainly right that the events of the seventeenthcentury secured upper-class dominance of theEnglish political regime. But unlike them, wethink the crucial development in this processwas the settlement fashioned by previouslydisunified, mainly upper-class elite factions in1688-89. As Jones (1972) demonstrates, thissettlement was much more than a simplepostscript to the events of 1640-60. The CivilWar and its Cromwellian aftermath had under-scored the depth of disunity among elitefactions. And, as evidenced by their numerousand bitter conflicts during the reigns of CharlesII and James II, disunity persisted for nearly 30more years before the major elite factions wereable to transcendtheirdeadly divisions. Withoutthe elite settlement of 1688-89, the accumulat-ing conflicts of the Restoration period couldwell have produced another civil war, some-thing elites seriously feared at the time(Schwoerer 1981, p. 211). More than anythingelse, then, the unification of predominantlyupper-class elite factions via a sudden anddeliberate settlement secured upper-classcontrolof the regime, enabled England to avoid thecivil wars, revolutions, and coups that subse-quently plagued most other European nations,facilitated England's rapid rise to world domi-nance, and permitted its peaceful evolutiontoward democracy.As noted, Moore and Skocpol stress thepeasantry's weakness in the seventeenth centuryas a key factor in England's subsequent politicaldevelopment. Although a weak peasantry wasdoubtless important, viewed comparatively, itcannot sustain the weight they place on it; for,

as Castles (1973) has pointed out, Sweden had astrong peasantry whose "free" members consti-tuted one of the four traditional Estates and as aclass comprised an important political forcethroughout the country's modem history. Yetafter the elite settlement of 1809, Sweden'spolitical development was even more placid anduntrammeled by social revolution or fascistreaction than England's. The point seems clear:whatever role one assigns to a weak or a strongpeasantry, the key variable for the establishmentof political stability and an eventual peacefultransition to democracy is the unification ofpreviously disuniflied elites, a unification that inEngland and Sweden occurred through highlydistinctive elite settlements.On the other hand, our analysis parallels anumber of approaches to democratic transitionsthat focus on elites (e.g., Rustow 1970; Rokkan1970; Almond et al. 1973; Linz 1978; Hartlyn1984; Peeler 1985). Thus, we find few indica-tions that in arranging settlements elites weremainly reflecting broader social or economic orcultural forces. Rather, the settlements appar-ently grew out of deliberate, relatively autono-mous elite choices among an array of possiblestrategies for protecting their diverse factionalinterests. And success in creating settlementswas due in part to the skills of specific elitepersons who happened to occupy pivotalpositions at the time.One of the most important of these ap-proaches is Dankwart Rustow's (1970) attempt,drawing on the cases of Sweden and Turkey, todevelop an ideal-typical description of demo-cratic transitions. Rustow (pp. 352-61) empha-sizes the importance of a "prolonged andinconclusive struggle" among well-entrenchedelite factions, deliberate elite choices, a willing-ness to compromise, agreement on proceduresrather than on fundamentals, and a period of"habituation" as indispensable to democratictransitions. All these elements are evident inthe cases we examined. A crucial differencebetween our analysis and Rustow's is that he isconcerned with democratic transitions per se,whereas we are focusing on elite settlements asa precondition for democratic transitions. Webelieve it is necessary to separate the twophenomena, that is, to extract elite settlementsfrom their embeddedness in the concept ofdemocratic transitions. Doing this allows one tosee that elite settlements sometimes longantedate democratic transitions but are stillapparentlya precondition for them. At least twocenturies separated the English elite settlementfrom the flowering of English democracy, andone can plausibly argue that while an elitesettlement probably occurred in Mexico duringthe late 1920s or early 1930s-involving thecreation of the InstitutionalRevolutionary Party



10/14

ELITE SETTLEMENTS 303(PRI) as an omnibus vehicle for elite coopera-tion-the country is only today beginning itsdemocratic transition.Rustow's treatment of the Swedish andTurkish cases actually illustrates the utility ofthe distinction we urge. Concerned with demo-cratic transitions, Rustow concentrates on theperiod around 1907, when Swedish elitesadopted universal suffrage and proportionalrepresentation. He does not mention the 1809elite settlement, which produced a century ofpeaceful conflict management-including theSwedish Estates' voluntarydissolution in 1866-and thus laid the foundations for Sweden'sdemocratic transition. In other words, Rustow'scharacterization of conditions and phases ofdemocratic transitions omits the fundamentalprecondition. Failure to recognize the centralityof an elite settlement also accounts for Rustow'sjudgment that Turkey had accomplished ademocratic transitionby the late 1960s, when hewrote. Turkey's 1971 coup and subsequentpolitical instabilityandrepressionclearly showedthat its shift to democratic politics during the1960s was merely temporary. This might wellhave been Rustow's conclusion had he seen thata democratic transition depends on the prioroccurrence of an elite settlement, of which therewas little or no evidence in Turkey.Another work akin to ours is Stein Rokkan's(1970) treatment of stable democracy as theoutcome of elite choices at particularhistoricaljunctures to create elite structures that incorpo-rate rather than exclude challenger elites. Weare attempting to clarify what such an elitestructure entails and to pinpoint one way itoriginates. Samuel Huntington's (1984) recentassessment of the prospects for more countriesbecoming democratic also has a definite elite-centeredthrust:"[Democratic] institutions comeinto existence through negotiations and compro-mises among political elites calculating theirown interests and desires" (p. 212). Huntingtonimplicitly recognizes that a fundamental changein elite structure must occur for there to be atransition to stable democracy, and he arguesthat this change may happen either through"transformation"or "replacement"of the exist-ing elite. But he does not elaborate on thecharacteristicsof the new elite structure that iscreated, and, like Rustow, he treats elitechanges as part and parcel of democratictransitions. Moreover, military conquests apart,we disagree with Huntington's suggestion thatdemocratic transitions can occur through thereplacement of an existing elite. Probably, ourdisagreement mainly involves semantics, butsemantic clarity is crucial here. Huntington'sapparentmeaning of "elite" when he speaks ofreplacement is those controlling a government.By contrast, we construe elites as including all

persons with power to make serious trouble evenif they cannot make government policy. Thougha settlement may involve changes in topgovernment personnel, it is primarilya transfor-mation of relations among existing elite fac-tions.Greater awareness of this basic characteristicof elite settlements can be seen in some casestudies. For example, in his study of Venezuela,Levine (1978, p. 93) contrasts the behavior ofessentially the same elite persons before andafter the 1958 settlement, concluding that "themost striking feature of Venezuelan politicsafter 1958 is the conscious, explicit decision ofpolitical elites to reduce interparty tension andviolence, accentuate common interests andprocedures, and remove, insofar as possible,issues of survival and legitimacy from thepolitical scene." This shift to elite cooperation,conciliation, and the muting of conflict is alsostressed by Peeler (1985) in his analysis ofCosta Rica, Colombia, and Venezuela. Examin-ing the mechanisms by which elites in thesecountries maintain liberal democratic regimesand give them legitimacy, Peeler observes,"Fundamentally, they have replaced the old,highly visible partyor personal hegemonies by amuch more subtle and flexible joint hegemonyof elites who quietly cooperate on the big issueswhile publicly competing with each other on thesmaller issues" (p. 123; emphasis added). Moreclearly than most other analysts, Peeler distin-guishes the condition of elite consensual unity,or accommodation, from the institutions ofliberaldemocracy, hypothesizing that "an ongo-ing spirit of elite accommodation is requisite tothe establishment and maintenance of liberaldemocracy in Latin America" (p. 145, emphasisadded). We concur with Peeler's thesis, and wesuggest further that it holds for all nations withdisunified elites and undemocratic regimes. Wewould stress, however, that a sudden anddeliberate elite settlement is possibly the onlyway an "ongoing spiritof elite accommodation"can be achieved in the less developed countriesof today's postcolonial world.In his important, still-evolving analysis ofSpain during the late 1970s, Richard Gunther(1985; Gunther et al. 1986) sharpens this focuson the causal role elite settlements play intransitions to democracy. He highlights thesecret negotiations that occurred among allsignificant Spanish elite groups during 1977-78,the care they took to give each other more orless proportional representation in these negoti-ations, and their deliberate avoidance of recrim-inations over the actions of the Franquistregime. The result was "a textbook case of elite



11/14

304 AMERICAN SOCIOLOGICAL REVIEWsettlement,"2 which was followed by Spain'ssuccessful transition to a stable, representativedemocratic regime during the early 1980s.Nevertheless, we think a serious limitation ofmost elite-centered work on the conditions forstable democracy is its ad hoc theoretical status.Awareness of this problem can be seen in theefforts of students of the Colombian andVenezuelan democratic transitions (e.g., Wilde1978; Dix 1980; Hartlyn 1984; Levine 1978) tofit their analyses into Lijphart's (1968, 1977)"consociational democracy" framework. Butthe fit is loose at best because Colombia andVenezuela cannot be considered plural societies(Lijphart 1977, p. 33), and their democraciescannot be considered consociational (Linz 1978,p. 8), though Colombia did evince consocia-tional forms during the 16-year period ofNational Front government. Lijphart's frame-work is presumably attractive to these scholarsbecause it shows the link between elite cooper-ation and stable democracy. With Di Palma(1973), however, we contend that the samemode of elite behavior is found in majoritariandemocracies. A more systematic approachwould postulate that (1) underlying both conso-ciational and majoritariandemocratic regimes isthe same type of elite structure,which we thinkof as "consensually unified"; (2) the consensu-ally unified elite structure constitutes one ofseveral distinct variants of elite structurein themodern world; (3) consensually unified elitesoriginate in only a few specifiable ways, ofwhich the elite settlement is today probably themost likely; and (4) consensual elite unityconstitutes a necessary precondition for stablerepresentative regimes which tend to evolvealong democratic lines. These postulationsextend the elite paradigm, and in doing so theypromise a more systematic theory of politicalchange.CONCLUDING REMARKSWe have had two main goals in this article, onefairly modest, the other more ambitious. Ourfirst goal has been to show that the elitesettlement is a distinct form of major politicalchange which merits serious attention. Theresimply is no literature that examines elitesettlements as a discrete phenomenon. Atpresent, what we know about them must begleaned from historical narratives on particularcountries, from social science case studies,which are loosely theoretical at best, and fromworks that embed and thus obscure the settle-ment in some other supposedly more central

phenomenon such as democratic transitions orbourgeois revolutions. So there is a need forextensive research and theorizing that focusesdirectly on elite settlements. Indeed, we thinkelite settlements merit a level of attentionapproximating that given to social revolutions.Our more ambitious goal has been to advancea particulartheoretical view of elite settlements.We have approached the phenomenon from theconceptual framework and basic assumptions ofthe elite paradigm, contending that (1) elitesettlements are the result of relatively autono-mous elite choices and thus cannot be predictedor explained in terms of social, economic, andcultural forces; (2) the consensually unified elitestructure created by a settlement constitutes theprimary basis for subsequent political stability;which (3) is a necessary condition for theemergence and sustained practice of representa-tive democratic politics. The exploratory natureof this undertaking, the complexity of thesubject, and space constraints prevent us fromconsidering many of the questions we may haveraised. But two clear limitations in our approachto the historical evidence should be addressed.We have focused only on positive cases ofelite settlements. Although this is perfectlylegitimate, the obvious next step is to examinenegative cases. Prime candidates would becountries that have experienced democraticbreakdowns. In some cases, it appears thatdemocratic institutions were initiated by failedor incomplete elite settlements, such as Spain'sturnopacifico agreementof 1875, the establish-ment of France's Third Republic in 1875, andColombia's Republican Union of 1909. Otherinstances of democratic transition and subse-quent breakdown reveal, on brief examination,much less evidence of elite efforts to reach asettlement: Argentina's Saenz Pena law of1912, the establishment of the Spanish SecondRepublic in 1931, Chile's returnto democracyin 1932, Turkey's democratic reforms of 1950.Examination of such cases will both clarify thedistinctive features of lasting elite settlementsand provide opportunities to evaluate ourargument about their causal significance.Another limitation involves our treatment ofthe historical data primarily as illustrations ofthe conditions and features of elite settlements.We have lacked space for discussing any case indetail. In addition to glossing over rich andintrinsically important information about eachindividual case, this approach has not allowedus to sort out and look for causal connectionsbetween the various social, economic, cultural,and political forces that may affect elitesettlements. Thus we cannot claim to havetested our propositions about the causes andconsequences of elite settlements. But, follow-ing an accepted strategy in this genre of2 Personal communication from Richard Gunther,October24, 1986.



12/14

ELITE SETTLEMENTS 305historicalsociology (Skocpol 1984, pp. 366-67),we have demonstrated the plausibility of ourargument by contrasting it with leading alterna-tive approaches. In concluding, we offer onemore such contrast, juxtaposing our positionwith the more societally based approachadoptedby Jack Goldstone (1986) in his importantworkon "state breakdown" in mid-seventeenth-century England.Population is the driving force in Goldstone'smodel of state breakdown and revolution.Drawing on an impressive array of evidence,Goldstone contends that sustained growth of theEnglish population during the period 1500-1650stimulated price inflation and a fiscal crisis ofthe state, increased the volume of upward anddownward mobility, causing heightened elitecompetition for scarce positions, and raised thecountry's mass mobilization potential by caus-ing a drop in real wages, rapid urban growth,and expansion of younger age cohorts. Theresult was state breakdown in the 1640s. Incontrast, he argues, the "relatively peacefulchange of rulers" in 1688 was "at least partlydue to the greater social peace that thenprevailed: in 1688 England had behind it ageneration of sharply reduced social mobility,stable prices, rising real wages, and slowerurban growth" (1986, pp. 305-6).By combining this well-grounded model witha persuasive critique of neo-Marxian theories,Goldstone has presented a formidable socialexplanation of political conflict and transforma-tion, which appears at least to fit seventeenth-century England, and may fit other countriestoo. Is it compatible with our more narrowlypolitical explanation? We think so, if the eliteconflict variable in his model is seen as afunction of much more than population and theother variables he employs. Specifically, theseverity of elite conflict must be seen asprimarily a function of the pre-existingg elitestructure, of whether elites are unified ordisunified. Elite structureis not a variable thatresponds mechanistically to various societalconditions. Disunified elites tend to remaindisunified, even when social conditions wouldseem to favor unity. Unified elites tend toremain unified even when social conditionswould seem to favor disunity.English elites, like all the other elites of earlymodem Europe (except the Dutch, who achievedunity in winning independence from Spain),were disunified at the time England emerged asa nation-state. Though some periods were, ofcourse, more peaceful than others-which maywell be partially explained by the variables inGoldstone' s model-deadly interelite warfarewas the prevailing fact of political life through-out the fifteenth and sixteenth centuries. Elitesnot only fought over scarce positions and

personal grievances, they also fought overissues of economic policy, religion, localautonomy, the line of monarchical succession,and, indeed, the use of monarchical power itselfto further their individual interests. ThusEnglish elites confronted the especially divisiveconditions of the mid-seventeenth century with-out widely accepted rules of the game forpeacefully managing their conflicts and those ofthe larger society. Neither the Civil War nor theRestoration of 1660 altered their basic disunity.As Jones (1972, p. x) argues in his authoritativehistory of the period, "It is my thesis that theevents of the 1640s and 1650s had not beendecisive or conclusive, that the major constitu-tional and political issues were still open andundecided in James's time, and that the victoryof parliament, representative government andthe common law was by no means predeter-mined or inevitable."In our view, it is quite plausible that theapparently more benign societal conditions thatGoldstone says prevailed in the late 1680shelpedfacilitate the English elite settlement. Onthe other hand, it would be implausible to arguethat knowledge of such conditions could enableone to predict the English or any other elitesettlement with even a modest degree ofaccuracy. Indeed, in the favorable conditionsthat Goldstone suggests prevailed throughoutEurope during the period 1650-1750, Englandwas the only country to experience an elitesettlement; all other disunified national elitesremained disunified. Moreover, during theperiod 1750-1850, which Goldstone says broughta return to the disruptive demographic, eco-nomic, and mobility conditions that character-ized Europe in the period 1500-1650 andresulted in widespread state breakdowns, En-glish elites remained unified and politicalinstitutionsremained stable andrepresentativeinnature (as did Dutch elites and institutionsbefore and after the French occupation of1795-1813). It is also noteworthy that duringthe same period Swedish elites achieved theirsettlement, seemingly in the face of many of theconditions that were associated with statebreakdowns in most European countries.In short, the variable of elite structureshouldbe included in Goldstone's model. The modelmight turn out to be a good predictorof certainkinds of state breakdown when elites aredisunified. But we should expect that whereunified elites exist, whether of the consensual orideological kind, they will manage the disrup-tive societal conditions on which Goldstonefocuses in ways that avoid state breakdowns orother irregularseizures of power.To conclude, there are strong reasons tobelieve that a robust conceptualization of basicvariations in elite structure must be given a



13/14

306 AMERICAN SOCIOLOGICAL REVIEWcentral place in explanations of political conflictand change. Although our three fold distinctionbetween disunified elites and two types ofunified elites is widely used, albeit withdifferent labels, it has not been seen as partof ageneral theory of political change. By focusingon the elite settlement as a fundamentaltransformation of elite structure from thecondition of disunity to thatof consensual unity,we hope not only to have directed attentiontoward a neglected, yet extremely important,political phenomenon, but also to have gonesome way toward demonstratingthe explanatorypotential of the elite paradigm when it isextended in this manner.

REFERENCESAlford, Robert R. and RogerFriedland. 1985. Powers ofTheory: Capitalism, the State, and Democracy.London:CambridgeUniversityPress.Almond, GabrielA., Scott C. Flanagan, and Robert J.Mundt, eds. 1973. Crisis, Choice, and Change:Historical Studies of Political Development. Boston:Little, Brown.Beck, C. and J. Malloy. 1964. "PoliticalElites: A Modeof Analysis." Presentedat the Sixth World Confer-ence, International Political Science Association,Geneva.Bialer, Seweryn. 1980. Stalin's Successors: Leadership,Stability,and Changein the Soviet Union. New York:CambridgeUniversityPress.Blank, David E. 1973. Politics in Venezuela. Boston:Little, Brown.Brown, John. 1895. OriginalMemoirs of the Sovereignsof Swedenand Denmark.London:H.S. Nichols.Burggraaff,Winfield J. 1972. The VenezuelanArmedForces in Politics, 1935-1959. Columbia:Universityof Missouri Press.Burton, Michael G. and John Higley. 1987. "Invitationto Elite Theory." Pp. 219-38 in Power Elites andOrganizations, edited by G. William Domhoff and

ThomasR. Dye. NewburyPark,CA: Sage.Castles, Francis G. 1973. "BarringtonMoore's ThesisandSwedishPolitical Development."Government ndOpposition8:313-31.Dahrendorf, Ralf. 1969. Society and Democracy inGermany.GardenCity, NY: Anchor.Di Palma, Giuseppe. 1973. The Study of Conflict inWesternSociety: A Critiqueof the End of Ideology.Morristown,NJ: GeneralLearningPress.Dix, Robert. 1980. "Consociational Democracy: TheCase of Colombia." ComparativePolitics 12:303-21.Elstob, Eric. 1979. Sweden, A Political and CulturalHistory. Totowa, NJ: Rowman and Littlefield.Field, G. Lowell and John Higley. 1973. Elites andNon-Elites: The Possibilities and their Side Effects.Andover,MA: WarnerModularPublications._ 1980. Elitism. Boston: Routledge & KeganPaul.. 1985. "NationalElites and Political Stability."Pp. 1-44 in Researchin Politics and Society: Studiesof the Structure of National Elite Groups, vol. 1,editedby Gwen Moore. Greenwich, CT: JAI Press.Fleron,Frederick ., Jr. 1969. "Toward Reconceptualiza-

tion of Political Change in the Soviet Union: ThePolitical LeadershipSystem." Pp. 222-43 in Commu-nist Studies and the Social Sciences, editedby FredericJ. Fleron, Jr. Chicago: Rand McNally.Goldstone, Jack A. 1986. "State Breakdown in theEnglish Revolution: A New Synthesis." AmericanJournal of Sociology 92:257-322.Gunther, Richard. 1985. "Democratization and PartyBuilding: Contradictionsand Conflicts Facing PartyElites in the SpanishTransition o Democracy." Paperpresented at the annual meeting of the AmericanPolitical Science Association, New Orleans.Gunther, Richard, Giacomo Sani, and Goldie Shabad.1986. Spain AfterFranco: The Making of a Competi-tive Party System. Berkeley: University of CaliforniaPress.Hartlyn,Jonathan. 1984. "MilitaryGovernmentsand theTransition to Civilian Rule: The Colombian Experi-ence of 1957-1958." Journal of InteramericanStudiesand WorldAffairs26:245-81.Higley, John and Gwen Moore. 1981. "Elite Integrationin the United Statesand Australia."AmericanPoliticalScience Review 75:581-97.Hill, Christopher. 1972. The World Turned UpsideDown: Radical Ideas During the English Revolution.New York: Viking.Horowitz, Henry. 1977. Parliament, Policy and Politicsin the Reign of WilliamIII. Newark: University ofDelaware Press.Huntington, Samuel P. 1984. "Will More CountriesBecome Democratic? Political Science Quarterly99: 193-218.

Jones, J. R. 1972. The Revolutionof 1688 in England.New York: W. W. Norton.Karl, Terry. 1981. "Petroleumand Political Pacts: TheTransition to Democracy in Venezuela." WorkingPaper No. 107, The Wilson Center, Latin AmericanProgram.Washington,D.C.Levine, Daniel H. 1978. "Venezuela Since 1958: TheConsolidationof DemocraticPolitics." Pp. 82-109 inThe Breakdownof DemocraticRegimes: Latin Amer-ica, edited by Juan J. Linz and Alfred Stepan.Baltimore:JohnsHopkins UniversityPress.Lijphart,Arend. 1968. The Politics of Accommodation:Pluralism and Democracy in the Netherlands. Berk-eley: Universityof CaliforniaPress.. 1977 Democracy in Plural Societies: AComparative Exploration. New Haven, CT: YaleUniversityPress.Linz, JuanJ. 1978. "Crisis, Breakdown, and Reequili-bration."Pp. 3-124 in The Breakdownof DemocraticRegimes, edited by Juan J. Linz and Alfred Stepan.Baltimore:JohnsHopkins UniversityPress.Manley,John F. 1965. "The House Committee on Waysand Means: Conflict Management n a CongressionalCommittee." American Political Science Review59:927-39.Maravall, Jose. 1982. The Transition to Democracy inSpain. London: CroomHelm.Martz,John D. 1966. AccionDemocratica:Evolutionofa ModernPolitical Party in Venezuela.Princeton,NJ:PrincetonUniversityPress.Montealegre, Flora. 1983. "Costa Rica at the Cross-roads."Developmentand Change 14:277-96.Moore, Barrington,Jr. 1966. Social Origins of Dictator-ship and Democracy:Lordand Peasant in the Makingof the ModernWorld. Boston: Beacon.



14/14

ELITE SETTLEMENTS 307Mosca, Gaetano. 1939. The Ruling Class. New York:McGraw-Hill.Pareto, Vilfredo. 1935. TheMindand Society:A Treatiseon GeneralSociology. New York:Dover.Peeler, John A. 1985. Latin American Democracies:

Colombia, Costa Rica, Venezuela. Chapel Hill:Universityof North CarolinaPress.Plumb, JohnH. 1967. The Origins of Political Stabilityin England. EnglewoodCliffs, NJ: Prentice-Hall.Purcell, Susan K. and John F.H. Purcell. 1980. "Stateand Society in Mexico: Must a Stable Polity BeInstitutionalized?"WorldPolitics 32(2):194-227.Putnam, Robert D. 1976. The ComparativeStudy ofPolitical Elites. Englewood Cliffs, NJ: Prentice-Hall.Rokkan,Stein, Angus Campbell,Per Torsvik, and HenryValen. 1970. Citizens, Elections, Parties. New York:McKay.Rustow, Dankwart. 1970. "Transitionsto Democracy:Toward a Dynamic Model." Comparative Politics(April):337-63.Schwoerer, Lois G. 1981. The Declaration of Rights,1689. Baltimore:Johns HopkinsUniversityPress.Scott, FranklinD. 1977. Sweden:The Nation's History.Minneapolis:Universityof Minnesota Press.Skocpol, Theda. 1979. States and Social Revolutions.Cambridge:CambridgeUniversityPress.. 1984. "Emerging Agendas and Recurrent

Strategies in Historical Sociology. " Pp. 356-91 inVision and Method in Historical Sociology, edited byTheda Skocpol. Cambridge: Cambridge UniversityPress.Steiner, Kurt. 1972. Politics in Austria. Boston: Little,Brown.Stiefbold, Rodney P. 1974. "Segmented Pluralism andConsociational Democracy in Austria: Problems ofPolitical Stability and Change." Pp. 117-77 in Politicsin Europe, edited by Martin0. Heisler. New York:McKay.Tardanico, Richard. 1982. "State, Dependency, andNationalism: Revolutionary Mexico, 1924-1928."Comparative Studies in Society and History24:400-423.

Taylor, PhilipB. 1968. The VenezuelanGolpede Estadoof 1958: The Fall of Marcos Perez Jimenez.Washington,D.C.: Institute or the ComparativeStudyof Political Systems.Trevelyan, George M. 1938. The English Revolution,1688-1689. London: ThorntonButterworth.Wilde, Alexander W. 1978. "Conversations AmongGentlemen: Oligarchical Democracy in Colombia."Pp. 28-81 in The Breakdownof DemocraticRegimes:Latin America, edited by Juan J. Linz and AlfredStepan. Baltimore:JohnsHopkins UniversityPress.

elite settlements

Documents