89 In amuary 17 the align pine mine, Giulio Andel, epored to patoment that 120 ces of ola vikne had ecu inthe peeing yer i 69, Sani, “Selar Tends and Party Realignnents” 81, The percentage of respondents going to chuch weekly has dropped from a high print o 70 percent in 156 to approximately $0 percent in the 17 $2. Sn, “eur Trends and Party Re Alignments” ‘93, Ricerche Demascopiche 2 (1974); Bollettino DOXA, $9, ns. 7-8(1975). 4, Ree changes in the image ofthe Communit party are dcumeted in my esay, “La nora immagine dl PCIe Feletorat ting” in Dold L Bachmer and Sidney Tarow cs, UI Comunismo in Hala ¢ in Franca (Mian; Ea), 17) 45, Ricerche Denosopiche 7 (73). $6, Thi arguments developed in my say, “The PC oth Theo" Problems of Communism (Norember-December 1879, pp. 27-1, CHAPTER IX Political Culture in Mexico: Continuities and Revisionist Interpretations Ann L, Craig and Wayne A. Cornelius Univnty of Califo, Sm Diego Gaanret Atatoxo avo Sidney Verba’ eleventhhour decision to substitute Mexico for Sweden in their cross-national survey re- sulted in the fst lngescale empiric study of police ele in Mexico. Unil then, the dscusion about pois and politcal culture in Mexico revolved around eloquent, reflective esays based on subjective interpretations of Mexican history or very limited psychoanalytic data from clinical experiences. Almond and Verba's work shifted the discussion to empirical studies based on research explicitly designed to explore Mexican att tules and copitons which might inuene poll pari tion. Nearly all survey studies of political attitudes and behavior conducted in Mevico since 1959 have atempte to replicate (though usually in modified form) some portions ofthe Cixe Culture survey, And even among Mexicanists whose own work thas not been influenced by the study, it has generated consider. able debate about the theoretical tility and normative asump- tions of the study of political culture. In reviewing this debate Copyright @ 1980 by Ann L. Craig and Wayne A. Cornelius3 Political Culture in Mexico in the Mexian context, we must begin by acknowledging that the posbility ofthe debate depends largely upon having studies like The Civic Culture as specific points of reference. ‘This volume is designed as.a critique of The Civic Culture and a summary of subsequent research on politcal culture in each of the five countries included in the original study. In fulling this dual task, we have devoted the fist section of this chapter con Mexico toa discussion of the principal methodological and substantive weaknesses of Almond and Verba's treatment of the ‘Mexican case, In the second section, we tum our attention to three general themes — authoritarian politics, subjective politi- cal competence (politcal eficay, and political cynicism —in recent empirical reearch on political attitudes and behavior in Mexico. These themes also provide useful points of comparison with the Civic Cullure analysis In the concluding section, we ‘outline some of the central issues which might be pursued in further research on political culture in Mexico, MEXICO IN THE CIVIC CULTURE: A CRITIQUE ‘We have carefully reviewed the original volume with attention to the major methodological and interpretive problems. Some of the methodological problems emphased here became ap- patent as researchers have become sensitized through experience with sample surveys conducted in Mexico since 1959. Some of ‘our criticisms address weaknesses which are an outgrowth of the complexity and technical demands of cross-national survey re search. We focus on three major areas of methodological difficul- tis: the sample, transaton ambiguities or errors, and problems of validity or equivalence. These problems are covered in some detail because they involve the accuracy or generalizability of some interpretations in the Almond and Verba study. The first half of this section is devoted to these methodological problems. In the second half, we tur to substantive criticisms of the Mexican analysis, METHODOLOGICAL PROBLEMS The Sample. In Appendix A, we are informed that the Mexican sample for The Civic Culture was drawn in 27 cies with populations of 1000 or nor, dispersed throughout the Political Culture in Mexico "7 nation! We assume that the sample was limited to this urban population because ofthe cost and relative dificly of sampling and interviewing rural dwellers. Nonetheless, the exclusion of rural dwellers which constituted 63 percent of the Mexican population in 1960), produces @ major gap in the data and requires serious qualification of many interpretation, ven in ros-national comparisons, In a footnote on lalian—Mevican iferences in politica cognition and communication, wear tld thatthe Mexican sample was compared with tat part of the Ttaian sample living in towns of 10,000 an! larger, "The results differed only by a small number of percentage points and con firmed the conclusion discussed here." * This statistical double check asures us that on the point of politcal cognition the comparisons between the Mexican and Italian samples are valid Similar checks are not reported elsewhere in the study, despite the possibility that other comparisons involving the Mexican results may have been affected by the exclusion of rural dwellers from the sample. In the absence of extensive empitical research on political culture in rural Mexico, we can only suggest some posible consequences ofthe urban bias of the sample. In Mexico there are great advantages accruing to urban residence, in tems of ‘acces to basic services, education, and opportunities for eco- nomic improvement. As Gonzilez Casanova has written: “There isa Kind of integral margnality. The population which is mar inal in terms of one factor is highly likly to be marginal in terms of all the others, Thus there is an immense number of Mexicans who have nothing of nothing.” * Access to political information and personal conact with poital and gover: mental oficial, especially ouside the arena of eecoal politics, are also strongly associated with urban residence Opportunities for involvement in most types of politically relevant onganiza- tions pari, labor unions, community action group, ek. — are also relatively more numerous in urban areas. In short, the average urban dweller is likely to be much better endowed with the opportunities and resources which translate into politcal involvement than is his rural counterpart. His sense of political ¢fficacy and more generally his image and evaluations of the government may therefore differ significantly. Owing to the much sreater visibility of poital and governmental activites in ut8 Political Culture in Mexico. ‘an areas, the urban dweller may be better able to distinguish among political parties, government agencies, and incumbents | of certain politcal roles, ‘The rural population is more strongly influenced by the Catholic Church and its enduring adversary relationship with the civil authorities. In general, then, much of the diversity in Mexican politica culture may have been obscured by the urban bias of the sample, In adttion, by excluding the rural population, the Civic Culture sample alo introduced an ethnic bis, since the vast majority of Mexicols Indian population is beaily concentrated in small rural com ‘munities Furthermore, although the Mexican sample is described as a representatie sample of the national urdor population, 2 carcul reading of the description of the sampling procedure ‘raises questions about the reptesentativeness of the urban sample, We are told that “the assignment of interviews in each city... . ‘was made through the use of the latest block maps available,” * Our experience has demonstrated that avallable Block maps are | invariably several years out of date and, most significantly, ‘exclule major, low.income zones ofthe urban periphery which ‘are usually the most recently setled areas, often without urban services, street layout, or dicrete Blacks of housing! It is pas sible, therefore, that not only rural dwellers but also a significant. portion of the urban lowincome population were exduded from the Give Culture sample Another source of bias in Almond and Verba’s Mexican sample is the high rate of sample mortality, The rate at which inter- views were completed with originally selected elements of the Mexican sample—only 60 pereent— compares. unfavorably with the rate achieved in three other countries included in the study (4 percent in Germany and Italy, 88 percent in the United States), and with other recent survey studies of political attitudes and behavior in Mexico The Translation. Our examination of the English and Spanish versions of the questionnaire used by Almond and Verba re vealed several apparent Engst Spanish trandation eros, which may have aected the responses obtained in Mexion. The sot significant Spanish deviations from the Engh equivalent are presnted in Table 1.1. Inthe ce of item 7, the second Political Culture in Mexico 9 statement presented othe Spanih speaker s much more strongly worded than the English original (“Others say that one should dlstust the majority of people”, “Others say you cal be too carefl in your dealings with people’). In items Sla through 32 the English version asks about government “activities,” whereas the Mexican respondent is referred to government “laws and regulations.” In a country where laws and regulations are relatively unfamiliar compared with government projects and polices (a more likely connotation of “activities”, the differences in terminology may have aflected response pattems, The most significant translation error occurs in item 36. In Spanish, the respondent was asked whether the ideals and goals ofthe Mexi- can Revolution had been realized, whether “it is still necessary to work in order to realize them,” or whether they have been forgotten, The English version of the second alternative aks whether "the people holding {the ideals and goals of the Revolu tion—preuinably government officials] ave sill working to reuie them.” In this instance, the Spanish version probably inflated the proportion of respondents selecting the second alter native, Sixtyone percent of the respondents chose this alterna tive, Almond and Verba’ interpretation of the responses, stresing the relationship between this question and other expresions of symbolic identification with the Mexican Revolution,” suggests that they were unaware ofthe translation error. Validity and Conceptual Equivalence, Several questions in cluded in the Ciuie Culture quevionnire may have diferent meanings or experiential referents in. Mexico compared with the other four counties incuded in the study, A-number of thes items em to be inappropriate to the sxioplitical con texto subject toa variety of interpretations by respondent ‘Two ofthe most important items of questionable appropriate ness form part of the pliccal competence sae, which plas an important role in the Civic Cullae analysis. Questions % through 29 asked the respondent to consider what he might do ifthe locl and national governments were considering a law or regulon thatthe respondent considered unjust or harmful, the fielbood that he/she would actually attempt to secure a change in the law or regulation, the probability that this attempt would succeed, and whether the respondent had ever attemptedat Gnamqonin 2p) C 2p pepe sy ap) omnes, fa tt Weore onues gn consmyonsyy ‘ze 2m) Cop BEpNP step) oulsqod [a us opuesusd exe vie eis ‘ Se deguce ppend otin anb wasp senoried eearBty ‘ sou wary = Soptioa ieyeamela ou wont sapounornon® 23RD PMID 247 euOLs musT pores fo uoMeea YsnTug pur ysrucds Jo eduog -VXE HWE 30ae Politica! Culture in Mexico such influence. In Mexico 52 percent ofthe respondents believed that they could do something about a local regulation and $8 percent believed they could do something about a national regulation. These figures are surprisingly high when one con- siers the way in which such regulations are actually established ‘or modified in Mexico. The highly centralized, presidemtally initiated patter of decision making in Mexico andthe general absence of publicity about poicy deliberations have the effect of shifting citizen infuence on government decision making from the “rule making’ to the “rule application” (distributive, implementation) stage. Approaching public officials to seek par- ticalaristic (personal or local communityspecific) benefits to ob- tain panticularitic “exemptions” from public policies or regu- lations is both permitted and encouraged by government officials; but demanding major changes in public policies ("hws and regulations’) is viewed as threatening and illegitimate activity The average cten seems quite avare of the kinds of plital action which are likely to be rewarded by the authorities and. the kinds which are likely to be ignored or violently repressed. He therefore auempts to influence the government decsion making proces only on the output (policy implementation) side" In view of this, we question the utility in the Mexican context of a measure of political competence which ostensibly focuses on citizen influence on policymaking to the exclusion of perceived influence on allocative decisions, Almond and Verba's interpretation of Mexican politcal cu ture as containing “a large aspirational component” * depends in part on data demonstrating that Mesican are willing to expres politcal opinions but ave lite information on which to base them. This analysis relies on the responses to questions whose appropriateness for measuring political information levels in the Mexican context is doubtful, It would seem most useful to ‘measure political information as that knowledge which respon- dents would require for the most common or essential forms of political involvement; that is, basic and useful information, If this tandard were applied, how appropriate isi, for example, to limit measures of political information in Mexico to the ability to name party leaders or cabinet ministries (‘ecrelarias 0 puestos del gabinete")? The first is a rather esoteric piece of information, since the president of Mexico is also the de facto, Political Culture in Mexico 8 head of the official party (Partido Revolucionario Institucional — PRI), and the strength and visibility of opposition partes ae tore regional than national. Such questions could be answered with greater facility in counties where there are numerous promi tent party Hades with nationwide constituencies, so thatthe respondent's level of information need not be limited by the ability to recall a very limited range of alternatives, Further: amore, occupants of most ministerial posts are not key political personages for most Mexicans, but even some of the reasonably wellinformed respondents may have been confused by the uncommon vord usge ("gabinete’), We do not have any estimates of the functional literacy rate for the Almond and Verba sample. According to the official ‘Mexican census in 1960, about 75 percent of the national urban population yas “literate,” but it is widely acknowledged that the oficial census statistics include huge numbers of people who ate actually functional illiterate? Our concer here is whether functional illiteracy may have reduced the ability of numerous interviewees to respond to a substantial number of items included in the Give Culture questionnaire. The survey instrument con tains fourteen different questions which asked respondents to read alist of alternatives from a card provided by the interviewer and to eect among them. They include items asking about the most effective technique of influencing government decisions (30a), the characteristics of political party supporters (6), pattems of decision making within the family (4a, 4b, 55a, 8b, 5c), the rost important problems facing the country (fa), and other topics (62,11, 17, 4, 65, 86), In all of the more recent surveys of political attitudes and behavior in Mexico of which we are aware, inerviewers were required to read and, if necessary, reread to respondents the alternatives for closed, multiplechoice items. No asumption of literacy was mate Also problematic in the Mexican context are two additional items used by Almond and Verba as measures of politica avare ness; questions aking the respondent to assess the impact of goierament outputs on his daily life (ee items Sha and 82a in Table IX.1), The authors report that two-thirds of the Mexican respondents felt that rules and regulations of the national and Joaal governments had no impact on their daily life. In the other four countries, the respondents who perceived no efletaH Political Culture in Mexico of government on their daily Lives ranged from 10 0 28 per: cent. In their interpretation of the data, Almond and Verba suggest that a respondent who believes that the local andar national government's activites have no eft on his dayto-day life is uninformed! or rejets the significance of government output Infact, when one considers that the Spanih transla tion of these items asks about lows and regulations rather than government projects or activi, i is not suprising that rela tively few Mexican respondents perceived any relationship be: tween goverment and ther dil lives. Even if the question were understood to refer to government programs and activites, given the extremely unequal distribution of government benefits and services in Mexico, the responses obtained may reflect highly accurate perception of actual government impact, rather than a Tack of “awarenes.” This is particularly true with regard to local government outputs, which outside of Mexico City, Guada- lajara, and one or two oiber large cites, are of ‘minimal ‘importance. Teis perhaps inevitable chat the country specialist will peer rote contextually sensitive measures of certain concepts than the crossnational researcher is able to supply. The measure- ment problems discussed above were selected, however, with a view toward highlighting the weakneses thewe problems intro: duce into crossnational comparisons based on the resulting data, We donot seek to minimize the importance ofthe Mexican responses to some of the questionnaire items we have singled ‘out for criticism here. A strong possibility remaitis, however, that in the Mexican context these questions may have tapped rather different dimensions from those Almond and Verba in- tended to measure, or that they provide only very pari ameasures of Key attitudes and cognitions SUBSTANTIVE CRITICISMS: PATHS NOT TAKEN In this section we shall consider what we regard as two major substantive deficiencies in Almond and Verba's treatment of the Mexican cae (1) a lack of attention to important witia-nation variations, expecially in terms of social clas and ‘region; and @) an excesively truncated examination of the proces of politcal socialization, Both of thse analytical paths, had they been fol- lowed by Almond and Verba, might have led the authors to Politica! Culture in Mexico 385 interpre thir data wth pete elit the dynamic of polit al life in contemporary Mexico WithinNation Variations Students of politics in all fie of the nations included in The Civic Culture have been critical of its failure sufficiently to examine internal or subnational ci ferences. Such criticisms must be tempered with the acknowledg. ment that inclusion of the analysis critics find lacking would have resulted in a multivolume work, We believe, however, that 2 thorough examination of the diversities within each country would have portrayed more accurately the complexity of the national political culture (or cultures), thus satisiying important theoretical concerns and perhaps even amenting Almond and Verba conclusions about democratic participation and stability and the political culture which supports them. The Civic Culture dogs contain some examination of internal differences based on ed uation, party support competence levels and type, sex, age, and level of participation in school and family decision making, How ever this analysis of subnational diferences seems inadequate par ticularly forthe Mexican case, in two respects: the limited analysis, by socal cas, an the failure to examine regional variations First, Almond and Verba's presentation of the data fails to emphasie, as subsequent research in Mexico has demonstrated, that there isan intimate relationship between social cas, edu cation, and sex, on the one hand, and politcal copitions, ati tudes, and behavior, on the other. We consider the absence of catenive class analysis a particularly serious omision based on cour reading of subsequent research, although not necessarily because behavior or attitude differences may inhere in class statu itself, More important is the inleraction of sex, income, education, occupational status, and family postion, which com bine in Mexico to determine ones resources for political partic: pation a5 well as the terms of individual contact with polities and government. The result, as described by Fagen and Tuohy, is “sharply cumulative patterns of opportuni, resoures, and [political] activity,"** which perst over time because of the absence of corrective mechanisms. We shall elaborate subse quently on the effects of the social status variables as demon- stated in recent research on authoritarianism, political elicacy, and cynicism,86 Political Culture in Mexin In some respects, Almond and Verba's extensive use of educar tion as an independent and control variable might be taken as quasi-class analysis, were it not that their discussion of the re- sults implies that it is education itself that produces the dif- ferences in attitudes and thehaviors which they observe, The intetrelationships among education and other social status vati- ables alluded to above suggest that the attitudinal and behavioral differences observed may be the result of more complex social processes than Almond and Verba's analysis would indicate, Evidence from recent surveys tend to support sucha conclusion Coleman reports a.65 correlation between education and family income in his sample of Mexico City resdents** Fagen and ‘Tuohy found, in ther Jala sample, a coreaton oft be | tween eduction and a css index baed on the respondent’ income and the interviewer's estimate of socioeconomic status, ‘The same authors report a comeaton of between occupation and eduction, The research in Jalapa also reports marked differences in educational achievement by sex: 56 percent of the ‘upper-class women, compared with 85 percent of the upper-class ren, hae! more than primary education; the figures for the middle class are 18 percent famong women) and 36 percent {among men}, and for the lower clas, 7 percent and 17 percent, respectively.” Fromm and Maccoby report a .55 correlation beaneen family scoeconomic status and sons scooting, and 2M conelation between daughter’ schooling and the same class measure. ‘There is, therefore, some suggestive evidence that being male, of middle or upperdas background, and residing in an urban area contribute powerfully to educational advantage in Mexico, This advantage has reprcusions on pital culture, most ‘immediately on information and interest in politics. "Consumers of politically relevant information constitute a very small pro portion of the total Mexican people. These consumer, .. the concemed public’ ate principally urban, literate, Spanish speaking mestos"* Coleman preents evidence that, even in ‘Mexico City, the ability and the desire to acquire information about politi increas significantly with education and fail income level. He was particularly concerned with knowledge related to elections ot elective ofice in Mexico, testing his re: spondents on questions about the terms of elected offiials, the Political Culture in Mexico wt patty afliaton of political personalities, andthe sectors ofthe PRI. Of his two samples, the upperclas sample was far beter ‘informed than the random sample, with differences of 20 to 80 percentage points in the percent of comet answered. ques tions ‘This evidence on Mexican adults is paralleled by the results cof a recent study of the politicization of Mexican schoolchildren, Rafael Segovi's analysis of data from si states relies heavily ‘on comparisons between groups of schoolchildren defined by father’s occupational status, Segovia demonstrates that as the father’s occupational status increas, children increasingly report discussing politic, a fact which he attributes to greater interest and information in higherstaus homes, Similarly, higherstatus children could more frequently name sate governors and could comrectly answer other informational questions! Segovia di casas predictably similar findings in his comparisons of children ‘attending public and private schools, showing that attendance in private school is associated with much higher frequencies of ds casing polities at home, with parents, and with pers, and with Higher levels of information, pariculnly in primary grades where it might be expected that factors outside the schooroom axe particularly influential These resuls, and those summar- ined in the second half ofthis chapter, sages that simple cross tabulations of education with individual item responses a5 used in The Givi Culture do not adequately convey the cumulative impact of educational and clas diferences on politcal cognitions and attitudes in Mexico. Another shortcoming of the subnational analysis provided in The Civic Culture is its lack of attention to regional ferences Mexicans and students of Mexico ae very avare of the pro nounced differences among regions in that country in terms of economic developmen, landtenute pattems, occupational struc tur, etic, eis, politcal partipaon, relaconships with the federal government, and other dimensions There is alko some evidence of impottant intraregional variations in patterns of community participation, interest-group formation, and competitiveness of polis, associated primarily with intra regional diferences in resource base, economic development, and government contact There is every reason to belive that there may be an interregional parallel for these findings.58 Political Culture in Mexico So far, however, systematic regional comparisons of politcal culture rarely have been undertaken, owing to the lack of comparable data from diferent regions. Segovia’ data, which do permit such comparisons, demonstrate that the severe in- equalities in regional and tural/urban development in Mexico are reflected in the amount and accuracy of the political infore ration that children posts?" Almond and Verba provide only 2 hint of regional diferencs: ‘There is some difference within nations on the regional distribu tion of sores on the subjective competence sal, but thee di ferences parallel other differences between regions in levels of rational competence, politic activity, and the lke, Ths itis probable that repona diferences in attitades toward lcs govern sme ar funtion, not so mucho ferences in attuds tard the local goverment, but of ferences in general attitudes tovard goverment and pols Analysis of regional variations in the Cite Culture data for Mexico may have been limited somewhat by the heavy concen- tration of respondents in the central and northern regions of the country (66 percent and 2%. percent, respective), but the analytic posites remain largely unexplored Political Socialization and Participation. At several points in their analysis, Almond and Verba refer to the importance of adult experiences as being equally or potentially more significant than childhood socialization in understanding political culture. Although they do consider childhood (amily and schoolrelated) experiences, they suggest that the recency of adult experiences and their more diet relationship to plitis and citien-gover- ‘ment relations make adulthood a fertile field for socialization research. Their recognition of the importance of adult learning ‘experiences represented a significant advancement over “national character’ and personality studies, which tended to discount the importance of learning alter childhood Tn dealing vith politcal socaiation during adulthood, however, Almond and Verba limited themselves primarily to workelated experiences, We would argue that, at leas in the ‘Mexican case, the study of adult political socialization could very usefully be extended beyond the workplace, to encompass the influence of the local community or neighborhood, peer groups, Political Culture in Mexico 39 and individual or collective experiences in dealing with political and governmental agencies, The importance of the local com munity as an agent of adult political socialization has been demonstrated in research conducted in both urban and rural areas of Mexico. There is also considerable evidence suggest ing that personal experience with government structures and the oficial party contributes not only to the development of cognitions but also to afective and evaluative elements of politi cal culture in Mexico. Attitudes such as political eficacy (or subjective politcal competence), political eyicsm, trast in government, affect for national politcal institutions, and evalua tions of government output performance often seem to have an important experiential basis” Fuller consideration of the process of adult socialization would also have revealed the importance of various nonelectoral forms of political participation in Mexico, In the Mexican context, for example, voting is relatively unimportant 2s a political learning experience compared with individual and group peti- tioning for government benefits, involvement in pattoncient relationships, and participation in oficially sponsored. group gatherings to demonstrate support for the regime: Unfortu rately, these modes of political participation receive virtually no attention in The Civie Culture, Almond and Verba did ‘include in their survey several items intended to measure “atten- tiveness to politcal input.” which they regard as a degree of involvement in the politcal input structure of government." Their measures of political involvement are therefore limited to questions about frequency and perccved freedom of dicus- ing politic, attentiveness to and feelings about political cam- peigns, perceptions of ctzens’ responsibilities for community involvement, nd an asorment of questions about partisanship, Tn shor, they are questions which esenlly tap cogitzein- Yolvement in politics and public af Finally, at no point do Almond and Verba examine the various forms of “deviant” political behavior — involvement in protest demonstrations or anistem political movements, partcpation in urban or rural land invasions, politically motivated violence, and other behaviors which place the individual in direct con- frontation with the regime. Although still relatively rare in Mexico (xcept for participation in land invasions) such acts0 Politica! Culture in Mexico are quit significant when they do occur; and the characteris ‘and motivations of those who engage in them have not received. adequate attention in empirical research." ‘The multidimensionality of politcal participation and the impact of various participation experiences on attitudes such as perceived politcal efcacy and eynicom are discused at greater length in the following section of this chapter. At this point wwe wish only to underscore the fact that the narrow range of participatory behaviors and socialization experiences considered in The Cisie Culture excludes many ofthe experiences of ctaen interaction with government which have helped to produce or reinforce citizenship attitudes that Almond and Verba attribute to more individualistic phenomena, MEXICAN POLITICAL CULTURE: A PROFILE FROM RECENT RESEARCH In comparing Almond and Verb's concusions about poi cal culture in Mexico with the finding of subsequent research, vie must be specific about our working dentin of politcal culture, We stall consider the politcal culture of a group to be the set of cognitions, perceptions, evaluations, attitudes, and lwhavioral predispositions through which member individuals and/or subgroups order and interpret politcal institutions and processes, and their own relationships with such institutions and process, I shouldbe noted that his definition encompases cog nitive and behavioral components of politcal culture as wel as the more frequently studied value and attitudinal dimensions. General studies of political culture have been infrequent and usually unpopular among Latin Americans and Latin Ameri canists. Many Latin Americanists deliberately chose to avoid. broad national studies of political culture, generally for one of two reason: (I) that internal divenstes based on ca, ethnicity, region, ruralurban diferences, or diferences of participation opportunities militated against the possibilities of arriving at ‘meaningful, national generalizations about political cular; and (2 thatthe basic terms defining the context and outcomes of political activity were determined more by political structures ‘and economic relations than by values or attitudinal orienta- tions and that, therefor, the sigfcant research questions ly ‘outside the realm of political culture. Although Mexico has been Political Culture in Mexico ul studied by more social scientists than pertaps any other Latin Anetican nation, the Mexican literature is far richer in ethno- graphic materials than in research relating to political culture ‘Most of the ethnographic studies make only passing reference to politics and political values, Among the survey studies we shall cite in this section with reference to selected aspects of political culture in Mexico, The Civic Culture and Rafael Segovia's study of the politicization of Mexican children are the only studies based on samples not confined to a particular region or locality. Another group of studies to which we stall reler has focused on public polices, political change, or int tutional processes at the national level, using data from sub- rational studies of Mexican political culture to help explain these phenomena, Early politcal culture studies on Mexico (oughly prior to 1968) were based largely on fick! observations, clinical data, newspaper reportage, and elite interviews, conducted primarily in Mexico City, Statements about political culture tended to parallel the national character literature, stresing certain psycho: Topical variables or personality characteristics as obstacles to evelopment or modernization. These studies were concerned mainly with questions of national and ethnic identity, integra tion, and nationalism. Their explanations of Mexican politics and society tended to stress psychological, racial cultural (mis- cepenaion), religious, and historical factors More recent scholarship largely bypass this earlier tradition of political culture analysis, regarding it es impresionst, un able to deal with questions of change, and inattentive to key Structural variables, This second wave of research generally avoids explanations which rely on personality, turing instead to attitudes and behaviors for which fundamental explanations ate most often sought in structural or temic features, Te tends to be both more micro- and more macroanalytic in approach: tmicroanalytic because itis baed on surveys ot indepth ite: ‘viewing of individuals clustered in a limited number of com- munities or regions; macroanalytic bectue it attempts to explain how the politcal sytem operate, Among the contbu tors to this literature are a number of researchers who have ‘conducted studies of “community power” or local community polit The principal survey studies conducted in Mexicowe Political Gultwre in Mexico since 1959 which have any appreciable politcal content are listed in Table 1X.2. Since surveys re, by virtue oftheir scope and design the most appropriate research instruments for making statements about the distribution of attitudes; cognitions, values, and behaviors within the populations from which their samples are drawn, most of the data to be discussed in this section will be drawn from survey studies. 1 is important to be aware of several features of the surveys conducted in Mexico, however, Of the studies included in Table IX.2, seven were done in urban areas (all but three exclusively in the Mexico City metropolitan area). All of them include respondents from two or more social classes, bt four of the samples are predominantly low-income, Seven include both males and females, All but one are studies of exclusively adult populations. Of the seven studies which include 2 rural ot provincial town sample all were conducted in rela tively accessible localities which could not be considered isolated. or primarily Indian, Although these surveys were conducted over a period of eighteen yar, they donot provide a basis for preci statements about changes in Mexican politcal culture over time, owing to differences in sampling sites, sampling procedures, and sample composition, Comparative analysis of the survey results is aso limited by significant differences in question wording, response alteratives, and coding categorie, With thee caveats in mind, we tarn now to dicwsion of three broad themes emerging from this body of research which ate important to an understanding of the basis of political involvement in Mexico: authoritarian politic, subjective politcal competence (or eficac), and politi cal gniciam AUTHORITARIAN POLITICS Almond and Verba selected Mexico as “one ‘non-Atlantic community’ democracy,” a country in which the democratic politcal stem was relatively new, but in which the people's hopes and aspirations for democratization and. modernization were high?” Given Almond and Verba's theoretical concerns with democratic political panicipation and the stability of democratic systems, as well asthe prevailing academic image of the Mexican sytem at the time they were iting, i isnot su 3 8 é 2 5 : 8 s g Sample Surveys of Potiteal “ramen 1X2. Sampte 1 950% of vata s5e4 e wspoe -Aoeoura 1 Kouoiuon pus Sees ee Be eats - nies suet oumuon suoqin ysesead 9 song wire pone srmEO, Recon eee tare See ee ale ey oe oor = eee Jop1o paw et somom pus Uap st O2Pore sot pe “ueumstod pus cous iteyrenanepont ep erensnamean ce we sucpepes Jo sapwiuoeaidos oa "o) borers dpanpeodaznd sopranos ore 108 (oor) Z001 are exer somsrowsen Tames Temdoee manure some fouwonM6 Political Culture in Mexico Political Culture in Mexico M0 pring that they cose to compare their Mexican data with a model of democratic systems. Since 1968, however, students of Mexico have increasingly stresed the authoritarian features of Mexican pits and he formiabe obstacles wo genuine demor. ratation ofthe gstem.* ‘Fhewe scholars argue that although earlier optimistic asesments of politcal democracy in Mexico ecgnzed the high concentration of polit pover, they tended | sees one ofthe mos etd explorations ofthese topics toconve) a basicly pluralist image of competition and interest ithe Nfeican content. Fromm and Maccoby found that the agzregtion within the oficial pny, and were therefore mit 1 yy frequent mode of socipoitcal elation in ther village was Alirected at bet, fundamentally inaccurate at wort” “ubmisivenes” (49 percent ofthe residents) followed by “ta Most staies of authortaranisn in pols have employed ions authority relations (20 percent) and “authortarinisn” ane of wo approaches (I) the study of individual atades oF geen) nly percent ofthe vile’ population exhibited behav which re inert “amor and@) the) redonnanydenocatcendencs By "smite" the stucly of authoritarian pattems of decision making and politcal | 2 hors refer to the “lack of hope, and fatal about the foe rgaizton, In general the fist, “psychological,” approah "ture the feng of poweresness to change events... atributes more weight to personality and politcal culture (in- uding nonrational and individual phenomena) as the basis of authoritarian politi. The second, “Structural,” approach relegate politcal cute to a telavely more drvative or | supportive status, emphasizing instead authoritarian structures, the tactics of rung elites, and rationality as the basis of political behavior. the bases of interpersonal and leaderfollower relations and pref- erences for certain patterns of social organization in Mexico, This research deals with such topics as power relations (wb missivenes and dominance), cynicism, intolerance of ambiguity or opposition viewpoints, and fatalism. Fromm and Maccoby's Social Character in a Mexican Village submisiveness to nature and God's will... the submissive “nes of young ool, women to men and the por tothe rich" # “Authoritarian” individuals are characterized by a concern with obedience or disobedience in children, admiration for persons who have power and we fore, and ¢ tendency to justify in equality by attributing merit to strength. Fromm and Maccoby contrast this form of authoritarianism with traditionalism. The traditional peasant stresses custom and tradition in his desire for respect and obedience and in his acceptance of dominance relations. The extremely subtle distinctions drawn by Fromm and Maccoby among submisivenes, authoritarianism, and tra ditionalism are often problematic. On the basis of their guide. lines it would be difficult to determine whether peasants’ re- sponses to “authoritarianism” items represent expresions of individual needs and preferences or simply rection of ta tional patterns of authority relations, The Fromm and Maccoby study represents a significant eft to link social character to the evolution of political and eco- ‘nomic structures in rural Mexico. They base their conclusions on historia data, participant oberation, and data gathered with a standardized questionnaire, They conclude that the character orientation of the villagers is adaptive to the mode of production; the social character motivates to behavior which fulfills socioeconomic (and, we would add, political) functions. In general, they found that landless day laborers have a recep Prchological Interpretations, In their seminal study, Adorno and his associates set forth a theory of the “authoritarian pet- sonality." * They conceptualized authoritarianism as personality syndrome characterized by implicit anidemocratic (pefacist and ethnocentric) tendencies; that is, the authoritarian person: ality type has a predisposition to glorify, be subservient to and uncritical of authority figures in his “ingroup,” and to punish members of the “outgroup.” Adorno described nine dimensions of authoritarianism, and designed an “Pscale” composed of agee/disagree questionnaire items to tp thee nine dimen sions The Adorno “F-cale” and the concept of “the authori- tarian personality” have been widely criticized in social science on both methodological and conceptual grounds. Except for Fromm and Maccoby,** students of Mexican politics do not refer to Adomo; yet some of the recent research clearly seeks to tp several ofthe dimensions of autboritrianism as defined by Adorno (albeit under diferent Ibe), inorder to explain48 Political Culture in Mexico tive, nonproductive orientation; free landowners tend to have a boarding, productive orientation; and “new entrepreneurs" have a productiveexploitative orientation® Other studies examine only selected values or attitudes which might be considered. dimensions of authoritarianism, For ex- ample, some authors have explained power relations in Meco by focusing on macksmo, a preoccupation with male power, expresed in displays of bravery, sexual prowess, and domi: nance. Much of the early literature which discussed Mexican pteoccupation with power, assertiveness, and strength interpreted these phenomena as a reflection of insecurity or inferiority feel- ings rooted in national historical experiencess* Such an inter pretation, although not based on extensive empitcal data, would be consistent with the concept ofthe authoritarian personality. ‘There are few hard data on the incidence of machismo within the Mexican popolation. Lola Romanuec'Ros, writing about the same village studied by Fromm and. Maccoby, found that machismo was generally less prevalent than gossip and folk myth would suggest, and that i is found les fequently among young generations By contrast, Fromm and Macoby found no rela tionship (comlation of 08) between machi and age * There is aso litle empirical evidence liking machismo with politcal behavior. The presumption has been that machismo probably afets politcal leadership tle, participation in poi tics (limiting participation by women), and family socialization to authority relations.” Octavio Paz, in his reflections on Mexi- can character, relates machismo to perceptions of men in power: ~ «ima world of chingones fliteally, sexual violaton), of ificalt relationships, ruled by violence and supicon. . [the] only thing of value is manlines, personal strength, a capacity for imposing oneself on others. . . . This is the model —more mythical than real—that determines the images the Mexican people form of men in power... . They are all machos, chin ones," In survey research in Mexico, authoritarianism has more often been treated as a set of opinions on a range of “democratic” practices and preferences for particular sles of political leader ship. Fagen and Tuohy, for example, report the responses to a seties of items measuring support for free expresion, universal suftage, and minority rights. They found that members ofall Political Culture in Mexico My socal clases in the cy of Jalapa agred (0 percent or more on al items) that “democracy isthe bes form of gorernment; pubic offcials should be chosen by majority vote; {and that) every citien should have an equal chance to infuence government policy.” However, the responses of Jaapeis to eight items about the rights of minorities to free expresion and the franchise revealed marked “anticlemocratic tendencies,” especially among the middle and lower cass. Jalapeios placed very litle value upon the rghs to fre expresion of such groups as Communists, speakers against the PRI or the Catholic Church, and critics of “‘Iife in Mexico,” Fagen and Tuohy interpret these results (in which specifi democratie practices do not receive the same sup port as general “plattudinous formulation") asa refection ofa statusquo orientation, a desire to restrict elements perceived 4s “disruptive.” particularly among the les educated and sci politically marginal members of the population for whom costs of disruption have generally been highest in the past? Evidence fom surveys on authoritarian tendencies among the Mexican lower clases is contradictory, depending on which dimensions of authoritarianism are being tapped, For example Coleman constructed a civil libertarianism index and found that in simple correlational analysis there was no relationship be {ween civil libertarianism and income or education that could ‘support the thesis of “working-class authoritarianism” ** in Mex- ico. However, in multiplicative analysis, the combination of high income and more education was positively asociated with civil libertarianism. ‘The same study demonstrates some interesting class differences in replies to specific items, evidence which could support the Fagen and Tuohy thesis that responses t0 such questions may reflect group interests, Coleman's respondents were asked whether they agreed that “democracy demands that iliterates have the Hight toot." Sevenythe pret hana spl aed or agred strongly with the statement, wheres only 528 percent ofthe upperlass sample expresed similar sentiment. By con trast, when asked for their reaction to the statement that “Even though he may be an agitator, everyone has « right to say Publicly what he thinks about political matters” 67.2 percent ofthe random sample agreed or agreed strongly, whereas 789 percent ofthe uppers sample expressed such agreement30 Political Culture in Mexico Among a predominantly low-income sample of Mexico ity residents, Cornelius found that those who had been born and raised in a rural area showed significantly more authoritarian tendencies (defined as a preference for strong, autocratic leader ship anda low level of tolerance for minority opinion) than ity-born respondents, a difference which was independent of educational level. He also found that authoritarian tendencies were strongest among the most politically active stratum of loverclass migrants to the city. This finding may reflect the gantclar patter of authoritarian leadership —caciguimma — which had prevailed in several of the settlement zones included in the study: “A cacigue may devote considerable attention to political mobilization within the community under his. con- trol. . . . Authoritariansminded residents may be more receptive than others to his inducements, and may thus be drawn more realy into politcal activity” The same pattem of local community leadership is widely encountered in rural areas of Mexico" ‘Theories of working-class authoritarianism in Mexico receive some support in Rafael Segovia’ book on politicization of Mexi- can schoolchildren, Summarizing authoritarianism among the children he surveyed, Segovia describes them as characterized by “desconfianza,’—a general lack of trust, asociated with de- cisions not to partcpate and with pattems of iret, personal dependence; a strong (not benevolent) leader image of the presi- ent; a lack of confidence in political parties, such that a ma jority belive that the government should have jurisdiction over parts; intolerance of Communiss; rejection ofthe rights of political dissidents; and the belief that laws should be taken as givens, without citizen participation in their formulation, Se- sgovia’s data indicate that most of these orientations are more prevalent among children from lowerstats families * We are inclined to agte with Segovia that a sulwtanial proportion of these differences may be due to variations in exposure and in- formation, as well as to classrelated differences in political efficacy (subjective political competence) to be diseussed below, ‘Structural Interpretations. Many recent treatments of Mexican polities have suesed the authoritarian character of politcal decisionmaking proceses and strucures for political organiza- Political Culture in Mexico 31 tion. Authoritarianism in this structural senwe may be considered in conjunction with, or independently of, authoritarian tenden: cies among the general population. Discussions of structural authoritarianism genetally rely upon Juan Linas model of authoritarian regimes, which stresses the following characteris. tio: limited (not responsible) pluralism, low popular mobiliza tion, weak ideological conaints on elite decision making, and the frequently arbitrary exercise of power by a single leader ot small group. Among Mexicanists there is now wide agreement that these features describe the Mexican political system, and that previous scholarship which focused on the possibilities for interest articuation, government acountabiliy, and increasing mas particpaton in politics within the one-partydominant system misinterpreted the way Mexican politial institutions operate, This reconceptualiation of political life in. Mexico becomes important in explaining various aspects of Mexican political culture, In structural interpretations of authoritarian ism, centralization of authority and decision making is epitomized by the key institution of Mexico's political structure: the presi: ency, Some students of Mexico prefer to describe the Mexican regime as “presidential,” stressing the relatively unlimited powers of the president as well asthe “cule of the gloifica- tion of the President.”* In this view, government decisions can be influenced to ome extent by certain poveral interes roups (national and foreign entrepreneurs landowners, organized labor, the military) and regional ot statelevel power brokers, either through direct inuence attempts or, more frequently, “preemp. tive concesons by the government, Bu the president hinsel isthe ey initiator and shaper of all major public polices: In Mev, the decionsmaking proces is formal inte bythe execute. Inthe fis stage, the president commits himself to 2 pmtularidea that he may or may nt have originated. Te actual ovigin ofthe dea not important, however. What mates i the president's commitment to it tir couse of action rarely ithe neslt fdtet presi by con cered groups. The coupation of group lars (which reduce the Aawonomy of interest group) and the low lee of mobiliaion of the rankandle membership makes i dificult for groups to pres sute the executive... The main device the authoritarian die 1635 to demobilnitscries i the incmporation of the maleon {This} commiment to 2 par32 Political Culture in Mesico tents into. the decisionmaking proces. Incorporation occurs however, or the inital vague version ofthe Iegiation has been approved by Congreso, if no lation is involved, after the vague venion of te decison as been publidy announced. Participation is... confined 0 the elaboration of technical de- tals and implcs the acceptance bythe groups of the President's political commitment Congresiona ratification (eitimation) of the presdent’s wil has been virtually automatic in most of the period since 1980. Stevens points out that until 1968, when President Gustavo Diar Ordaz ordered the violent repression of student demon- strators in Mexico City, resulting in the loss of hundreds of lives, Mexican presidents were virtually immune to public crim, Tn ac, Stevens's analysis shows that glorification ofthe president vas one ofthe three dominant themes in newspaper reportage between January 1965 and March 1966, the period she selected for content analysis Since the president is regarded as per- sonally responsible forall government policy, the media exhiit remarkable selfcensoship in withholding critical news and problems in the execution of announced polices or programs, ules publication of such material has receive prior authori tion." Press coverage tends to emphasize cultural and material Progress and benefis to come. AS a consequence, to the extent that critical opinions of government are based on facts and figures, the information must come from personal experiences, contacts within the regime, or other privileged soures. Today the president remains the embodiment of paternalistic, authoritarian rule in Mexico, The regime continues to cultivate an image of presidental acesibilty to citizen influence at tempts, and, responding to this image, a large proportion of | petitions for governmental action originating in lowincome groups is routinely directed to the president himself This Teliance on the president reflects two factors in addition to his “accesible” image: extreme centralization of authority not only in the executive but also in the federal government, and the saliency ofthe presidency asa political institution, Thus the coms mon practice of directing petitions for government benefits as close to “the top’ as posible rellects popular awareness of the fact that “the resources, opportunities, and decisional latitude Political Culture in Mexico 88 anailable at the local level are everywhere in Mexico sharply diminished though the mechanisms of executive central" The salieny ofthe presidency over and above all other Mexi ‘can political institutions is reflected in the results from various surveys. Summarizing data on politcal information levels, Cole. ‘man concludes that “the offce of the chief executive is much tore salient to the average urban Mexican than are other poli cal phenomena.” In contrast to the relatively low levels of political information Goleman found among his respondents, he found that 100 percent ofthe upper-class sample, and 89 percent ofthe random sample could identify the preceding president by tame and that (respectively) 98 percent and 0 percent of his samples knew the length of the presidential term of ofices* Cornelius found that $9 percent of his low-income respondents in Mexico City could evaluate the overall performance of the incumbent president (Diaz Ontaz) and his immediate prede- cessor (Liper Mateos), while 88 percent could evaluate the per formance of President Miguel Alemén, who held ofice from 196 to 1952. Segovia also found that the highest levels of political information among the Mexican children he studied were sociated with the office of the president. Ninety percent of his ‘respondents could identify the incumbent president by name; expresidents vere the next category of oficials most readily identified. Eighty-six percent knew that the President had been the PRI nominee five years before. Seventynine percent knew that he serves a sicyear term." Each incumbent fosters the popular sense of dependence upon his ofice, One of the purposes of the highly elaborate, nine- to tenmonth “electoral” campaigns conducted every six years by the presidential candidate of the oficial party, involving visits to hundreds of localities throughout the nation, is to enable Citizens to make direct appeals to the president-to-be for various Kinds of government assistance While only a small. portion of these petitions is ever acted upon by the government bureau cracy, the symbolic impact of the extensive pettiongathering effort should not be underestimated, When government aid is not forthcoming, the ction is likely to absolve the president of blame, attributing these failures of performance to corrupt, in- competent subordinate officials who thwart his will, This kindut Politica! Culture in Mexico of rationalization is clearly reflected in the sharp contrast be tween highly positive mass orientations toward national politi- cal institutions (especially th presidency) and strongly negative citizen evaluations of government output performance, revealed by The Civic Culture and every subsequent survey study which has explored these aspects of Mexican politcal culture. We shall retum to this point in out disussion of politcal cynicism in Mexico. The image of the president as an authority figure, exenial not only in policy making, but aio in maintaining public order and containing the divisive forces (latent or overt) within the nation, sems deply embedded in the Mexican politcal culture, and serves to reinforce and legitimize the authoritarian features of the regime. This vision of the president is already fixed in childhood: this capacity to govern, to retan all ofthe power in his hands iswhat mot awracs the young minds, ... He mot the benevolent leader; he is above all he authoritarian leader, capable not oly of governing but of eating 8 material world ~ subway, dams, highways athletic sain ~ by his own wil. Fr the Mevian cil, authority in a single person, and i coming to hi? Segovia’ research demonstrates convincingly that this presden- tial image is internalized at a relatively early point in the life cycle. Filiyone perent of his sample of Mexican children in the fifth year of primary school expressed the belief that the president’ principal {unction i to “maintain order in the coun- try,” while 80 percent felt that “the President fulfills the will of the people” and 17 percent belived that “the President makes the laws? ® This widely beld view ofthe president as the chi authority figue charged. with guaraneeing "law and order” helps to explain the high degre of popular approval ~espe: cially among the poor — acorded President Diaz Onda mas sac of hundteds of unarmed student demonstrators and by standrs in Mexico ity in 1958. By a margin of 89 perent to 15 percent (with 26 percent expressing no opinion), the low income adults in Mexico City interviewed by Comelus in 1970 approved the government’ actions agaist the students, the snajorty leling that thes actions were ncesay to restore ode, Politica C 8 on because the students had been openly “delving” the pres dent" The imprint of authoritarian pots on mas; orientations tovard political participation is ako clearly visible, Mexicans seem to have asessed the utility of participation in such limited “democratic” institutions as are available to them, and to have concluded that their interests are best served by abstention or participation through brokers. This pattern of mediated partic pation, or interest articulation through patron-lient networks, is extremely important in Mexico, The widespread view of govern mental authority (xcept that exercised by the president) as arbitrary and capricious, combined with awareness of the in ficiency, graft, and personalise interpretation of rules which pervade Mexican public administration, leads to the extensive ue of politically connected intermediaries (persons with palace, ar leverage) to facilitate dealings with government agencies, In some instances the brokerage function is performed by lower: level Partido Revolucionario Institucional officials others by local community leaders who ate themselves the “lens” of goverment bureaucrats or poitians”* While providing im- portant channels for demand making by the poor, therefore increasing the responsiveness of the sistem, the brokers ako asist the incumbent regime by helping to limit demands for excesively costly services or benefits and by fragmenting the politcal action ofthe lower cases into requests for highly pa ticularistig, shorter benefits. The potential for lagescale demand making aimed at influencing overall government pri aris or the browd outlines of public policy is thereby reduce, POLITICAL, EFFICACY (UBJECTIVE POLITICAL COMPETENCE). Almond and Verba accord great importance to the general concept of subjective competence; indeed, one might regard it 48 the single mos. important atitudinal orientation in their roe of democratic participation and the stably of democratic systems, Much of their analysis is devoted to distinguishing several types of competence and to relating subjective compe. tence to pattems of politcal sodaiaton, political avarenes, civic cooperation, and other politically relevant attitudes. Other scholars have prefered the label “political efcacy,” and have ‘generally accorded it less centrality in their analysesPolitical Culture in Mexico 386 In their indexes of competence, Almond and Verba distinguish between political or citizen competence (the individual's per- ception of his or her ability to influence the formation of laws and policies), and administrative or subject competence (the inividuals perception of hfs or her ability to afet the applic tion of general policy in specific, personally relevant situation), Questions used to measure the various forms of political com. petence tap the individual’ anticipated response to situations of stress in which rules or regulations harmful to the person's interests are being considered by local or national government. Administrative competence is measured by items which ask the individual to consider his or her relations with the bureaucracy anc police inthe case of personal problems or minor oferses ‘Throughout their analysis, Almond and Verba contend that the main distinguishing characteristic of Mexican politcal cul ture is that it is “aspirational” (and alienated), In part this characterization rests on their finding of a substantial difference between levels of political and administrative competence in Mexico. Fifty-two percent of their respondents were “Iocally” competent; that is, they felt able to affect a regulation being considered by the local government, Thirty-eight percent were “naonally competent, believing themselses able to affect regu lations under consideration by the national legislature. Thirty three percent were both locally and nationally competent. ‘By contrast, on the dimension of administrative competence, nly 14 perent of the sample expected their point of view to receive serious consideration from the bureaucracy and 12 per cen expected similar attention from the police. The gap between politcal competence and adminiarative competence vas greater in Mexico than in the other four countries surveyed, both in the total sample and among thse with only primary education ‘When sex and secondary education are controlled, however, the gap in Mexico between the two kinds of competence begins to approximate the diferences in competence levels encountered in the other countries Almond and Verba attach grat significance to the relation: ship between the two kinds of competence in the five nation, because they regard it sa refection ofthe process of the spread of competence and, by iniplication, of democratization, They attribute the relatively higher levels of political competence in| Political Culture in Mexico 87 ‘Mexico to the Revolutionary experience, which focused atten: tion on national politics and political leaders, initiating mem- bership ina politcal system which rejected traditional authority and promoted a democratic ideology, Lower levels of adminis trative competence are attributed to contacts with a bureaucracy which has remained eventally arbitrary and subject to politcal forces.* We take issue with these assessments of the significance of the pattems of subjective competence in Mexico revealed! by the Cie Culture sry, We suspect that Almond and Verba have ot tapped two kinds of competence, but rather, thatthe items measuring administrative competence (questions about the ex pected treatment and attention from the police and buraucrac) really measure more concrete cases of the same estimates of politcal stem responsiveness tapped by their Guttman scale questions on local and national political competence. Viewed in this context, it would be logical for responsiveness expacta- tions to be reduced, as the hypothetical cases relerred to in the questions become eter more concete and proximate 10 the individual's experience. Although Almond and Verba refer to the effect on administrative competence of government respon- siveness and the existence of channels of influence, these vari- ables are not mentioned as important determinants of other forms of competence: This omision of government responsive. ness a5 an important contributor to the patterns of subjective competence found in Mexico isa serious oversight, We agree with Carole Pateman’s criticism that the assumption in Almond and Verb (and most other studies of eficacy is that “efcacious” replies accurately reflect system responsinenes, and that all ‘other replies (including nonefficacious replies) must be indica- tive of individual or group psychology. Pateman contends that the subjective compeience sale used in Te Civic Culture re fects, insted, a cognitive element of efficacy, that i, that all replies to thee questions ree the respondents eximates of system responsiveness and opportunities Coleman and Davis, also convinced of the importance of dis tinguishing among various dimensions of political eficacy in Mexico, have argued that efcacy should be studied in terms of internal and exteral dimensions, depending on whether the individual is asvesing his own skills or resourees for takingPolitical Culture in Mexico 8B advantage of influence opportunities (“internal efficacy"), oF the instrumental payoff of inuence attempts (‘extemal ef- cacy”) This may be a sel theoretical distinction, but the practical consequences for the political involvement of Mexi- cans (in the absence of sibvantial system changes) are not significantly affected by the distinction. With thee asessments ofthe concept of politcal eficacy (or competence) in mind, we turn to the relevant research findings from Mexico. After reviewing the available data on levels and. determinants of efficacy, we shall examine the relationships be- tween eficacy and participation inthe Mexican poi ssi, Political Eficacy in Mesico, Altiond and Verba, Fagen and ‘Tuohy, and Landsberger have all used roughly similar questions to tap the dimension of local political efficacy, In answer to the question “Suppose a regulation were being considered by the most local governmental unit which you considered very unjust o harmful... , what do you think you could do?” 52 percent ‘of Almond and Verba's Mexican respondents believed they could do something about the unjust law, Using a similar ques- tion, but with a graduated scale of responses, Landsberger found that 80 percent of the rural ceditsociety members he studied in the Laguna region believed that it was quite likely or very likely that they would be able to stop the local authorities from taking steps injurious to themselves Fagen and Tuohy asked their respondents to consider a similar hypothetical situation, and then asked, “Do you think you could do anything about i” Only 22 percent of their respondents replied in the affirmative, Probing further, Almond and Verba asked their respondents about the likelihood of their success, should they attempt to change the regulation, Forty-one percent considered it likely or very likely that they would sued, OF ther Jalapa respondents who thought they might really do something to ty 10 change the regulation, Fagen and Tuchy asked about the anticipated probability of success; only 11 percent thought it probable or very probable that chy might succeed’ ‘There are substantial differences in the responses obtained in cach of these surveys which tequieexplanaion, In par, word ing diferences have created esentilly distinct question: Al mond and Verba ask for hypothetically weful strategies, while a) Political Culiure in Mexico Fagen and Tuohy ask more bluntly about the posibility of infuence, Landsberger’s report is quite unclear about the pre ce content of his question, We suspect, however, that the much higher semse of eficacy among rural creitsociety members might be attributed tothe specific natute of the local (jal) govern ment to which Landsberger’s question presumably rele, All ofthe ejdatarios(andsholdng residents of communities created through the government's agrarian reform program) together constitute the local government. Although one may expect that influence is unevenly distributed among the membership, the feasibility of an influence attempt is nevertheless greater in such a situation than in- an urban area such as Jalapa, where the question referred to the municipio (county) government, Other efforts to probe for manifestations of political eficacy in Mexico have more directly addressed the issue of government responsiveness to popular participation, Table 1X.S reports re: sponses to identically worded items administered in. Mexico City by Comelius and in rural Michoacin by Arterton, along with data from the Almond and Verba survey, In general, we are inclined to accord reduced importance to the absolute values ofthe Almond and Verba replies because ofthe severe response set bias problems often encountered (especially among. poorly ducated respondents) with these and other questions in the “agree disagree” format. Both rural and urban samples are in general agreement about the elect of the vote on government actions. The respondents’ asessments are probably reasonably accurate, considering that the responsiveness of government functionaries and the ese ot Aificulty of acquiring government benefits may be influenced, at least in some localities, by vor turnout and rates of voting for opposition paris, Avery diferent view, hovever, results from questions asking ahout the efet of the vote on election outcomes. Fagen and Tuohy asked their respondents whether they considered it useless to vote in municipal elections, since leaders are preselected by the official party, Eightytwo percent oftheir respondents agreed that municipal voting is useless in this respec. Commelius, asking a similar question, found that 50 percent of his sample believed that itis true or partially true that elections do not offer the voter a meaningful choice because the winners have already been selected by the PRI. The dis-30 ‘Taset INS, Seleced Poli Efeacy en (bn percentges) Political Cullure in Mexico Meio City Rum ew wer las Michatin | 1, Popular iafuence on gover “Weare spose for bud goverment” IAB “People ike me have no infence cn what erent des cs) 2 Het of popular vote cm what government docs: “Much fle” mo 8 "A little effect," m8 aug Noel.” i ug 4 Undertning pic ise “Plc it oo empl for he average anf unde.” or We ppl pay atenton or want to) they ‘an undertat polis mB 4 Abit to infloencegorernent programs: Can ony ait and cept goveraent progam” iM “Cin influence and make goverment hep” 4 1 Almoad sod Vera survey inladed ini items a an ageeiagre forma, The vests for thir naa sample (vbich in no ase dilered by more than ne percentage point om the rponse of a male, loves subsample of the Give Culture data from Mein which we anal sep arate) ae TROY, agreed that people ke themseles ave no iauene on what government des (item 1 shoe). (5, ard tat the way pope vote isthe main thing determining how the goverment san item 2 abo TM agucd tat pollicis too complated for the average man to undesand (cite 3 aoe) Sources: [Mexico City lover cas} Wayne A. Comelv, Polis ond the Migrant Poor in Meco City. Stanford: Sanford Uni. Pres, 197) [Rural Michoaci) F, Christopher Aretn, “Poll Paiciation as tempted Interpersonal Infene: Tes ofa Theoret Model Ung Dat from Rural Mean Villages” unpublished PRD. dixertoe, Masschusts Intute & Technology, Caige Mas, 19 parity between the Mexico City and Jalapa results may be due to the greater visbility and importance of opposition pty candidates inthe capital, swell sto the importance in Mexico ity of voter mobilization by local community Headers** ‘The differences between the Mexico City and rural Michoacin Political Culture in Mexico 461 samples on questions dealing with the ability to understand pol tis and to influence government programs can probably be attributed to dlferences in participation opportuites, gover ment resources, and information sources—all of which leave the rural dwellers at a relative disadvantage, By comparison, the Mexico City poor are a relatively eficacious group, For example, half of Cornelius’ respondents believed that the “really power ful public oficial and politicians” pay agent dal of attention ‘or at least some attention, to the opinions of men like themselves, This response probably reflects an expectation of attention to the opinions of a general social category (the poor, suater settlement dwellers, etc.) rather than expectations of attention to the particular respondent Vith approximately a third of the rural and urban Mexicans included in these surveys believing that they have no influence on what government does and that politics is too dificult for them to understand, as well as one-half or more believing that their vote has little or no effect on government actions and that they can only wait and accept government programs, the di: Terma ofthe average citven becomes particularly aut, con sidering his or her dependence on the government for sats: tion of baie needs, Fagen and Tuohy found that 91 peteent of the Jalapeios considered themselves helpless to. do. anything about the most serious problems facing Jaapa; 87 percent of them felt that the government had primaty responsibility for solving their problems Similarly, in Mexico City, Comneius found that 66 percent of his low-income respondents believed that the government was principally to blame for mest of their neighborhood's problems. Yet when asked who outside their community could help to solve its problems, more than 97 pe: cent ofthe respondents mentioned some government agency or official, When Arterton and Cornelius asked their respondents whether they considered! good luck, government help, God's help, cr the hard work of the residents as the most important factor for improving local condition, $8 percent ofthe rural Michoa cn residents and 41 percent of the Mexico City residents replied that government help was mos important These data showing relatively low estimations of personal eficacy combined with a heavy sense of dependence on government assistance and pro gram to improve local conditions serve asa wel badtropPolitical Culture in Mexico sie as we turn our attention to the relationship between eficacy an politcal participation in Mexico. Fficacy and Political Portcpation, Earlier in this chapter, reference was made to the rélative absence of behavioral data in The Givic Culture, and to the narrow range of participatory behaviors which are examined by Almond and Verba, Recent research on politcal participation in a variety of counties, in- cluding Mexico, bas cleatly demonstrated the need for more ifereniaed! analysis of participation and the determinants of participant behavior, Verba and his associates, for example, in 4 comparative study of political participation in Austria, India, Japan, Nigeria, and the United States, isolated four basic modes of participation, and found that (I) people who tend to “speci ize” in a given mode of participation often have diferent pet= sonal characteristics and attitudinal orientations toward politics than people who participate mosly in other modes of pari pation; (2) that participation in one mode does not necessarily ead the individual to participate in other modes; and (8) that the causal paths or proses which lead people to part aay differ substantially from one mode of participation to | i another." In their research, the standard socioeconomic moclel of partipation —in which higher income, education, or ocu- patonal status are che principal characteris that lead some people to participate more than othert~ proved les adequate to exphin some moves of participation (voting and particulars tic contacting of public ofa) than others (amipaign activity — and community involvement}, | The Verba and Nie research, and stuies which have partially replicated it in Mexico, Costa Rica, Peru, and Japan, suggest — that at ast some forms of politcal participation among the tual and urban poot may be better explained by variables hich are not as relevant for explanations of participation in general populations or among the middle and upper classes, Specifically, these studies find that the resources or characteris- tics of the individual (including personal sense of political eficay) ae often far les important a determinants of partic: pation than certain apects of the national or local political structure, or more genealy, the group context in which the j poor citizen finds himselt* Political Culture in Mexico 68 Similarly, Mathiason and Powell's research in Venevuela and Colombia has demonstrated that among peasants, personal ef icy —the perceived ability to influence government decisions individually and direcly—is not a prerequisite for political activity. They found that much particpaton could be explained by “mediated political eficacy,” wherein the peasant fels able to influence the government because of his relationship with a network of mediators or brokers who ate, in turn, influential in government cicles: ‘The syndicate is something concretely visible and undersandabe to peasns. No ability to imagine onesl nvidally tying to influence noma government is require fora sese of fia, A peasant ned only lean thatthe scat exis as broker be- tween himself and the goerament and his Keson is larned by obuerving and especially by participating in the brokerage ac tvs ofthe union... Consequently, if the union i adept at brokerage... there isa greater likelihood that its member will be taught that they ar, in fc, eficacious* ‘Thus the sense of fica) sted more tothe mediating structure, and the successful outcomes of its influence attempts, than to the individual. The authors also demonstrate that mediated po- litical effcacy is not a product of childhood socialization but, rather, of direct political leaning experiences during adulthood A varity of other factors have been found to be important in explaining politial participation among the rural and urban poor: sire experiences of a cllecive nature (eg, land in ‘vasions), concrete experiences in dealing with political and gov ernmental agencies, intensive mobilization by lcal-evel leaders or government representatives, patronlient linkages, specific concrete needs or problems particular to a community, and awareness ofthe represve capacity of the government and of the overall politcal opportunity structure atthe national level — thats, perceptions of the kind of polit activity which are likely to be rewarded or punished by elites"* ‘These kinds of perceptions are also quite important in explain- ing the generale preference among the poor for informal ma nipolaton of the political stem, as opposed to diet confront tio of protest tacts. Informal manipulation through mediating structures such as centage networks may simply be a more efficient and less risky strategy for extracting benefits from the