You are on page 1of 23

AREND LIJPHART

University of Leiden

Among the several fields or subdisciplines into which the discipline of political
science is usually divided, comparative politics is the only one that carries a
methodological instead of a substantive label. The term “comparative poli¬tics”
indicates the how but does not specify the what of the analysis. The label is
somewhat misleading because both explicit methodologi¬cal concern and implicit
methodological aware¬ness among students of comparative politics have generally
not been very high.1 Indeed, too nmny students of the field have been what
Giovanni Sartori calls “unconscious thinkers” —unaware of and not guided by the
logic and methods of empirical science, although perhaps well versed in
quantitative research techniques. One reason for this unconscious thinking is
un¬doubtedly that the comparative method is such a basic, and basically simple,
approach, that a methodology of comparative political analysis does not really exist.
As Sartori points but, the other extreme —-that of the “overconscious thinkers,”
whose “standards of method and the¬ory are drawn from the physical paradigmatic
sciences”—is equally unsound.* The purpose of this paper is to contribute to
“conscious think¬ing- in comparative politics by focusing on comparison as a
method of political inquiry. The paper will attempt to analyze not only the inevitable
weaknesses and limitations of the comparative method but also its great strengths
and potentialities.
In the literature of comparative politics, a wide variety of meanings is attached to
the terms “comparison” and “comparative method.” The comparative method is
defined here as one of the‘basic methods—the others being the experimental,
statistical, and case study methods—of establishing general empiri¬cal
propositions. It is, in the first place, defi¬nitely a method, not just “a convenient
term vaguely symbolizing the focus of one’s research interests.”3 Nor is it a special
set of substantive concerns in the sense of Shmuel N. Eisenstadt’s definition of the
comparative approach in social research; he states that the term does not “properly
designate a specific method . .., but rather a special focus on cross-societal, Institutional, or macrosocietal aspects of societies and social analysis.”4
Second, the comparative method is here de-. fined as one of the basic scientific
methods, not the scientific method. It is, therefore, narrower in scope than what
Harold D. Lasswell has in mind when he argues that “for anyone with a scientific
approach to political phenomena the idea of an independent comparative method
seems redundant,” because the scientific ap¬proach is “unavoidably
comparative.,,<J Like¬wise, the definition used here differs from the very similar
broad interpretation given by Ga-briel A. Almond, who also equates the

compar¬ative with the scientific method: “It makes no sense to speak of a
comparative politics in po-litical science since if it is a science, it goes without
saying that it is comparative in its ap-proach.”6
•Arthur L. Kalleberg, “The Logic of Comparison:
A Methodological Note on the Comparative Study of Political Systems,” World
Politics, 19 (October 1966), p. 72.
4 Shmuel N. Eisenstadt, “Social Institutions: Com-parative Study,” in David L. Sills,
ed., International Encyclopedia of the Social Sciences (N<ew York: Macmillan & Free
Press, 1968), Vol. 14, p. 423. See also Eisenstadt, “Problems in the Comparative
Analy¬sis of Total Societies,*’ Transactions of the Sixth World Congress of Sociology
(Evian: International Sociological Association, 1966), Vol. 1, esp. p. 188.
Harold D. Lasswell, “The Future of the Com¬parative Method,** Comparative
Politics, 1 (October, 1968), p. 3.
•Gabriel A. Almond, “Political Theory and Po-litical Science,** American Political
Science Review, 60 (December, 1966), pp. 877-78. Almond also ar-gues that
comparative politics is a “movement” in political science rather th^n a
subdiscipline. See his
Third, the cpmparative method is here re¬garded as a method of discovering
empirical re¬lationships Among variables, not as a method of measurement. These
two kinds of methods should be clearly distinguished It is the latter that Kaileberg
lias in mindi when he discusses the “logic of comparison/ He defines the
com¬parative method as “a form of itieasurement”; comparison means
“nonmetrical ordering,-' or in other words, ordinal measurement.7 Simi¬larly,
Sartori is thinking in terms of measure¬ment on nominal, ordinal (or comparative),
and cardinal scales when he describes the con¬scious thiiiker as “the man that
realizes the lim¬itations of not having a thermometer and still manages to say a
great deal simply by saying hot and cold, warmer and cooler.”8 This impor¬tant
step of measuring variables is logipaily prior to the step of finding relationships
among them. It is the second of these steps to which the term “comparative
method” refers in this pap?r.
Finally, a clear distinction should be made between method and technique. The
cpmparative method is a broad-gauge, general method, not a narrow, specialized
technique. In this vein, Gurinar Heckscher cautiously refers to “the method (or at
least the procedure) of compari-son,-*9 and Waiter Goldschmidt prefers the term
comparative approach, ibepause “it lacks the preciseness to call it a rhethbd,”^ The
cpm¬parative ixibthod may also be thought of as a basic research strategyf in
contrast with a mere tactical aid to research. This will become clear in the
discussion that follows.

. 1. 7 (January-Jurie. uses two equivalent groups. 1954). 1961). Nadel.ode comparative eri Science Politique. 1970). one of which (the experimental group) is exposed to a stim-ulus while the other (the control group) is not. see L$o Moulin. as the ex¬perimental and the statistical methods. Turner. Readings in Cross-Cultural Methodology (New Haven. ‘This Crbss-pultural Method.” Revue Inter¬nationale d'Htstoire Politique et CoristUutionelle. Conn. the condition that the cetera are indeed paria—c&vi be achieved by a process of deliberate randomization. Holt and John E. but unfortunately it u For thes idea of discussing the comparative meth¬od in relation to these qtheir basic methods. 1970). The Logic of Comparative Social Inquiry (New York: Wiley-Iriterscicnce. Maurice Duverger. Thus pne knows the relationship between two variables— with the important assurance that no other variables were involved* because in all respects but one the two groups were alike.: HRAF tvess. Handbook of Social Psy¬chology (Reading. 57-71. Equivalence—that is. which consists of two basic elements: (1) the establishment of general emjpirical relation¬ships among two or more variables. 101-17. 1966). M The case study method will be discussed below.13 while (2) all other variables are controlled. For other general discussions of the comparative method.*' Trans¬actions of the Sixth World Congress of Sociology (Evian: Internatibhal Sociological Association. 1951). The Methodology of Comparative Research (New York: Free Press.. Vol. Thp exper¬imental method is the most nearly ideal method for scientific explanation. pp. . 222-55. 1957). pp. S. Adam Przeworski and Henry Teune. The two groups are then compared.: Addisbn-XVesley. that is. Sriieiser’s example. Moore. pp. Afttftodes des Sciences Sociales (3rd ed„ Paris: Pfe&ses Universitaires de France. 1964). The Foundations of Social Anthropology (London. 523-31. following Neil J. “The Methodology of Com¬parative Research. pp. in its simplest form. The ceteris paribus condition is vital to empirical generalizations. and any difference can be attributed to the stimulus. these will be referred to.*’ m Holt and Turner. F. 2.11 All three methods (as well as certain forms of the case study method1*) aim at scientific explana¬tion. eds. 1-20. pp. Cohen arid West. “La M6th.. Statistical. ed. Shielser’s outstanding and most enUghtehing article *‘Note$ on the Methodology of Comparative Analysis of Economic Activity. and Comparative Methods The nature of the comparative method can be understood best if it is compared arid con¬trasted with the two other fundamental strate¬gies of research. held constant. John W. I am in* deb ted tp Neil J.M in Gardner Liiidzey. Frank W. pp. The experimental method. Vol. Mf Whiting. arid Robert T. Mass.The Experiment#!. These two elernents are inseparable: one cannot be sure that a relationship is a true one unless the influence of other variables is controlled. 375-99. ed.

eds. Moreover. . This can be done by partialihg^dividing the samjple into a nuhiber of different age groups and looking at the correlations between participation and education within each sepa-rate age group. 1955). control by means of partial correlations does not allow' for the effects of measurement error or unique factor cpnjjppnems. par.quiry.. 4<Testirig a. to consider the role of further vari¬ables. The Theory and Method of Political Analysis (Homewood.”1* The statistical method can be regarded. Sod.’ i. 43). 115.: Dorsey Press. He expresses this idea in three short sentences: “Science seeks to establish relationships” (p. Single.j elatibns do hpt resolve the problem of the codiirusion of characteristics. Galton’s problem”. as an approximation of the experi¬mental metlipd. Meehan. The Langucige of Social Re¬search: A Readier in the Methodology of Social Re¬search (Glericoe. „? Lazarsfeld. Brewer.. 1970). It entails the conceptual (mathematical) manipulation of empiricially ob¬served data—which cannot be manipulated sit. William D... is empirical” (p. “every bi'aiach of inquiry aiming at reliable gen¬eral laws concerning empirical subject matter must employ a procedure that. IU. pp. merely for the other key variables that are known or suspected to exert influence. Cranp ttnd Donald T. ‘Science is a generalizing activity” (p. therefore. “Interpretation of Statistical Relations as a Research Operation. IU. 33 (March. Strictly speaking.: Free Press. he immediately starts to ‘crosstabu¬late. How¬ler. It cannot control for all other variables. if it is not strictly controlled experimentation. “Two . Whenever an investigator finds him¬self faced with the relationship between two variables. Paul F. Has the es-sential logical functions of experiment in iri. can only rarely bfe used in political science be¬cause of practical and ethical impediments. An alternative to the experimental method is the statistical method. Campbell. As Ernest Hagel emphasizes. see Marilynn B. “Science .” in Lazarsfeld and Morris Rosenberg. 111. 37). one should control for the in-fluence of age because younger generations have received more education than older genera¬tions. p. 35).omefry.e. but it is not as strong a method as experimentation because it cannot handle the problem of control as well. when one wants to inquire into the relationship between political participation and level of edu¬cation attained.tial Correlations in Hypothesis-Testing Research. known in anthropology fs.uationally as in experimental design—in order to discover controlled relationships among vari-ables.Factor I^odel as an Alternative to the Misuse of Par. Lazarsfeld states that this is such a basic research procedure that it “is applied almost automatically in empirical research.** Eugene J. see Raoul Naroll. For instance.”15 The statistical method does have these essential logical functions. It handles the problem of control by means of partial correlations.

More¬over. Blalock empha¬sises “the underiymg similarity between the logic of making causal inferences on the basis of experimental arid nonexperimental designs” (p. and World.” p. 1961). too. but nipt with absolute certainty. 51-53.16 But experimental design provides the closest appi'oximation to this ideal. “Iiitferpfetation of Statistical Relations as a Research Operation. the investigator knows that they are alike with a very high degree of probability.18 It follows that in many re. The crucial difference is that the number of cases it deals with is too small to permit systematic control by means of partial correlations. 26). the difference depends entirely on the • number of cases. which is (especially importpt if one seeks to establish causal relationships. The comparative method resembles the statistical method in all respects except one. In statistical design. is an approximation—not the equiyaient—of the experimental method.17 The logic of the comparative method is. Brac6. one quickly “runs out of cases.# For instance. In general. 452f. 1961). The statistical method. Another advantage of the experimental method is that the time variable is controlled. 743. even the experimental method does not handle the problem of control perfectly. 1964). 119. in accordance with the general standard ex¬pounded by Nagel. in turn. op. . 17 Lazarsfeld. 28 (January. states. this control can be approximated by means of the panel method. Talcptt Paribus makes a. so-called “forcing variables” cannot be controlled by randomi¬zation. See his Causal Inferences in Nonexperimental Research (Chapel Hill: University of North Carolina Press. as Hubert M. consequently. one can also argue.So¬lutions to Galton’s Problem. pp. 23-26.” Philosophy of Science. Con-versely. There is. 1539. Jr. . no clear dividing line between the statistical and comparative meth¬ods. as Lazarsfeld does. but only if one adds that it is an especially potent form. nothing but the comparative method where the cases to be compared are produced to order and urfder controlled condi¬tions. actually alike in every respect. This problem occurs in statistical operations. similar statement with regard to the com¬parative method: “Experiment is . The Structure of Science (New York.*’ See his The Structure of Social Action (2nd ed. because the investigator can never be completely sure that his groups are. especially when one wants to control simultaneously for many vari¬ables. if the groups are made equivalent by means of deliberate randomization. 1949).” The com¬parative method should be resorted to when the number of cases available for analysis is so small that cross-tabulating them further in or¬der to establish credible controls is not feasible. . Blalock. Harcourt. pp. New York: Free Press. pjp.. cit. 15 Ernest Nagel. and Przeworski and Teune. that the experimental method constitutes a spe¬cial form of the statistical method.. also the same as the logic of the experimental method. p.t pp.

the statistical method. Th& KuUs of '':S$&olpgtefli Method.*’ (Durkheim. 15. it can be argued with equal justice that the comparative and statistical methods should be regarded as tWo aspects of a single method. one can consider the comparative method in proper perspective and answer such questions as the following.** See his The Comparative Meth-od in Social Anthropology (London: Athlone Press. Land* eh Voikenfciiride.) femile Durkheim also follows this usage When he declares that "comparative sociology is not a particular br^ch of spciolcigyj it is sociology itself... in so far as it ceases to be purely descriptive and aspires to account for facts.82. because. a combination of the statistical and cpmparative methods is appropriate. . Rpdnij?y Needham combines the two terms. 160.) On the other hand. when he iriakes a distinction between “small-scale comparative studies’* arid “large. 118 [1962]. pp. as has sometimes been claimed?”20 The answer is that the com-parative method is not the equivalent of the ex-perimental method but only a veiy imperfect substitute. really an adequate substitute for experimen¬tation in the natural sciences. 22. A clear awareness of the limitations of the cpmparative method is necessary but need not be disabling. Evans-Pritchard uses exactly the same terminology as used by Smelser and as adopted in this paper.*. 1963). and speaks of “large-scale statistical comparison. .» including both the comparative arid statistical methods as defined in this p#per. . these weaknesses can be minimized. Of course. p. 193&L. Many authors use the term “comparative method’* in the search situations.” Le. Sbipvay arid John H.: Frde Presis. “Notes on Comparative Method and Prescriptive Alliance.” (Brown. 81 [1951]. [8th ed. the number of cases is necessarily so restricted that the comparative method has to be used. p. translated by Sarah A. Where the cases are national political systems. analysis. but npi^periraehtal. R. 22. with an intermediate number of cases. The “conscious thinker” in comparative politics should realize the limitations of the compara¬tive method. raised by Samuel H. as they of¬ten are in the field of comparative politics. E. 139. (Needham. “The Comparative Method in Social Anthropology. From the vantage point of the general aims and the alternative methods of scientific in¬quiry. p. 111. but he should also recognize and take advantage of its possibilities. pueller. as we shall see.scale statistical dries.” Bifdragen tot de TaaU. This is how A. Glencpe. Radcliffe-Brown uses the term when he argues that “only the comparative method can give us general propositions. /.) See also the statements by Lasswell arid Almphd cited above. Beer and by Harry Eckstein: Can comparison be regarded as ‘‘the social scientist’s equivalent of the natural scientist’s laboratory?”19 and: “Is the comparative method in the social sciences .” Joiirnpl of the Rdyat Anthropological Institute of Great Britain and Ireland.M In order to highlight the special problems arising from the availability of only a small number of cases. broad sense of the method of multivariate empirical. the comparative method is discussed as a distinct method.

the most fruitful approach would be to regard the comparative analysis as the first stage of re¬search.** in Richard L. Before turning to a discussion of specific sug-gestions for minimizing these problems.** Comparative Politics. and the statistical analysis as the sec¬ond stage. The other is “micro replications. 1 (October. if at all possible one should generally use the statistical (or perhaps even the experimental) method in-stead of the weaker comparative method. nationality can simply be treated as an additional variable on a par with other individual attributes such as oc-cupation. “Comparative Cross-National Re¬search: The Context of Current Efforts. Rokkan.. 1966).” de¬signed “to test out in other national and cul¬tural settings a proposition already validated in one setting. The Comparative Method: Weaknesses and Strengths The principal problems facing the compara¬tive method can be succinctly stated as: many variables. age. In one type of comparative cross-national re¬search. eds. Comparative Politics: A Reader (New York: Free Press of Glencoe. 1963). one can also use the statistical method. Past and Present. p. in Cross-National Research (New Haven: Yale University Press. sex. One is the testing of "macro hypotheses’* concerning the ‘liiterfelatiohs of structural elements of total systems”. see his “Methods and Models in the Comparative Stiidy of Nation- . too. etc. 3. the intensive comparative analysis of a few cases may be more promising than a more superficial statisti¬cal analysis of many cases. “The Comparative Method and the Study of British Politics. two geheral comments are in order. in which hypotheses are carefully for¬mulated.. These two problems are closely interrelated.” in Eckstein and David E. type of neighborhood. Beer. 19. one-case” approach.22 n Stein . Stein Rofekan distinguishes two aims of cross-national analysis. here the number of cases tends t6 be limited.20. i968).”21 Here. First. Apter. “A Perspective on Comparative Politics. But often. small number of cases. eds. the latter is peculiar to the comparative method and renders the problem of handling many variables iapi*e diffipult to spive. The former is common to virtually all social science research regardless of the particular method applied to it.nation. Comparing Nations: The Use of Quantitative Bata. 20'Harry Eckstem. and financial resources. Rokkan specifically recommends the use of “paired comparisons” for this purpose. one can use the com¬parative method. p. In such a situation. and one has to rely on the comparative method. byt if the proposition in ques¬tion focuses on individuals as uhits of analysis. as Mer¬ritt and Rokkan point out. Merritt and Rokkan. en¬ergy. instead of the “one. % is logically possible and may ad¬vantageous to shift from the comparative to the statistical method. 19. in which these hypotheses are tested in as large a sample as ppssible. given the inevitable scarcity of time.19 Samuel H. pp.

1 (April. Elections. See Arend Lijphart.2® ” Terence K.” in Citizens.33 (italics added). In my critique of Giovanni Sartori1? proposition relating political instability to extreme mutyipartism (systems with six or more significant parties). not universal. “Typol¬ogies of Democratic Systems/* Comparative Political Studies. where cases are picked that are in accordance with the hypothesis-r-and hypotheses are? re¬jected if one deviant case is found All cases should.TH'dividual reseaifcTi?^ The second general comment concerns a dangerous but tempting fallacy in the applica-tion of the comparative method: the fallacy of attaching top much significance to negative findings. Terenpe K. 1970).”25' Deviant cases weaken a probabi¬listic hypothesis. Parties: Approaches to the Comparative Study of the Processes of Development (Oslo: Universitetsfdrlfiiget. 1967). of course. Hopkins and Immanuel Wallerstein make a similar d^iincjtipn between truly “cross.4r43. One or two deviant cases obviously constitute a much less serious problem in a statistical analysis of very many cases than in a comparative study of only a few —perhaps less than ten—cases. But it is never-theless a mistake to reject a hypothesis “be¬cause one can think pretty quickly of a con¬trary case. 505.” Johan Galtunfe. cit. 52. 1967). 193. pp. 6 (October. The comparative method should not lapse into what Johan Qaltung calls “the tradi¬tional quotation/ illustration methodology. p. 1968).national studies” in which total systems are the units of analysis. op.. 3. 52.. if only perfect correlations should be permitted social science wbuld hot haVe come very far. See also Przeworski and Teune. M. J. pp. and tjhe scientific search should be aimed at probabilistic. pp. one of my argu¬ments consists of the deviance of a single historical case: the stable six-party system of the Netherlands during the interwar years. 27. “Merritt and Rokkan. 32-35.” So¬cial Science Information. but in the comparative analysis of a small number of cases even a single deviant finding tends to loom large. I have beeri guilty of committing this fallacy myself.Build¬ing. p. Hopkins and Immanuel Wallerstein. The functions of deviant case analysis will be discussed below. generalizations. be selected systematically. Politics and Social Science (Harrnondsworth: Penguin Books. and “multi-national but cross. op. gether. Theory ia'nd Methods of Social Research (Oslo: Uniyersitet^forliaget. 33 W. cit. The erroneous tendency to reject a hypothesis on the basis of a single deviant case is rare when the statistical method is used to analyze a large sample. “ He adds: “This is a very naive conception of so¬cial science propositions. 1967). . p. “The CpmparatiYe Study of National Societies. but they can only invalidate it if they turn up in sufficient numbers to make the hypothesized relationship disappear alto-. p. Mackenzie.

Even though in most situations it is impossible to augment the number of cases suf¬ficiently to shift to the statistical method. Nadel. pp. 1 (Octo. let us turn to a discussion of specific ways and means of minimizing the “many variables. 104-09. cf. .^It is clearly incorrect. 1968). Richardson. Dowse* “A Functionalist’s Logic.27 Modern comparative politics has made great progress in this respect as a result of the efforts of the field’s innovators to fashion universally applicable vocabularies of basic po¬litically relevant concepts. Holt and John M. therefore. Miller. ‘‘Con¬cept Misformation in Comparative Politics”. op.” In¬ternational journal of Psychology. but logically such evidence would not compel its withdrawal. The test of the hypothesis by way of a cohfrphtation with empirical or historical data remains inconclusive. 607- . however small. one should also consider the possibilities of “longi-tudinal” (cross-historical) extension by in¬cluding as many historical cases as possible. any enlargement of the sample. Robert E. Douglas Oliver and Walter B. pp.the. 109. one is faced with the problem of how repre-sentative one's limited sample is of the universe of cases.. ber.29 It was the promise of discpvering universal laws through global and longitudinal compari¬sons that made Edward A. cit. pp. small N” problem . “Suggestions for a More Sykt&atic Method of Comparing Political Units. and Nico Frijda and Gustav Jahoda.28 Such a restatement of variables in comparable terms makes many previously inaccessible cases avail¬able for comparative analysis. Freeman enthusias¬tically espouse the comparative method almost proposition. p. Guenter Lewy’s statement: “To be sure. unless one investigates all available cases. Jr. “Historical Data in Comparative Political Analysis: A Note on Some Problems of Theory. see Sartori. 1968). .16. pp. op. a finding of a very lar£e number of .0 World Pdlitics. The State of The* ory in Comparative Politics (Minneapolis: Center for Comparative Studies in Technological Development and Social Change.” American Anthropologist. “On tJie Scbpe and Methods of CrossCultural Research. notably the ap-proaches based on Parsonian theory and Ga¬briel Af Almond’s functional approach. 18 (July.. In addition to extending the analysis geographically. For critiques of recent attempts at terminological innovation in comparative politics. 237-38. cit. to argue that on logical grounds a probabilistic generalization can never be invalidated. im-proves the chances of instituting at least some control. 1 (1966). 1955). comparative method.” JLeyvy. 1966). Robert T. [deviant cases] would cast doubt tipori the value of the After these introductpry observations.These may be divided into four categories: Increase the number of cases as much as possible.. see Siiielser.of . aT Furtherinore. pp.” Comparative Politics.. “On the necessity of* establishing general concepts not tied to particular cultures. 118-21. 114. 57 (February.

however. 5 (Winter. published in 1873. . “Almond's Concept of ‘The Political System’: A Textual Critique.ernment and Opposition. and stated that it could lead to the formulatipc of “analogies . Such adduction of what Lazarsfeld calls the “prpp. 1.” Gov. because of the/serious lack of information concerning most political systems. op. 19. 15 (June.t pp. and thus to achieve the same objective of increasing the average num¬ber of cases per 011.*3? The field of cpihparative politics has not yet achieved-r-and may never achieve—the goals that Freeman set for it with such opti¬mism. “Comparative Analysis. pp. 103-10. by simplifying a set of several catego¬ries into a dichotomy). (The value of this suggestion is somewhat dimin¬ished. See also Lewy. of course. 1955) .” Western . but dissimilar as far 10 Edward A. . Reduce the "property^space” of the analysis.-' Comparative poli¬tics could thus discover “a World in which times and tongues and nations which before seemed parted poles asunder. . Political Quarterly. Finer. See also Gideon Sjpberg's argument in favor of global com¬parative research: “The Comparative Method in the Social Sciences.” Philosophy of Science. 3-21. and the number of cases in each cell increased correspondingly. 298n. Comparative Politics (Lon¬don: Macmillan. cit. ifihe latter procedure. 1969-70). If the sample of cases cannot be in¬creased. w Michael Haas. and should not be used lightly.31 It may also be advisable in certain instances to reduce the number of classes into which the variables are dividend (for instance. “compara¬ble” means: similar in a large number of im-portant characteristics (variables) which one wants to treat as constants. and Samuel E. it may be possible to combine two or i more variables that express an essentially sirni. 302. he called the comparative rnethod ‘‘the greatest intellectual achievement” of his time.erty-space” increases the possibilities of further cross-tabulation and control without increasing the sample itself. ojie its own place. 1962). its own relation to every other. has the disadvantage of sacrificing a part of the information at the investigator’s dis¬posal.I lar uncterlying characteristic into a single vari¬able. In this context. between the po-? litical institutions of times and countries most remote frpm pne another. But his words can remind us Of the fre¬quent utility of extending comparative analyses both geographically and historically. Freeman. Factor analysis can Often be a J useful technique to achieve this objective.23. now find each . Tmis the number of cells in the matrix representing the relationship is reduced. pp. for historical cases in particular this problem is often irremediable. In his Comparative Politics. 1873). p. pp. Focus the comparative analysis on “comparable” cases. 22 (April.) . 106-17. a century ago.

1967). 56 (October. 172-75. cit. Sch^uch. and Erwin Ackerknecht. i954). Spencer. See also Erwin JC. op.” The 'method of difference con-sists of “comparing instances in which [a] phe-nomenon does occur. 1968). If such compara¬ble cases can be found. The focus on comparable cases differs from the first recommendation not only in its preoccupation with the problem of “many variables” rather than with “small N. 151.82 As Ralph Braibanti states. op. pp. “Social Aiithbjpology and tine Method of Controlled Cornparison.. by using comparable cases in which many variables are constant.experimental nature’* (op.” but also in the fact that as a byproduct of the search for comparable cases. p. although both are compatible with the second (and also the fourth) recommendation. pp. Holt and Turner refer to this strategy as the process of ‘specification” (op..”33 but it is often more practical to accord priority to the focus on a limited number of comparable cases and the discovery of partial generalizations.. Fred Eggan. pp.” this third approach focuses on the problem of “many variables. as those variables are concerned which one wants to relate to each other.-* American Anthro¬pologist. “Comparative Political Analytics Reconsidered.” While the total number of variables cannot be reduced.. cit. Whereas the first two ways of strengthening the comparative method were mainly con¬cerned with the problem of “small N. op. they offer particularly good opportunities fpr the application of the comparative method because they allow the es¬tablishment of relationships among a few vari¬ables while many other variables are con¬trolled. one can re¬duce considerably the number of operative vari-ables and study their relationships under con-trolled conditions without the problem of run¬ning out of cases." Journal of Politics. “Society as Context in Cross-Cul¬tural Comparison/' Social Science information. “The Concept of PropertySpace in Social Re¬search. 743-63." in Lazarsfeld and Rosenberg. The two recommendations thus point in fundamentally different directions. 45-50. Method and Perspective in Anthropology (Minneapolis: Univer¬sity of Minnesota Press. of difference” and as the “method of concomi-tant variations. cit.. 30 (February. It is probably also tvhat Eisenstadt has in mind when he mentions the possibility of constructing “special jntepsiye 'comparisons of a quasi. “ Ralph Braibanti.” in Robert F. “On the Comparative Method in An¬thropology. li-* 13). cit. This form of the comparative method is what John Stuart Mill described as the “method .. Mackenzie. 6 (Oc¬tober. Bar¬ton. p. 36. 113. ed. p. the number of cases subject to analysis will usually be decreased.” The M Smelser. with instances in other respects similar in which it does not. . “the move¬ment from hypothesis to theory is contingent upon analysis of the total range of political sys¬tems.M Lazarsfeld arid Barton. esp.. 20-23. pp. 424). 117-25. p. 1954). pp. pp. cit. op* cit.

222-23. therefore be used as controls. pp. 1958). cit. . the concomitance itself must be proved by the Method of Difference. this standard should be approximated as closely as possible. as “the instrument par excellence of sociological research” (p. 132). pp. . all other factors must be kept constant. Booi lll. op.35 It should be pointed out. A System of Logic (8th ed. op.”34 Mill’s method of concomitant variations is often claimed to be the first systematic formu¬lation of the modern comparative method.method of concomitant variations is a more so-phisticated version of the method of difference: instead of observing merely the presence or ab¬sence of the operative variables. *T iptirkheim.” Revue Francajse de Science Politique.38 But opinions on the utility of the area approach u John Stuart Mill. 1872). "that we may be warranted in inferring causation from concomitance of variations. As in the case of the method of difference. See also Franjois Bourricaud. it observes and measures the quantitative variations of the op¬erative variables and relates these to each other. But he hailed the method of concomitant variations. one can never be even approximately certain that two societies agree or differ in all respects save one..”87 These objections are founded on ^ too exacting scientific standard—what Sartori calls “over. chapter 7. that in looking for comparable cases. that Mill him¬self thought that the methods of difference arid of conqOmitaiit variations could not be applied in tlie social sciences because sufficiently simi¬lar cases could not be found. p. and Dyer. 251-63. . ciu. .”36 Diirkh^im agreed with Mill’s negative judgment: “The absolute elimination of adventitious elements is an ideal which can¬not r(gaily be attained. *• ilill. Green. see also Book III. 8 (June..” It is important to remem¬ber. Bock. 272.. ‘‘Science Politique et So. which he evidently interpreted to mean a combination of the statistical and comparative methods. . <>p.conscious thinking. Book VI. chapter 10. “The Comparative Method of Anthropology” Com¬parative Studies in Society and History. however. T'Tadel. Reader. Kenneth E. in Mill’s words. however.ciolojgie: Reflexions d’uri Sociologue. chapter 8. 129^-30. pp. Loadpn: Longmans. fie stated that their application in political science was “com¬pletely out of the question” and branded any attempt to do so as a "gross misconception of tlie mode of investigation proper to political phenomena. 8 (April. cit. The area approach appears to lend itself quite well to this way of applying the compara¬tive method because of the cluster of character-istics that areas tend to have in common and that can.

since certain political processes will be com¬pared between units within the area against a common background of similar trait configura¬tion”.” and points out that “the number of variables. while frequently still very large.44 An alternative way of maximizing compara¬bility is to analyze a single country diachroni.42 Comparability is indeed not inherent in any given area.caliy.wart A Rustow declares in a recent article that area. It seems unwise.”43 It is against this danger that the thrust of Rustow’s argument is directed. rather it is a quality imparted to them by the observer’s per¬spective. but it is more likely within an area than in a randomly selected set of coun¬tries. By means of an in-ductive process—a factor analysis of 54 social and pultural variables bn $2 countries—Bruce M.. for example. Within the same area) at the same time. they cite Latin America as an example of an are$ offering the prospect of “fruitful intra. First. It is not true that areas reflect merely geo¬graphic proxirnity. the differ sharply: Gunnar Heckscher states that “area studies are of the very essence of com¬parative government. some of the smaller areas may offer more advantages than the larger ones—-Scandinavia. but only where it offers the possibility of establish* ing crucial controls. which correspond closely to areas or regions of the world as usually de¬ fined.” and furthermore that “comparability is a quality that is not in¬herent in any given set of objects. the same country is not really the same at different times. Macridis and Richard Cox also argue that if areas are characterized by political as well as non-political uniformi¬ties. which have received greater comparative attention (but wJtiich dp not constitute ah area in the literal sense).area cp^parispn. to give up the area approach in comparative politics.g. therefore. Otherwise. they tend to be similar in many other basic respects.” and he shows little faith in it as a setting for “manageable comparative study. Cf. the era approach may be preferable to longitudinal analysis for the same reason. E>ank. Second.40‘ On the other hand. . study is “alihost obsolete. Rtissett discovered socio-culturally similar groupings of countries. But two im£ortaiat prpvisos should be attached tp this conclusion. or the Anglo-American countries. which has barely been exploited in this manner. not if it becomes an end in itself. is at least reduced in the case of a happy choice of area. although the con¬trol can never be perfect. the area approach can con¬tribute to comparative politics if it is an aid tp the comparative method.” He argues that “mere geo-graphic proximity does not necessarily furnish the best basis of comparison. . In this respect. Such comparison of the same unit at dif¬ferent times generally offers a better solutipn to the control problem than comparison of two or mpre different but similar units (e.”41 This is a compelling argument that shoiild be carefully considered. “the area concept will be of great value.”39 Roy C. the area ap¬proach should not be used indiscriminately. A good example of diachronic comparative analysis is Charles E. area study may indeed become “a form of imprisonment."If the area approach is often preferable to re¬search efforts with a global range in order to maximize comparability.

” American Sociological Review. 1963. 1 (January. than it would be to compare Germany as a whole with Italy as a whole. Qmntiicitiye Iriter. Causes. 7 (February.”47 As Juan J. 51531. larly good case [strictly speaking. type of governmental structure. 1959).” C&mparailye Stiedies in Society and History. Resemblances. Leslie Lipson. it is possible to move to the inter-unit comparisons to see if the same differences hold in the large.tion analysis can take advantage of the many similar national characteristics serving as con¬trols. a particularly promising approach may be the combination of intranation atid in¬ternation comparisons: “The comparison of| those sectors of two societies that have a greater number of . “The CdmparatiVe $tudy of Politics. ed. “The Value Patterns of Democracy: A Case Study in Comparative Analy¬sis. 12-31. and so on. 1968). having located what appear tp be operative factors in the in¬tra-unit comparisons.” The advantage of intra* unit comparison is that inter-unit differences can be held constant “Then. See also Riissett. pp. pp. Weimar and Bonn make a pariicu^4a Bruce M. Yet the differences could hardly be sharper. 53 (March. These two cpuntries differ not only in level of industrialization. 1953). “Delineating International Re¬gions” in J. and Variations.. 1959). 1963). Alford.Frye’s study of the empirical rela¬tionships among the party system. pp.^ Smelser illustrates the utility of this strategy with the example of a hypothetical re¬search project on industrialization in Germany and Italy.” American Anthropologist..”45 Unless the national political system itself constitutes the unit of analysis. 28 (August.tion instead of internation comparisons. pp.na. comparability can also be enhanced by focusing oh intrana. 195$). and political stability in Ger-many under the Weimar and Botin Republics.” Political Stud* ies. two cases] because there are mpre constants and relatively fewer variables than in many cross-national studies. esp. Neumanti. ‘Tor many purposes it would be more fruitful to compare northern Italy with southern Italy. pp. p. Frye argues that “for the study of these rela¬tionships. 360. 107-10. Schapera. In¬ternational Regions and the Interhdiidrial System (Chicago: Rand MbK&lly.natipnal Politics: Insights and Evidence (H<ew York: Free Press. Blahksten. See also Sigmund.” American Political Science Review. Robert R. 1967). 44 See Seymour Martin Lipset. Party and Society: The Anglo-American Democracies (Chicago: Rand Mc-Nally. Linz and Amando de Miguel point put. and I. 44 OePrge I. 317-52. I‘Party Systems in the United Kingdom and the Older Commonwealth. the interest group system. “Political Groups in Latin America. and the Ruhr with Bavaria. 55 (August. “Some Coiiiments on the Comparative Method in Sbciai Anthropology. 12$. 353—361. mvid Singer. The reason is again the same: comparative intra. but also in cultural traditions. p. Russett.

and calls for the exploration of more variables: the entire context~past> present. Heinz Eulau. Com¬parative politics should avoid the trap into which the decision- . Rokkan.) Th6 postwar division of Germany also oners the oppor¬tunity of analyzing th6 effects of democratic versus totalitarian development against a similar culiipral arid historical background. Frye. while Great Britain and the United States have many other differen¬ces. The Methodology of Comparative Research. “Within. pp. op.’’*9 Focus the comparative analysis on the “key” variables. “The New Gernmies: Restoration. Thriipp. “Comparative Political Analysis: A Methodological Note.natiori” bias of comparative research (“Methods and Models. .discovering controlled relation¬ships.”61 Scanning all variables is not the same as in¬cluding all variables. as a result. 115. 1964). 49). 22 (April. “Parties and Pressure Groups in Weimar and Botin. p. Linz and Amando d© Miguel. and it must therefore judiciously restrict itself to the really key variables. 6 (November. as long as one is pn one’s guarcj against an unrealistic and eventually self-defeating perfectionism. The nature of the comparative method land its special limitations constitute a strong argument against what Lass. 1962).well and Braibanti call “cohfigurative” or “con¬textual” analysis: “the identification and inter-pretation of factors in the whble social order which appfear to affect whatever political func. 1965). pp. Comparative analysis must avoid the danger of being overwhelmed by large nurribers of variables and. losing the possibility of. cit*. “Heckscher. and future—“must be continually scanned. Re¬construction.characteristics in cpmrnpn while differing pn some crucial ones may be more fniitful than overall national cottipari. pp. 635-55. See Ralf Dahrendorf. p. since with respect to all other variables Manitoba and North Dakota are very much alike. omitting those of only marginal importance. eds. “Diachronic Methods in Comparative Politics. (The quotation is from page 637. of course. f one is better advised to compare Manitoba and North Dakota than to compare Great Britain and the United States.” World Politics. 50. Revolutibn.tiprts and their institutional mahifestations have been identified and listed for comparispnM (Braib^nti’s definition).the presidential systems of government. 69. 49 Juan J.” in Holt find Turner.60 Lasswell argues that the comparative method as usually applied has been insufficiently cohfigurative. 397-407. 41 Smelser. pp. warns against the “whole. too.” p. the problem of “many variables*' may be alleviated not only by some of tlie specific approaches suggested above but also by a general commitment to theoretical parsimony.58.sons. See also Sylvia L.” Encounter. 17 (July. 343-58.”48 An illustrative example pf this ap¬proach in the political realm is suggested by Raoul Naroll: “If one wishes tp test theories about the difference between the cabinet and ^Charles E..” Midwest Journal of Political Science. Finally.

figurative” analysis is not synonymous With the tradi-tional single-country approach. ed. but can approximate it by focusing at¬tention on the key variables in comparative studies. op. Ap-proaches to Comparative and International Politics (Evanston* ill.. Barry Farrell. and the comparative method is also applicable in other fields and disciplines. 336-37. W. In primitive societies. pp. and the “Dulles .”56 Political science lacks this advantage. tne heaps of con¬cepts.” in R. and Burton suggests that Joseph Lapalpmbara’s call for a “segmented approach aiming at the formula¬tion of middle-range propositions concerning partial systems makes a great deal of sense. All rele¬vant factors can therefore be more easily sur¬veyed and analyzed. as in Eckstein’s defini¬tion of the term: “the analysis of particular political systems. cit. “con. other methods can often also be employed. A particularly in¬structive example is James N. op. *° Braibanti. p. ^ Parsimony Nation Differences and Comparisons: The Eight Spaliis..”6* It is iio accident that the most fruitful appli¬cations of the comparative method have been in anthropological research. op. of specifying and calling for the analysis of an exhaustive list of all variables that have any possible influence on the decision-making process. Snyder. 11). the number of variables is not as bewilderingly large as in more advanced societies. In this respect. cit.. cit. w See Richard C. and the mountains of data that seem at present to be required. A final comment is in order about the rela¬tionship of comparative politics as a substantive field and comparison as a method.: Northwestern University Press. and indeed to exist. The two are clearly not coterminous. H. Bruck. 268.. p. anthropol¬ogy can be said to provide “alinpst a laboratory for the quasiexperimental approach to social phenpmena. 49. 49 Naroll. In comparative poli¬tics. “Sciefttific Comparative Politics and In-ternational Reiations. p.63 Similarly. Eckstein’s urgent call for greater manageability of the field should be carefully heeded: “The most obvious need in the field at present is simplification-~and simplification on a rather grand scale—for human intelligence and scientific method can scarcely cope with the large numbers of variables.” in Merritt and Rokkan. treated either explicitly or implicitly as unique entities” (“A Perspective on Comparative Politics. in the field. Rosenau’s study of the relative influence of individual variables (personal policy beliefs and “personalizing ten¬dencies*’) and rple variables (party role and committee role) on the behavior of United States senators during two similar periods: the “Acheson era. 1966). 6.making approach to the study of international politics fell. In this context.” p. “ Lasswell.” 1949-1952.

a special form of the comparative method.” The method is called “quantitative” because the variables are operationally defined in quantitative terms. 5-li.” as an approximation of the experimental method. Strauss. “Discovery of Substantive lliebry: A Basic Strategy Underlying Qualitatiye Research. 8 (February. Potency of In-dividual and Role Variables ift the Behavior of U.57 The Comparative Method and the Case Study Method The discussion pf the comparative method is not complete without a consideration of the case study methdd. variables being examined. Indiredtly.** American Behavior. The great advantage of the case study is that cby focusing on a single case.’'’ One of its basic charac¬teristics is the testing of hypotheses by compar¬ing two eras (cases) that are “essentially com-parable .. . esp.sonsd qualities. then of' course the analyst inoves on to a third comparable period. .S. 17-5Q. should be closely connected with the comparative method (and sometimes also with the statistical method). If such a third or even more periods can be found—Which seems unlikely in the case of Rosenau’s particular Research problem—they should be included regardless of the outcome of the analysis of the first two eras (if the available resources permit it. isBut the case study method can and. Rosenau.al Scientist. The statistical method pan be applied to many cases. fepsenau argues that these two eras were characterized by a generally sim¬ilar international environment and that the two secretaries of state conducted similar foreign policies and also resenabled each other in ppr.era. He terms the method that he uses in his analysis the method of “quantitative historical coix^&fepn. hpweyer. of course). ed. arid “historical” because the two cases compared are historical eras. It il¬lustrates one of very many ways in which an imaginative investigator can devise fruitful ap* plications of the comparative method. because science is a generalizing activity. Glazer arid Anselm L. case studies can make M Jatufes N. Senators/* in Singer. and the case stiidy method to one case. 1965). the comparative method to relatively few (but at least two) cases. *T Se© also the proposed use of “multiple compari¬son groups. (p. “Private Preferences and Po-litical R^ponsibttities: The Relative. how-ever. Quantitative international Po^0sf pp. certain types of case studies can eVen be iconsidered implicit parts of the comparative method.. that case can be intensively examined even when the research resources at the investigator's disposal are rela¬tively limited. therefore.. 19).56 The method is. p.” 1953-1956.. in all respects except for the . pp. by Barney G. . The scientific status of the case study method is soriiewhat ambiguous. 1?. A single case can constitute neither the basis for a valid generalization nor the ground for disproving an established generalization. Itb^enau adds that if “thfe fmdings are npt so cleat as to confirm or iiegate thp hypotheses unmistakably.

and any particular study of a jingle case may fit more than one of the following catego¬ries: Athepretical case studies. As LaPaiombara emphasizes. but they also fit one or more of the other types (particularly the third. but this does not mean that they are altogether useless. Theory-confirming case studies. Deviant case studies.” pp. and fifth types) at ieast to some extent. “LaPaiombara. the dirept theoretical value of these case studies is ml. They are entirely descriptive and ruoye in a theoretical vacuum: they are neither guided by established or hypothesized generalizations nor motivated by a desire to forriiulate general hypotheses. 60-65. Theory-infirming case studies. Hypothesis-generating case studies. The Study of Comparative Government (New York: Random House. 7. " See Michael Curtis.cal case study probably does riot exist. and usually results in some vague hypotheses or cohciusipris that have a wider applicability. because almost any analysis of a sirigle case is guided by ait least aoitie vague theoretical notions and some anecdotal kriiowledge of other cases. A theoreiical case studies are the traditional sin¬gle-country or single-case analyses.building in political science.an important contribution to the establishment of general propositions and thus to theory. fourth. The first two types of cases belong to the former category. Interpretative case studies.orybiiilding. the atheoretical case study arid the other types of case studies are ideal types. It can even be claimed that “the cumulative effect of such studies will lead to fruitful generalization.” but only if it is recog¬nized that this depends on a theoretically ori¬ented secondary analysis of the data collected in athepretical case studies. p. Six types of c^se studies may be distinguished. These are ideal types. Cases m$y be selected for analysis because of an interest in the case per se or because of an interest in theory-building. An actual mstance of ^ri athepjreti. Therefore. Comparative Government and Politics: Ah Introductory Essay In Political Science (New York: Haiper and Row. See also Macridis. 1955).58 “Purely descriptive case studies do have great utility as basic data-gat!heririg opera¬tions.50 As indicated earlier. 1968). “Macrotheories and Microapplica¬tions. . the development of comparative politics is hampered by an appalling lack Pf in¬formation abotit almost all of the world’s politi¬cal systems. arid can thus contribute indirectly to the. Such actual case studies fit the first type to a large extent.

They may be particu¬larly valuable if the case selected for analysis provides what Naroll calls a sort of "crucial ex¬periment” in which certain variables of interest happen to be present in a special way. 336. In most cases. Hence they are studies in “applied science. ” Journal of Politics. statements ab6ut particular events from general statements concerning classes of events’* (p. These explanations consist of inferring. 19-36. “Michael C.0 pp. 821-37. but the case of West Irian is unique because of the complete absence of objective Dutch interests in the colony.ical case studies in one respect: they. 1966). are selected for analysis because of an interest in the case rather than an interest in the formula¬tion. Prior knowledge of the case is limited to a single 49 As Przeworski and Teune state: “The main role of a theory is to provide explanations of specific events. of general theory. . it is precisely the pur¬pose of empirical theory to make such interpre¬tative case studies possible. The Trauma of Decolonization: The Dutch and West New Guinea (New Haven: Yale University Press. pp. suich case stiidies are rare. One interesting example is Michael C. They differ. .*1 The remaining four types of case studies are all selected for t}xe purpose of theorybuilding. a general¬ization is applied to a specific case with the aim of throwing light on the case rather than of im¬proving the generalization in any way. Suph case studies are of great theoretical value.” Since they do not aim to contribute to empirical general¬izations. their value in terms of theory-building is nil On the other hand. 86) . Their objective is to de-velop theoretical generalizations in are&s where no theory exists yet.Interpretative case studies resemble atheoret. w “Naroll. “Scientific Comparative Politics and In-ternational Relations. “The Comparative Method and the Study of British Politics. in that they mate explicit use of establisheid theo¬retical propositions. “A Case of Political Under-development. In these studies. both objective (especially economic) and subjective factors can be discerned. Hyppihe^^neraiing case studies start but with a more or less vague notion of possible hy¬potheses.62 Theory-confirming and theory-infirniing case studies are analyses of single case? within the franmework of established generalizations. Hudson’s imaginative arid insightful case study of Lebanon in the light of existing development theories. and attempt to formulate definite hy¬potheses to be tested subsequently among a larger number of cases. 29 (November. See also Beer.” p. Hudson. An example of such a case study is my analysis of the determinants of Dutch colonialism in West Irian. however. See Lijphart. with a high degree of probability. top. in which he discovers a serious discrepancy between the country’s socio-economic and political development.60 Because of the still very limited degree of theoretical develop¬ment in political science.

pp. pp. arid Andr6 J.variable or to none of the variables that the proposition relates. 152-57. has quite different functions in respect to the¬orybuilding: Tlie hypothesis-generating case study serves to generate new hypotheses.. The case study is a test of the proposition. The theoretical value of both types of case studies is enhanced. 1964). eds. F. 191-96.. pp. “Sociological Law and the Deviant Case.. Robert M. 1970). Specifica¬tion should therefore definitely not be regarded as “the garbage bin” of comparative research. op. pp.” See his article “The Bearmg pf Com¬parative Analysis on Sociological Thepry. the hypothe. extreme on one of the variables: such studies can also be labeled “crucial experi-ments” or crucial tests of the propositions. ed. deviant case studies can have great theoretical value. 250-58. Kendall and Katherine M. see Conrad Phillip Kottak. 11415. in Rokkan.” in Lazarsfeld and Ffank Staritdn.64 Of the six types of case studies. the demonstration that one more case fits does not strengthen it a great deal. while the deviant case study refines and sharpens ex-isting hypotheses. which may turn out to be con¬firmed or ihfirmed by it. The deviant case study-—as ** See Patricia L. Wolf. The Politi&s of Accommodation: Plural¬ism and Dernotracy in the Netherlands (Berkeley: University of California Press. 1947). 10 (August. K6bben. or turn out to be.ometry. Each of these two types. 194?). “The Logic of Cross-Cultural Analysis: Why Exceptions?**. p.63 In this way. “The Analysis of Deviant Cases in Communications Research.” S&ci. Communications Research: i94§-49 (New York: Har¬per. and Lijphart. pp. however. . The validity of the proposition in its modified form must bfe established by further comparative analysis. to uncover relevant additional variables that were not considered previously. 12 (January. 17-53. 19^8). Ivlarsh calls • “specification. See also Milton M. Like¬wise.sis-generating and the deviant case studies have the greatest value in terms of their contribution to theory. . If the case study is of the theory-confirming type. cit. or to refine the (operational) definitions of some or all of the variables. if the cases are. " Thus process of refining generalizations through deviant case analysis is what. Sjoberg.” Social Forces. it strengthens the proposition in question. Comparative Re¬search Across Cultures and Nations (Paris: Mouton. but suggest a modified proposition that may be stronger. Gor-don.” Comparative Studies in Society and History. theory-infirming case studies merely weaken the generalizations marginally. Deviant case analyses are studies of single cases that are known to deviate from estab¬lished generalizations. They are selected in or¬der to reveal why the cases are deviant—that is. They weaken the original proposition. 43 (December. chapter 10. however. But. assuming that the proposition is solidly based on a large number of cases. 102. “Towards a Comparative Science of Society.

* tural cleavages. 41 (March.well as the theory-confirming and theory-in. Furthermore. ory-conflrming one. Eckstein simply drops it. 83-87.^ provoking case study of Norway may §erye as instructive examples. Diyision and Cohesion in Democracy: A Study of Norway (Princeton. From then on. In terms of the sixfold typology of case studies discussed above. sharp. NX: Princeton Uni¬versity Press. pp. In fact. Tru¬man’s proposition concerning “overlapping memberships. are clearly defined. that the position of the devi¬ant case on the variables under consideration. The deviant case may be likened to the “experimental group” with the remainder of the cases constituting the “control group. his analysis of the Norwegian case is only a theory-infirming one and is not made into a deviant case study. pp. Eckstein finds that the Norwegian case strikingly bears out his own “congruence” theory. The perfect fit . esp. 177-201.” Just as the analytical power of the comparative method increases the closer it ap¬proximates the statistical and experimental methods.” he ex-plicitly rules out any comparison with the cleavages in other countries. They focus on a particular case which is singled out for analysis from a rela¬tively large number of cases and which is ana¬lyzed within the theoretical and empirical con¬text of this set of cases. Truman. Some of the shortcomings in Eckstein’s otherwise insightful and thought. and persistent. of course. The Governmental Process: Political Interests and Public Opinion (New York: Knopf.”8® because Norway is a stable de. but that they fit the theory tpo perfectly. *8 David B.mqcracy in spite of the country’s deep and non. and cul. Eckstein argues that the Norwegian case deviates from David B. and consequently also its position relative to the other cases. so the analytical power of the case study method increases the more it approxi¬mates the comparative method in the fprm of deviant case analysis.firntiing case studies—are irriplicitly compara¬tive analyses. This exclusion se¬riously weakens the case study. Part of the critique which follows is included in my review of this book in the Journal of Modern History. the case study becomes a the“ Eckstein.67 He demonstrates persuasively that both governmental and social patterns of authority are strongly democratic in Norway and thus highly congruent. But he fails to place the case of Norway in relation to other cases. Such case analysis re¬quires. 1951). al¬though he describes Norway’s divisions as “as-tonishingly great.overlapping geographic. The problem here is not that the Norwegian facts do not fit the the¬ory. The different types of cases and their un¬equal potential contributions to theorybuilding should be kept in mind in selecting and analyz¬ing a single case. economic. instead of trying to refine Truman’s proposition with the help of the deviant findings. 6077. 1969). 1966). which states that govern-ments tend to be stable if there is considerable resemblance (congruence) between govern¬mental authority patterns and the authority pat¬terns in society.

‘'Concept Misforxnation in Com-parative Politics. 12. as Eckstein himself suggests. NJ. Black: “There is much greater value in comparing contemporary events and institutions than those that are widely separated in time. “Coxnparative Politics/' in International Encyclopedia of the Sdcial Sciences. but does not contribute to its refinement. Perhaps it should be Called a “hypothesis-strengthen. The comparison of societies or smaller groups that are concerned with reasonably similar problems is more likely to lead to satisfactory conclusions than comparisons between societies existing many Centuries apart” Black. 75r78.68 Because the Norwe¬gian case turns out to be a perfect theorycon-firming one. In one respect.” American Political Science Review. *** The comparative method and the case study method have major drawbacks.strengthens the theory margin¬ally. cit. of course. following statement by C. it is* not altogether correct to call the Norwegian base study a tHeoiy-cbhfir^mirig study. Thus. E. 1961). and he fails to take full ad¬vantage of the case study method in analyzing the case in terms of Truman*s theory of over-lapping memberships. 331-36. The Dytiafitics of . Because the congruence theory has a father narrow empirical basis. they pan be highly useful instruments in scientific political inquiry. September 1969). it is a hypothesis rather than an established theory. Vol. A Theory of Stable Democracy. see alsb pp. p. 1033. The case stiidy of Norway is. pp. The theory does not hold that complete congru¬ence of authority patterns is required for stable democracy. 10 (Princeton. ** Eckstein. 64 (December. not a hypothesis-generating study either. op. it cannot be used to refine the the¬ory in any of these respects. Therefore. 2 Giovanni Sartori. In his original statement of the congruence theory. consisting chiefly of only two cases (Britain arid GerManjr).t pp. T KaUebetg. But precisely because of the inevitable limitations of these methods. it is the challenging task of the inves¬tigator in the field of comparative politics to apply these methods in such a way as to mini¬mize their weaknesses and to capitalize on their inherent strengths. 1970). ory is concerned.' mg” case study or. a "plausibility probe” (oral comment at the IPSA Round Table Conference in Turin. Eck¬stein was unlucky in his selection of this case as far as the development of his congruence the-. Eckstein himself points out the necessity of further work on the important questions of how much disparity can be toler¬ated and how degrees of congruence and dis¬parity can be measured. Re¬search Monograph No.: Center of International Studies. 72-73.

Sapin. *1962).Moderniiatidn: A Study in C&mparatiye History (New York: Harper arid Rpiy. 1966). 39. eds. p. . Foreign Policy Decision-Making (New York: Friee Press of Olencoe..