The Ideology of the Mexican Revolution,1910-40

ALAN KNIGHT St. Antony's College, Oxford

Some Clarifications

Ideology is a slippery substance. It can mean many different things; one assiduous scholar has counted 27 different meanings.[1] (A similar plasticity is evident within the particular field I am covering: the 'socialism' which informed Mexican education in the 1930s supposedly embodied up to 35 different meanings).[2] Frequently, 'ideology' carries negative connotations: one man's 'scientific' meat is another man's `ideological' poison; or, in the terms of Clifford Geertz's 'familiar parodic paradigm': "I have a social philosophy; you have opinions; he has an ideology".[3] In part, it would seem, this slipperiness derives from the fact -a fact widely, if not unanimously, recognized- that the 'great' ideologies tend to combine 'objective' and normative elements. Marxists believe that class struggle is (objectively) the motor of human history, but they also believe that Marxists have a moral commitment to one side in that struggle.[4] Economic liberals are likely to favour neo-classical economic principles, or rational choice models of political analysis, while at the same time recommending the market

 And self-seeking individualism as the best means to maximize utility -as they would put it.[5]

The ideologues of the Mexican Revolution tended to stress the normative over the analytical. They advocated particular principles chiefly on the grounds that those principles were best for the country; not because they embodied timeless truths or offered universal analyses of the human condition. Such principles might carry some `scientific', analytical, weight: Madero, who placed great faith in free elections (sufragio efectivo), interpreted Mexican history in terms of a struggle of civilian reform against praetorian authoritarianism (La Sucesión Presidencial); yet more clearly, the radical ideologues of the 1930s -Lombardo Toledano, for example- combined a commitment to class struggle (para una sociedad sin clases) with a belief in `scientific' socialism. But, as I shall argue, the ideology of the Revolution was usually unashamedly normative, did not claim a universal validity (as Marxism or neoclassical economic theory do), and, indeed, often drew inspiration from an ideological tradition -roughly that of popular patriotic liberalism- which was quintessentially national, mythical, and affective (as opposed to universal, rational and cerebral).

A second source of ideological slipperiness is both more obvious and more problematic. It concerns the gap between theory and practice: a gap evident throughout the world, but particularly chasmal in Mexico. Not for the first time, I am drawn to quote Ernest Gruening: "in labor, as in all else Mexican, things are often not what they seem or what they are declared to be". [6]It is a commonplace that, throughout the history of Mexico, political practice has not corresponded faithfully to political rhetoric. Laws and constitutions have served as statements of intent rather than as blueprints of government. The old colonial precept -obedezco pero no cumplo- has its modern counterparts: in regard to electoral practices, agrarian reform, anticlericalism, labour legislation, economic nationalism. Elections are fixed; `fictitious' ejidos are created;[7] latifundia are disguised as pequeña propiedad, Catholic education and ritual are discreetly tolerated; the provisions of the Labour Code are discreetly ignored. In the words of James Scott, the `public' transcript -laws, regulations and rhetoric- diverge from what really happens, and what people know really happens.[8] The historian of `ideology' therefore has to decide whether to address both transcripts - public and hidden, rhetorical and real- or whether to give precedence to one or other. Intellectual historians, almost by definition, may favour public, stated, ideology. I admit to a preference in favour of the hidden and the real. That is to say, I think it is important to see what ideology meant in practice; not to be taken in by outward forms; not to fall victim to that sterile formalism which chases up ideological origins, blithely disregarding historical practice.[9]

Thus, when I come to categorize currents within the Revolution, I note rhetorical stances, but I stress real practices. Political liberalism -the implementation of free and fair elections- was a common rhetorical stance, which underpinned the formal Constitution of 1917. But it did not underpin political practice in revolutionary Mexico, since elections were usually controlled, often corrupt, and sometimes violent.[10] This does not mean that they were irrelevant; but it does mean that they failed to conform to liberal- democratic norms (in which respect, of course, Mexico was far from unique). For that reason, I do not consider political liberalism -and its touchstone, free and fair elections- to be a hallmark of every revolutionary faction or current; on the contrary, I would tend to confine it to Maderismo (and, with qualification, to Zapatismo) since, in both theory and practice, in prwer and out, these movements displayed a consistent attachment to this principle.[11] Similar examples of ideological facades can be cited: by c. 1930 Callista agrarismo was a hollow shell, even though article 27 of the Constitution had not been formally repudiated.[12]

A third preliminary point concerns the purpose and character of 'intellectual' history -meaning by that, history which deals with ideas, including therefore ideology. As Robert Darnton has noted, intellectual history is something of a dethroned queen; she has lost her old historiographical empire and gained, instead, a host of distinct and dissident provinces. Thus, Darnton argues, intellectual historians cover an enormous range: "at one extreme they analyze the systems of philosophers; at the other they examine the rituals of illiterates".[13] Within this Balkanized range, at least four provinces stand out: "the history of ideas (the study of systematic thought, usually in philosophical treatise), intellectual history proper (the study of informal thought, climates of opinion, and literary movements), the social history of ideas (the study of ideologies and idea diffusion), and cultural history (the study of culture in the anthropological sense, including world views and collective mentalités)".[14] In the case of revolutionary Mexico, `systematic thought' is at a discount: if I may be permitted another familiar quotation, this from Frank Tannenbaum: "the Mexican Revolution was anonymous. It was essentially the work of the common people. No organized party presided at its birth. No great intellectuals prescribed its program, formulated its doctrine, outlined its objectives... How different this from the French and Russian revolutions! There was not a (sic) Rousseau, a Montesquieu, a Diderot in Mexico".[15] Nor, we may add, was there a Lenin, a Trotsky, a Bukharin: all systematic thinkers, who left an intellectual legacy which commands attention irrespective of their political success or failure. (Compare Stalin, whose contribution to the history of ideas is as small as his contribution to political history is great.) I can think of no Mexican revolutionary thinker who compares to this Soviet triad; perhaps the strongest candidate -Lombardo Toledano- deserves mention within the Latin American intellectual tradition; but he even is not of the stature of, say, Mariátegui.[16]

 Indeed, the great protagonists of the armed Revolution (Madero, Zapata, Villa, Carranza) were unoriginal thinkers, at best; and the great architects of the revolutionary regime (Obregón, Calles, Cárdenas), produced a body of `thought' which is fairly vague, imprecise, rhetorical, and specific to Mexico. (Cárdenas, indeed, disfked intellectuals; Obregón's mordant wit was at times directed against ideas-mongers, like Vasconcelos).[17] Hence, as already mentioned, one obvious limitation of Mexican revolutionary ideology in general is its parochial pragmatism, its stubborn refusal to generalize beyond Mexico's borders: a display of modesty strikingly at odds with the `armed opinions' of the French jacobins, or the global pretensions of the Comintern.[18] However, this is a strictly intellectual limitation -which explains why Mexican revolutionary ideology and ideologues scarcely figure in global analyses of political theory; in historical terms, it is, in contrast, a source of strength and stamina, as I shall go on to argue.

The ideology of the Revolution, therefore, is chiefly notable for its practical impact, rather than its intellectual novelty. The revolution's ideologues are worthy figures of political -even social- but not intellectual biography; and the constituent ideas of the revolution are interesting inasmuch as they moved men -and women- to effective action, not because of their intellectual novelty or brilliance. Indeed, the most important ideas -sufragio efectivo, tierra y libertad, una sociedad sin clases- may be important precisely because they exerted a simple, down-to-earth, emotive, appeal -an appeal that was visceral rather than cerebral. As the school mistress Socorro Rivera, who began teaching as a seventeen year old in rural Puebla, recalled her experience fifty years later: "neither we nor the villagers had the slightest idea what socialist education was all about, but it was a powerful feeling, a spontaneous emotion"; at which point, the old lady spontaneously burst into song: "let us march, Agraristas, to the fields, to sow the seeds of progress...".[19] The `truth', consistency, and novelty of Mexican `socialism' were not crucial; it was its mobilizing power which counted. The same is true of Mexican `liberalism' - and other `isms', mentioned below.
However -and here I come to my final caveat-, measuring mobilizing power is not easy. A familiar quotation (not of Mexican provenance) attributes thaumaturgic power to ideas, speciflcally economic ideas: "practical men, who believe themselves to be quite exempt from any intellectual influences, are usually the slaves of some defunct economist. Madmen in authority, who hear voices in the air, are distilling their frenzy from some academic scribbler of a few year back"; in conclusion, "the power of vested interests is vastly exaggerated compared to the gradual encroachment of ideas".
[20] But Keynes was, I believe, hugely idealistic and hugely mistaken.[21] We do not know if, when political leaders cite particular `ideas', they are in fact admitting their real motivation or, rather, seeking ex post facto legitimation, justification, or mere intellectual respectability. Wars are habitually fought, economic policies pursued, in the name of grand principles rather than sordid self-interest. We must therefore beware of accepting Keynes's principle in general, or the self- serving rationalization of politicians in particular. In many cases, of course, intellectual motivation is not admitted; it is inferred by ingenious historians who, for example, discern a particular influence (such as Magonismo) running through the declarations, decrees and policies of later political leaders.[22] This is a familiar historical fallacy: it mistakes a certain loose congruence for strict causality; and congruence may arise from a shared cultural environment, rather than any causal relationship. Certain articles of the 1917 Constitution resemble some Magonista proposals, but both may be products of the contemporary milieu, branches of the same tree; it does not follow that the chronologically later causally derives from the earlier.[23]

Let me sum up there initial caveats: the ideology of the Mexican Revolution tends to be normative rather than analytical, national rather than global in its application; its importance resides less in any intellectual novelty or brilliance (indeed, its `great thinkers' are far from intellectual giants) than in its capacity to mobilize Mexicans; hence, the study of this ideology leads us logically to the `social history of ideas' - the theed of Darnton's four categories. Here, however, we encounter some specific problems: how to relate theory to practice, rhetoric to reality. Did ideas autonomously motivate Mexicans to action, or did they merely legitimate action taken for other (socio-economic, political, contingent) reasons? Here, the answer may vary: some policies (e.g., political liberalism and anticlericalism) strike me as more strongly `ideological', in that they responded to the power of ideas and did not, essentially, cloak ulterior motives or group interests (in some cases, indeed, they were even politically counter-productive). Other policies (such as labour and agrarian reform) responded to collective self-interest, and offered both material benefits to their mass constituencies and political pay-offs to their leaders. It is to there tricky questions of interpretation that I now turn.

A Working Model

Classic intellectual history -the great thoughts of great thinkers- can comfortably adopt a biographical approach, but we cannot. Even in the case of Mexico's more cerebral políticos -Lombardo Toledano, for example- the intellectual approach can be very partial and misleading.[24] But if we are to link ideas to broader processes of history, how should the analysis be organized: in terms of the ideas themselves (their origin, appeal, logic) or of the carriers of ideas (individuals, groups, institutions)? I think we have to do both: indeed, the pay-off of `intellectual history', in this broad, cultural sense, derives from our explaining the relationship between ideas and their carriers. Fortunately, this relationship -while it is certainly not simple and transparent- does display a measure of coherence and rationality; there is, if you like, an `elective affinity' between clusters of ideas, distinct social groups, and specific political functions. Some affinities are obvious: the peasantry espouses agrarian reform (e.g., article 27 of the 1917 Constitution), the urban working class looks to labour legislation (article 123). Even here, however, we must be cautious: some peasants supported agrarismo more eagerly than others -whether for socio-economic, cultural, or religious reasons; some peasants formed unions and sought the protection of article 123; and some urban workers (not to mention rural proletarians) petitioned for ejidal grants (a recurrent complaint).[25] Political and ideological attitudes cannot be straightforwardly inferred from socio-economic status.
lf such indeterminacy is evident in the socio-econonomic sphere -the sphere of property and labour market regulation- it is, not surprisingly, even more evident in the political and ideological realms. We can loosely correlate political liberalism with the growing urban middle class of the Porfiriato, which found its spokesman in Francisco Madero; but political liberalism was not confined to the middle class constituency -for, as Rodney Anderson rightly argues, a distinctive working class liberalism also flourished and fed into Maderismo.
[26] Furthermore, with the spread of armed revolution and social upheaval, middle class liberalism often gave way to middle class authoritarianism -in revolutionary Mexico as, decades later, in authoritarian Chile or Argentina.[27] Blective affmities', in other words, are loose and malleable. As Ernesto Laclau pointed out some years ago, citing pertinent examples, class identity and political ideology do not neatly correlate: "ideological `elements' taken in isolation have no necessary class connotation".[28]

What goes for classes and social groups also goes for political factions/ coalitions -and I do not accept the simple view that factions/coalitions merely mirror classes and social groups.[29] By `factions' I mean groups sharing some loose allegiance to a leadership (individual or collective), sometimes to a definite political project, and almost invariably to common political interests- a commonality which may be summed up in Benjamin Franklin's famous advice to the American revolutionaries: "we must all hang together or assuredly we shall all hang separately". Factions are therefore built up on the basis of clientelist, personalist, spatial and ideological loyalties. They are invariably multi-class associations; but some (such as Zapatismo) display greater class homogeneity than others (e.g., Villismo, Carrancismo).[30] Even more than classes, however, factions can display shifting ideological attachments. Zapatismo began as a relatively moderate movement, which recognized the place of the plantation in Morelos rural society; but it was radicalized by revolution and carne to adopt a more intransigent agrarista stance.[31] Villismo, increasingly friendly to US interests in 1913-14, veered towards anti-Americanism and a somewhat contrived anti-imperialism after 1915.[32] Callismo, relatively radical in its initial stance towards the church and the oil companies, became more moderate -even conservative- with time; the Callismo of the Maximato was, in general, a more conservative Callismo than that of the mid-1920s. Cardenismo wove a different course: initially uncertain (1934-5), it became increasingly radical (1936-7), but shifted to the centre after 1938, seeking detente with the Church, the prívate sector, and the United States.[33] In each of these cases, contingent factors affected the ideological colouring of revolutionary factions.

Given this contingency, it is difficult to discern any general `laws of motion' relating incumbency to ideology. Some factions grew more radical with time, some more moderate. In general, however, a couple of tendencies -not laws- suggest themselves. First, the bigger a faction, the more spatially diverse, the more, we might say, a faction became a coalition, so the more ideological variation it tended to display. The grand national coalitions -Villismo, Carrancismo, Callismo, Cardenismo- were more internally variable than their more limited, even local, counterparts: Magonismo, Zapatismo, Oaxaqueño soberanismo, Tejedista agrarismo in Veracruz. These could maintain a greater degree of coherence and consistency (recall the Zapatistas' fierce attachment to the Plan of Ayala); the national coalitions, in contrast, were more complex and shifting ideological hybrids.

The second tendency represents a revolutionary embodiment of an old and important principle: factions in power are different from (the same) factions out of power. It is not just that, as Lord Acton observed, power corrupts; it is also that power imparts a different -more centralized, authoritarian, perhaps 'responsible'- perspective. The classic nineteenth-century case was Juárez: a champion of liberal federalism as he perambulated about Mexico in his dusty black carriage, an architect of a more authoritarian centralism once he took power after 1867.[34] Madero drifted in a somewhat similar direction: in 1908- 11 he stood for fair elections, a free press, and a civilian government; in 1911- 13 he meddled in elections, curbed the press, and boosted the power of the military -ultimately with fatal results.[35] (We should note, in passing, an inverse process, which confirms the rule: Félix Díaz, nephew of Porfirio, collaborated happily in his uncle's dictatorship and was a key ally of Huerta and the army in early 1913; cheated of power by first Huerta, then by the rise of the Constitutionalists, Félix gave his narre to Felicismo -a catch-all coalition of conservatives who, in their opposition to the Carranza regime, suddenly discovered the virtues of liberal democracy and the 1857 Constitution -which, of course, Uncle Porfirio had flouted for a generation.[36]

This tendency for power to encourage a creeping authoritarianism -or for loss of power to bring democratic conversions- is hardly surprising; it derives from an obvious inner logic, and displays many global parallels: the Bolsheviks; the Indian National Congress; even the contemporary British Conservative party. But the acquisition of power could bring other consequences, other ideological mutations, which did not relate to this obvious liberal-authoritarian, federalist-centralist dichotomy. Let us shift the focus to a different dichotomy: that of (roughly) `social radicalism'.[37] Power often `moderated' factions -i.e., diluted popular inputs and made factions more sensitive to vested interests, both domestic and foreign: Maderismo, again, was a classic case; so, too, in varying degrees, were Carrancismo and Callismo.[38]

Cardenismo was more complex: initially seen as a `moderate' Callista clone, Cárdenas soon veered left, attracting popular support which was translated into radical reform; after 1938, however, as I have already mentioned, he tracked back to the centre. A cautious conclusion might be that power strongly encouraged centralist/authoritarian tendencies and somewhat less strongly favoured `moderation' (or, we might better put it, somewhat deterred radicalism). That power could, in certain circumstances, radicalise political movements is illustrated by mid-1930s Cardenismo. Furthermore, global examples of regimes which grow more radical with time are not hard to find: Stalinism represented a more radical economic project than NEP; Italian fascism grew more radical in the 1930s; the Chinese Communist Party veered left during the later 1950s and 1960s, as did Fidel Castro after 1959. (Counter-examples -of radical or progressive governments losing steam- are no less frequent: Batista; Perón; Bolivia's MNR; Popular Fronts, from France to Chile.)

These examples -the Mexican Revolution included- suggest a possible conclusion: authoritarian regimes committed to socio-economic transforma- tion (involving a serious challenge to vested interests) may grow more radical with time, as tenure of office enhances power, especially if vested interests are weakened by war or recession, and/or they provoke the hostility of the regime by their outright opposition. Thus, certain fascist and communist regimes have managed to accumulate power, curbing political opposition and vested interests: in Russia, China, Cuba, to a lesser extent Italy and Germany. The severity of the regime's attack on property rights would seem to be a crucial variable: radical socioeconomic reform is then accompanied by resistance, confrontation and -if the regime prevails (as it did in Cuba, but not Guatemala)- enhanced authoritarianism. The Cardenista project involved a move in this direction; but as political polarization increased, the government chose, or was forced, to retreat, lower its sights, and accommodate vested interests -landlords, businessmen, foreign interests, the Church. Its radicalism faded, its ideological stance moderated. In Hamilton's words, the Cardenista government could not go beyond the "limits of state autonomy" -in a way that, say, the Stalinist government could.[39]

Hence my tentative conclusion: while power usually encouraged centralism and authoritarianism, its relationship to social radicalism -in particular, challenges to property rights- was quite variable, subject to context and contingent factors (e.g., wars and depressions, domestic crises and confrontations). This relationship is therefore more amenable to jerky `catastrophe' theory than to smooth linear models of explanation. Power and centralism-authoritarianism often correlated, hence, power fostered ideologies of centralism and authoritarianism: e.g., in the Mexican context, Callismo and Cardenismo. But power and social radicalism displayed variant relations; Callismo and Cardenismo therefore differed: probably the biggest single causal factor was the Great Depression. A simple -loosely `catastrophic'- model, derived from Hamilton's thesis on the limits of state autonomy, would posit a regime -or `coalition-in-power'- linked to vested interests (property- owners, in particular), by a kind of elastic band: socially radical policies stretch the hand until a point is reached either where it snaps, allowing the government to fly off into the realm of extreme autonomy, where it can freely exert its sovereign will at the expense of vested interests (e.g., Stalinism, Fidelismo); or where the accumulated elastic force obliges the state to pull back, compromise its radical goals, and reach a new rapprochement with vested interests. This was the outcome as Mexico's revolutionary generation retired from the scene during the 1940s.[40]

The fact of entering on power can therefore have important consequences for policy, ergo for ideology. We may tend to assume that ideology -as proclaimed by out-of-power factions- determines policy, once those factions achieve power. But in the real, as opposed to the rhetorical, world, power can significantly affect ideology: often tending to make it more centralizing and authoritarian, sometimes affecting its social content, in the ambivalent manner described. This does not mean that ideology is irrelevant or that `public' and `hidden' transcripts can maintain an indefinite and extreme separation. Even Scott -a sceptic when it comes to notions of hegemony- recognises that too large a gap can create problems.[41] As Przeworski observed concerning Eastern Europe: people need a measure of cognitive coherence; a prolonged and pronounced gulf between words and deeds eventually proves intolerable.[42] For a successful regime -and the Mexican revolutionary regime was by many criteria successful- some congruence between ideology and practice had to be maintained. How can this congruence be conceptualized, given -as we have said- the loose, imprecise, many-stranded character of Mexican revolutionary ideology? To put it in the somewhat crude anthropomorphic terms of an old debate: if we are to ponder whether -at some fatal moment in history- the Mexican Revolution was `betrayed' or was justly pronounced `dead', what was the ideology whose betrayal or death was at issue?[43] For if we cannot identify the coupse, how can we pronounce it dead or decide who plunged the fatal dagger between its ribs?

My preferred conceptual model would be a genetic one. Let us assume that specific elements represent ideological `genes' which, replicated and trans- mitted through successive political factions/regimes, shape their character. These `genes' include: political liberalism, nationalism, indigenismo, agrarian and labour reform, economic nationalism, anticlericalism and `development- alism' (see accompanying diagram). Given eight such `genes' (some analysts might wish to add to the list), the possible combinations of one or more are huge (250). In fact, however, many potential combinations are historically irrelevant, since genes tend to combine in predictable -not random- clusters. Just as, in human genetics, blue eyes and fair hair tend to go together, so -for example- anticlericalism and economic nationalism tend to correlate. Thus, the number of actual combinations is quite limited.[44]

The genetic model also helps solve -or, better, remove- the riddle of the anthropomorphic revolution -the revolution which is conceived, born, matures arad dies (perhaps `betrayed'). As all historiaras now recognize, the revolution was multifaceted -in terms of its class, ethnic, regional and ideological make-up. Of course, there were some goals or attributes which `the Revolution' never embraced: Ultramontane Catholicism; a thoroughly laissez faire `nightwatchman' state; ara anti-national cosmopolitanism. But its embrace was broad and often included contradictory elements. Hence the recurrent debates concerning what was `truly' revolutionary, who were the real' revolutionaries, and when arad how the Revolution was `betrayed'. The genetic model has the advantage of offering several different but no less `true' revolutions: it therefore cautions against imputations of betrayal. Calles, judged by his own criteria (his own ideological `genetic make-up'), was as much a revolutionary as Zapata; he embodied a set of principles and policies which were integral to the Revolution, even though they were different from - and perhaps less socially radical than- Zapata's.

Finally, the genetic model makes talk of `the death of the Revolution' somewhat suspect. Ideological genes may survive some time in recession; a freak mutation or a change of environment may suddenly restore them to prominence. Agrarianism, placed ora the revolutionary agenda (we could say, 'incorporated into the Revolution's DNA') thanks to the efforts of Zapata and other agrarian rebels in 1910-20, enjoyed a brief efflorescence -esoccially in certain states, like Morelos- during the 1920s. It then lost its prominente during the Maximato, but was revived by the freak mutation of the Cárdenas presidency, coupled with the new environmental conditions of the 1930s depression.[45] Briefly, agrarianism -now a dominant gene- determined the basic character of the national reg¡me. After 1940 however, the environment turned hostile, and agrarianism -though it never entirely died out[46] - again receded. Similar cycles can be plotted for other elements in the Revolutionary repertoire, for example, anti-clericalism: absent under Madero; intermittently influential after 1913; dominant during the later 1920s and early 1930s; declining under Cárdenas; largely recessive after 1940. Hence a concluding caveat: we should not write off the Revolution -as dead, buried, with a stake through the heart- unless we are sure that its genetic material has been thoroughly destroyed or superseded. To some, the Revolution seemed spent in 1930; yet in the following decade it experienced its most radical fulfillment. Technocracy and neo-liberalism seemed to have supplanted the Revolution in the 1980s: but the 1988 presidential election produced the powerful phenomenon of neo-Cardenismo (a literal as well as metaphorical genetic revival!); and even President Salinas's Solidarity programme embodied bits of the old populist DNA, which was also evident in last year's (1996) Congress of the PRI. And, as I note in conclusion, the Chiapas rebellion draws inspiration from a revolutionary ¡con: Emiliano Zapata.

The genes of the Revolution

Using the genetic model, therefore, I shall try to set out -very briefly and superficially- the principal elements, the ideological DNA of the Revolution. I shall try to explain why these cohere in certain clusters (and not others); and I shall analyse the changing environment which, over the period 1910-40,[47] favoured some genes over others. To help elucidate the argument, I have located the eight genetic elements in the accompanying diagram, seeking to show how they cohere within particular historical `species': Maderismo, Zapatismo, Callismo and Cardenismo.[48] Both `elements' and `species' are plotted according to two political axes: state centralization as against decentralization (or parochialism); and social radicalism as against social conservatism - the key test of radicalism/conservatism being attitudes to property rights.[49]

(a) Political liberalism

If, as Tulio Halperin argues, Argentina was a "nation born liberal", Mexico was a nation which espoused liberalism as a result of prolonged and bitter conflict during the nineteenth century.[50] From the 1830s on, liberals battled to establish a liberal -¡.e., representative, constitutional- polity, a liberal economy, characterized by the free movement of the factors of production, and a liberal society, blessed with equality before the law, though not, of course, equality before the market. Victories over the conservatives in the 1850s, and over the imperialista and their Franco-Austrian allies in the 1860s, ensured a definitive liberal triumph. Several consequences followed: unfke Colombia, Mexico avoided a liberal-conservative dyarchy (with its implicit risk of civil war); conservativm, discredited, kept a low profile and pragmatic conservatives discreetly colonized the (liberal) Porfirian regime; the liberals, as saviours of the country, could both enjoy a monopoly of office and claim the mantle of patriotism. Hence liberalism, apart from providing an ideological blueprint for politics and economice, also acquired intense mythic and emotive power -which served as glue to hold quite disparate `liberals' and 'liberalism' together.[51] Political liberalism therefore long antedated the Revolution (it was an old bit of inherited DNA); the Revolution began, as Córdova notes, with a clarion call to the past; and its power -like the power of any historical ideology- depended on its emotive appeal and masa constituency, not just its internal intellectual coherence or originality.[52]

 After 1876, economic liberalism flourished, but political liberalism languished.[53] For, despite the retention of the 1857 Constitution, Díaz accelerated the trend towards positivistic authoritarianism which Juárez had begun. Mexico's liberals -like those of Bismarckian Germany- faced an awkward dilemma: in general, they approved of Díaz's economic project (its promotion of exporta, infrastructure, cash crop production, even manufac- turing industry), not least because they were often the direct beneficiaries; but they increasingly deplored his abandonment of political liberalism -the rigged elections, docile press, infractions of civil rights, and recurrent re-election of office-holders, above all of Díaz himself. In doing so, they drew on two distinct liberal traditions: first, indigenous Mexican tradition, with its emotive appeals to Juárez, the Revolution of Ayutla and the patriotic crusade against the French; and, second, foreign examples which caught the attention of an educated, literate, newspaper-reading public -French republicanism, Argen- tine radicalism, US Progressivism.[54]

These influences are clearly apparent in the key text of revolutionary political liberalism: Madero's Presidential Succession, a book which, citing historical presedent and foreign example, stressed the need for civilian government, civic virtue, and a functioning representative government. Similar concerns dominated Madero's Plan of San Luis and both the prívate correspondence and public statements of the Maderistas. It was not that the Maderistas ignored `social' questions, as sometimes alleged. Madero condemned the social abuses of the Porfiriato (Cananea, Río Blanco, the Yaqui War, the Valle Nacional).[55] But, as both opposition campaigner and elected president, he believed that, if a properly liberal, representative, polity could be established, social questions could then be addressed consensually and effectively. As he told the workers of Orizaba: "it is not up to the government to raise your wages or cut your working day and we, who embody your aspirations, promise no such thing, because that is not what you want; you want liberty, you want to have your rights respected... you do not want bread, but simply liberty, because liberty will enable you to win your bread".[56] And the workers responded positively: political liberalism exercised a strong appeal and could even cohabit with a moderate anarchism, since both shared values of libertarianism, anticlericalism, and self-impovement through education. Liberalism could also appeal to (some) peasants, for whom the promise of free elections stirred hopes of local democracy and autonomy; while invocations of Juárez struck the old chords of patriotic liberalism. Rational interest and affective allegiance thus combined.

Though narrow in its formal proposals, Maderista political liberalism was no feeble, minimalist programme. It offered solutions (effective suffrage, no re-election) which were novel, popular, even radical: solutions which had never been seriously implemented in Mexico before'[57] and whose implemen- tation implied considerable political change, even upheaval. Political liberal- ism was also a highly ideological creed, in that -notably in case of Madero and, I believe, many of his middle-class supporters- it did not serve ulterior interests; it was no mask for material advancement; it was sincerely, even idealistically, espoused. (Peasant and working class liberalism was probably more instrumental; and, as already mentioned, by 1913 middle-class material concerns -for order and property- were fast undermining disinterested liberal principles.)

Thus, the practical accomplishments were relatively modest and shortlived. The Madero presidency probably represented the high point of Mexican political liberalism judged in terms of free elections and political pluralism- at least prior to the 1980s and `90s. After 1913, however, political liberalism wilted, scorched by Huertista militarism and scorned by Carrancista Realpolitik.[58] It was never formally repudiated: it provided the political framework for the 1917 Constitution, and it continued to excite what might be called neo-Maderista enthusiasm; for example, on the part of Vasconcelos, his supporters in the 1929 campaign, and the aptly named Partido Antirreleccionista.[59] But these, of course, were movements of opposition, of political `outs', natural adherents of a liberal democratic programme. Political reality was otherwise: hardheaded, ruthless, caciquista, corrupt, demagogic, and -by the 1930s- corporatist. Political liberalism thus figured as the great lacuna of the Revolution: it had inspired the first wave of protest; it lived on in revolutionary rhetoric, for none dared openly repudiate it; but it was never translated into effective practice, not least because, as I suggested above, incumbency strongly favoured authoritarian and centralizing tendencies over democratic, decentralizing ones. Since, in my analysis,I stress actual practice over empty principie, I do not consider political liberalism as a functioning prerequisite of any revolutionary faction/movement/regime, save- with qualifications- Madero's and, perhaps, Zapata's. In short, political libéralism is chiefly characteristic of Maderismo, but fails to distinguish other post-1917 revolutionary `species'; not surprisingly, ardent advocates of political liberalism - Enrique Krauze, for example- admire Madero as much as they excoriate Calles or Cárdenas.[60]

(b) Patriotism

If political liberalism is notable for its limited representation, patriotism is ubiquitous and, of course, very old -at least as old as the colony.[61] For this reason, it is a poor criterion of differentiation, not only between revolutionary factions, but also between revolutionaries on the one hand and, let us say, `reactionaries' on the other. The Huertistas of 1913-14, the Cristeros of the 1920s, and the Sinarquistas of the 1930s all had plausible claims to be patriotic. Equally, it is difficult to differentiate between Zapatistas, Villistas and Carrancistas on this basis. All factions claimed to represent and defend the patria; few can be seen as genuine vendepatrias (whatever their opponents may Nave said). An important clarification must be made, however. I am talking here about patriotism, by which I mean respect and support for Mexico's sovereignty, autonomy, and (loosely) national culture. I differentiate this from both economic nationalism (considered separately below) and 'state-building' (which some might also wish to denote as `nationalism'). State-building certainly figures in my model, in that it helps set one of the basic axes: policies -or ideological positions- which lie to the right are, by definition, conducive to centralized state-building; for example, economic nationalism, anticlericalism, and, in practice if not in theory, both indigenismo and labour reform (all considered separately below). State-building thus involves a range of policies and positions and is inseparable from the Revolution, especially in its Callista/Cardenista forms.[62] But patriotism is not necessarily tied to a state-building project; on the contrary, it can happily co- exist with an attachment to the patria chica which may in turn block centralized state-building: witness the example of Zapatismo -or, in earlier times, the popular peasant liberalism of the nineteenth century. And state- building, though it embraces patriotism, involves a great deal more, including `modern' policies of mass mobilization (e.g., labour and agrarian reform) which have little to do with patriotism per se.

When, therefore, scholars assert the `nationalist' character of the Mexican Revolution, as they often do, they should feel some obligation to disaggregate and clarify this murky generalization. For if the `nationalism' in question is what I am terming patriotism, it is a pervasive sentiment, shared by virtually all revolutionary groups and not a few `reactionary' ones too. It is therefore a poor litmus test of `revolutionary' status. If, however, `nationalism' connotes a project of centralized stare-building -which it sometimes does, explicitly or implicitly-, then it is a more specific and useful criterion; but it is not one shared by all revolutionaries. Zapata and, probably, Madero were major exceptions; neither were -by design, inclination, or achievement- dedicated stare-builders. Arguably, they were more `stare-destroyers'. Furthermore, `stare-building' is best analysed not as a single ideological postulate, but rather in terms of particular stare-building policies (economic nationalism, anticlericalism, labour and agrarian reform). It is to these I now turn.

(c) Economic nationalism

By 'economic nationalism' I mean the doctrine which seeks to nationalize Mexico's economic resources and activities, by means of taxation, regulation and, on occasions, expropriation. It is quite different from xenophobia and, although it may be schematically regarded as the economic equivalent of patriotism (patriots promote their country's political sovereignty and autonomy, economic nationalists its economic sovereignty and autonomy), it by no means follows that all political nationalists were or are economic nationalists. For one thing, patriotism long antedated economic nationalism: creole patriots under the late colony or Juarista patriots battling the French did not necessarily entertain economic nationalist hopes; they sought to create or defend an autonomous (imagined) national political community. So, too, with the revolution: Madero -and Maderismo in general- was pretty tolerant of foreign investment in Mexico and advanced no grand project to curtail or control it; at most -and for pressing fiscal reasons- Madero made a modest start to oil production taxes.[63] Zapata, too, entertained no economic nationalist project (although, it is true, he shared a popular disfke of Spanish landlords, mayordomos and merchants). Peasant leaders in general eschewed economic nationalism: they usually confronted Mexican (occasionally Spanish) oppressors, rather than foreign (i.e., Anglo-American) capitalists; and, even where foreign companies flourished, their relations with local communities were better than often imagined -examples of collaboration (which tend to get overlooked) outnumber examples of conflict (which grab the historical headlines).[64] The strongest peasant movements, indeed, tended to be found in regions of Mexico -chiefly the rural centre- where foreign interests were sparse, compared to the north; and, of course, they steered clear of cities in general. Popular economic nationalism, to the extent that it existed, was a working class rather than a peasant phenomenon; but even working class economic nationalism was limited in scope and carne relatively late in the revolutionary cycle.[65]

 Much more important, however, was 'elite' economic nationalism, espoused by the rising leadership cadres of the revolution after 1913. In terms of timing, therefore, economic nationalism is a late starter -Compared to the pristine political liberalism of Madero. Economic nationalism, along with anticlericalism, forms part of a second ideological wave, which began to form following the fall of Madero (1913), impelled by the struggle against Huerta, by the sharp social polarization and the mounting financial difficulties which ensued; and it culminated in the policies of the 1920s and `30s: the war against the Church, `socialisf education, the struggle with the oil companies, the expropriation of 1938. Taken together, these were quintes- sential state-building policies: they were designed to bolster the new national regime, beat down its enemies (the Church and the oil companies were in some senses analogous anti-national institutions), and build firm bases of both popular support and financial stability. In more personal terms, they responded to Calles's almost obsessive desire to be 'master in his own house'.[66]

Integral to the revolutionary project of the 1920s and `30s, these policies were not, I repeat, coeval with the Revolution: neither economic nationalism nor anticlericalism figured in the pioneer projects of Maderismo or Zapatismo. And not only were the timing and motivation distinct. Economic nationalism was a minority interest (so, in some respects, was anticlericalism): it lacked the broad base of agrarianism and, one could say, was imposed from above, not driven from below. True, by 1938 the government could count on broad support for its oil expropriation.[67] But oil was - especially by the late 1930s- a somewhat special case, a perceived threat to national sovereignty and presidential authority, in a way that, say, foreign mining or manufacturing were not.

In general, the long-term policy of `nationalizing' Mexico's economic resources was the work of a minority group of políticos, técnicos, and some labour leaders: the políticos and técnicos entertained a grand project of state- building (which would have been anathema to Zapata and distasteful to Madero); the labour leaders utilized economic nationalism as a lever to pressure foreign employers.[68] It is also worth recalling that, while agrarianism was a distinct product of the Mexican Revolution, lacking close parallels elsewhere in Latin America at the time, economic nationalism was a commonplace: in Chile, Uruguay, even Bolivia. It represented a continent- wide reaction to foreign economic penetration, galvanized by `external shocks' (the First World War and, above all, the Great Depression), and advocated largely by literate, urban, official spokesmen.[69] The Mexican Revolution certainly helped stimulate a distinctive and quite powerful version of economic nationalism (the role of Manuel Peláez in the oil enclave was significant),[70] but this was not the most original aspect of the Revolution, and it is best seen as one of several threads in a skein of state-building measures which the Revolution provoked.

(d) Anticlericalism

Another important thread was anticlericalism. Here, the contrast between early indifference and later obsession is striking. Although the old PLM had begun (1900-1) as an anticlerical, `priest-baiting' organization, it rapidly mutated into a socially radical, quasi-anarchist, clandestine group, lacking a broad base in Mexican society.[71] The growth of political opposition in 1908- 10, although it was coloured by Catholic reformism, did not involve anticlerical agitation: Madero took a relaxed attitude toward the Church and, as President, welcomed and applauded Catholic political mobilization, which was particularly strong in the centre-west of Mexico.[72] Local (Maderista) liberals were less enamoured of Catholic political competition; but it was not until 1913-14, with the Huerta coup, the death of Madero, and the ensuing hardening of political enmities, that anticlericalism became a major item of revolutionary policy. Now, revolutionary leaders expelled priests, confiscated Church property, and began to enact formally anticlerical decrees -which culminated in the 1917 Constitution and its attempted subjugation of Church to State.[73]

The parallel with economic nationalism, already noted, is striking, in terms not only of timing, but also of motivation and constituency. Anticlericalism was not a mainstream, popular cause: in an overwhelmingly rural society, it emanated from the cities; it is hard to interpret it as a cynical bid for votes and mass support. On the contrary, like liberalism, anticlericalism was a highly 'ideological' creed, lacking a material rationale (at least among the middle class; the union leadership was another matter). It depended heavily on prior liberal/anticlerical `acculturation', hence garnered support among certain sociocultural groups -roughly, the literate, urban, petty bourgeosie, especially those raised in liberal/patriotic/Juarista homes and schools.[74] It tended to be stronger in the north, and particularly affected those `proconsular' northern leaders -Calles, Alvarado, Castro- who carne south to govern what they saw as a backward, Indian, priest-ridden society.[75]

Antielericalism also tended to be urban in origin (though there were a few anticlerical rancheros) and it strongly appealed to the infant labour movement, which -notably in Mexico City- was coloured by anarchist/ rationalist thinking.[76] Hence the making of the historic pact between the Mexico City workers and the anticlerical Constitutionalists in 1915: a pact which combined mutual self-interest with a shared urban, anticlerical, anti- rural ethos and which eventually evolved into the durable -and ardently anticlerical- alliance between Calles and the CROM in the 1920s. Calles, as I have already suggested, yearned to be `master in his own house', hence the Church, like the oil companies, had to be brought to heel. The CROM also had a direct interest in weakening the competition of Catholic labour unions. But, for Calles and for many anticlerical ideologues, the issue went beyond poltical self-interest -and it was certainly much more than a cynical smokescreen. Curtailing the power of the Church formed part of a broad and ambitious project to mould Mexican society, making Mexicans literate, sober, hard-working, thrifty, patriotic, and soberly scientific in their thinking. Thus anticlericalism was the central plank of a much broader platform which- for want of a better word- I term developmentalism.

(e) Developmentalism

Among my eight `genetic elements', this is, I suspect, the most original, contentious, and -semantically, at least- disconcerting. By `developmental- ism' I refer to a bundle of aims and policies designed to `rationalise and nationalize' the people of Mexico -a people whom the revolutionary leaders, for all their populism, often considered a low, inmoral, and degenerate breed.[77] The idea that the revolutionaries sought to `nation-build' -to inculcate sentiments of nationalism, to bring about a genuine integration of what, especially during the armed revolution, had seemed a fractured, fissiparous society- is a commonplace. Had not Manuel Gamio, one of the key ideologues of the Revolution, proclaimed the need to forjar patria in 1916?[78] But, in my view, `developmentalism' goes beyond this nation-building project and involves a yet more ambitious effort to remake Mexican society, to eliminate social vices and instill social virtues. Without such social engineering, many Mexican leaders feared, the country would lag behind the world and fall prey to more vigorous, expansionist nations, the US in particular.[79]

In one sense, developmentalism was old hat. As long ago as the colony, Mexico's governors had lamented popular vices and sought to impose a stricter morality and work ethic.[80] Nineteenth-century opinion-mongers, conservative as well as liberal, harped on this theme, alarmed by Mexico's dismal record of economic stagnation and national disaster; and, after 1876, it formed a staple of Porfirian/Científico thinking.[81] The Revolution of 1910, bringing further dislocation and disaster, gave added stimulus to these concerns: if Mexico was to survive, let alone prosper, it had to embark on a deliberate project of social regeneration. Quite possibly, the experience of the Revolution -the mass upheaval, death, and destruction- engendered a collective mood which, in Mexico as in other `post-trauma' societies, was broadly based and conducive to `reconstruction' and rebuilding.[82] If the popular mood was, to a degree, receptive, the revolutionary state was also capable and committed like none before it. Here lay the novelty of revolutionary developmentalism: not so much in the goals, which had a long history, as in the means, which, for reasons both particular and general, grew more effective after 1910.

A key example was education. Previous administrations had advocated education; some -Juárez's and Díaz's- had made some modest advances. The revolutionary regime, augmenting the power and scope of the federal government, committed more resources and efforts to education, especially in the countryside, where the bulk of the population still resided.[83] Education, too, was wedded to a clear ideology of nation-building, moralization, hygiene, hard work and production. Where an earlier generation of anticlericals attacked the Church for its wealth, or its illicit involvement in politics, the generation of 1910-40 inveighed against the entire value system and practices of Catholicism -particularly popular Catholicism [84]- which, they reiterated, encouraged sloth, drink, disease and superstition. Again, revolutionary ideology chimed in with working class anarchism, which stressed not only the evils of Catholicism, but also the need for workers to educate, discipline and morally uplift themselves.[85]

Revolutionary developmentalism, while updating and extending old concerns and objectives, could capitalize on new methods. The central government's commitment to a national education system reflected foreign examples (notably the French). Programmes of preventive medicine and national hygiene were, post-1918, a staple of European politics. State- sponsored art-for-the-masses, represented by the famous revolutionary murals, also had European -fascist and Soviet- counterparts. The new fashion for corporatist party organization and military-style coloured shirts, pioneered in Europe, soon found its way to Mexico: Garrido Canabal's Tabasco, where moralization, anticlericalism, and uniformed (Red Shirt) politics reached their apogee, became the so-called `laboratory of the Revolution'.[86] And, though the revolutionary generation placed inordinate faith in the power of the printed word, they also recognized the political potential of new technology, seeking to harness radio and film to the task of `rationalizing and nationalizing' the Mexican people.[87]

In short, revolutionary `developmentalism' represented an aggressive new twist to an old tale: the secular efforts of government to create hard-working, productive, model citizens. The Revolution brought to power new elites, desperately concerned both to consolidate their power and to `develop' the Mexico people, in the face of threats and obstacles both at home and abroad. The upheaval of the Revolution gave added cause for concern, but also, perhaps, afforded propitious circumstances for such a developmentalist project.[88] In addition, the global conjuncture -social, political and technological- favoured ambitious projects of state social engineering; the revolutionaries of 1910-40 were therefore more eager, less inhibited, than their Porfirian precedessors of 1880-1910 had been; the notion that the estado rector could assume the leadership of a national developmentalist crusade was now common wisdom around the world.

(f) Labour Reform

I turn, penultimately, to labour and agrarian reform. These are often seen, with some justification, as the twin ideological pillars of the Mexican Revolution, exemplified by articles 27 and 123 of the 1917 Constitution. Frequently, too, they are coupled, on the grounds that they represent the `social' -as opposed to the `political' conquests of the Revolution. While this makes some sense, it is also necessary to distinguish between-as well as lump together- these two items.

There is an initial paradox: although the organized working class was a small minority of the Mexican population and made a very limited contribution to the armed revolution, it rapidly became a favourite son of the revolutionary regime. The alliance of the Casa del Obrero Mundial with Obregón, or of the CROM with both Obregón and Calles, gave these organizations privileged access to govenment patronage; and, since the CROM/Calles alliance of the 1920s was more deep and durable, the pay-offs were correspondingly greater. The CROM won governorships, seats in the cabinet, pro-labour decrees and favourable verdicts in labour arbitration tribunals. Employers lamented this accession of working class power -which was unique in Latin America in the 1920s.[89] Of course, the CROM did not represent the entire organized working class, still less the entire Mexican working class: it tended to recruit weakly organized unions and state confederations, not the major industrial unions (the oil, mining and railway workers), whose relations with the CROM were ambivalent, even hostile. In addition, a large segment of Mexican labour -artisans, workers in small workshops in provincial towns, rural jornaleros- remained unorganized. The benefits accruing to labour were therefore skewed and depended on political deals, even a degree of institutional subordination to the state, which the state in turn justified in the narre of social justice and equilibrium. Recognising the existence of class struggle, the state undertook to support the weaker party (the workers), but it did not seek the victory of the proletariat, nor did it strive to eliminate capitalism. Under the guise of this benevolent tutelage, the revolutionary state -and its syndical allies- enjoyed a wide margin of manoeuvre. Labour policy veered right during the Maximato but swung left with Cárdenas, whose notion of social equilibrium implied a more radical stance than Calles's had.[90]

Hence the halts and detours in labour reform. The universal promises of article 123 viere -like the democratic provisions of the Constitution- honoured in the breach: it took fourteen years for a Labour Code (1931) to be produced; and, in some regions and industries, it remained a dead letter. During the subsequent decade, labour mobilization and reform quickened pace dramatically; the balance of political power shifted towards the infant CTM which, at the outset, represented the most radical and effective sector of the new PRM (1938). Briefly, Lombardo Toledano appeared to be second only to the President as a mover and shaker in Mexican politics. Business interests vocally deplored this precipitate slide into socialism -evidenced in radical rhetoric, proliferating strikes, and rash experiments in state or worker control of industry.The structural links between the regime and labour (now the CTM) strengthened, cementing a distinctive Mexican form of `labour incorporation' and, crucially, affording the revolutionary regime an important basis of mass support, which helped Obregón resist de la Huerta in 1923-4, and Cárdenas overcome Calles in 1934-6.[91] But incorporation did not (yet) imply either tight political control or an abdication of labour militancy. These came later, particularly following the charrazos of the later 1940s.

Labour reform and mobilization viere therefore key characteristics of revolutionary ideology. They viere also relatively, though not totally, original. Like most of the major Latin American countries, Mexico became aviare of the so-called `social question' around the turn of the century.[92] The strikes of 1906-7 illustrated the rise of more militant unions, which went beyond the modest mutualism of earlier years. The Porfirian state resorted to repression; but it also tempered repression with mediation, showing that, in Mexico as elsewhere, even `oligarchic' governments saw the necessity -even perhaps the advantage- of conciliating working class groups, especially those in the key export sectors.[93] The Revolution therefore had the effect not of introducing labour reform and concilation de novo, but rather of decisively accelerating an incipient trend.

(g) Agrarian Reform

Agrarian reform was a different matter. In 1910 the peasantry constituted the great bulk of Mexico's population; their demands posed a threat to property rights, particularly those of Mexican -rather than foreign- landowners. The peasants viere also the shock-troops of the Revolution: they forced themselves on the national agenda by virtue of their armed mobilization in the 1910s and their continued political mobilization in the 1920s and `30s. The regime of the Porfirian landed oligarchs collapsed; denials of the `agrarian problem' carne to shoud increasingly hollow; and the Sonoran regime of the 1920s, building on Carrancista precedent, had perforce to recognise agrarian demands. But its recognition was grudging, since the Sonorans -men of the north, tainted by northern racism- were skeptical about corporate (ejidal) landholding and saw efficient private farming as the way forward. Obregonista agrarianism was, therefore, limited and tactical: the unusually sweeping agrarian reform in Morelos reflected the force of Zapatismo and the need to conciliate Zapatista interests. But it remained a local reform; Zapatismo had failed to dictate the national agrarian agenda. While the Calles government may be fairly termed refomist in labour matters (Calles proudly described himself as a `labourist'), its agrarian credentials were far weaker, especially as the President, soon Jefe Máximo, turned against the agrarian reform and, in 1930, sought to terminate it.[94] At the local level, I would argue, the Revolution retained strongly agrarian characteristics throughout the 1920s; but at the national level, it was not until the Cárdenas presidency that agrarianism became an integral part of both revolutionary ideology and practice.

This was a decisive, original and radical step.[95] In distributing some 18m hectares of land, and establishing collective as well as individual ejidos, the Cardenista government reversed decades of Porfirian policy and achieved results unprecedented in Latin America. For, while labour reforms, even oil expropriations, had parallels elsewhere in the Continent, the Cardenista agrarian reform had none.[96] It dealt a blow to private property, reconstituted the Mexican peasantry, both economically and socially, and -over time, at least- subordinated that peasantry to the authority of the revolutionary state, thus enhancing political legitimacy and stability alike. While this could be seen as a belated triumph of Zapata's Plan de Ayala, such a conclusion should be qualified, in at least two respects. First, the Zapatistas had favoured small-scale family farming and had no brief for the giant collectives which Cárdenas set up in the hope -sometimes realized, sometimes disappointed- of fostering profitable cash-crop production. And, more important, the 1930s agrarian reform carne from on high, as the gift of a burgeoning central government, with political strings attached. While Zapatista agrarianism had seen access to land as a traditional right to be restored (restitución), Cardenista agrarianism conceded land as a conditional gift (dotación) of the revolutionary state. The former pointed towards a loose confederation of autonomous agrarian communities, the latter favoured the creation of a centralized paternalistic bureaucratic state. Both Zapatismo and Cardenismo were socially radical (see accompanying diagram); but their respective radicalisms had quite different implications for state-building, hence must be differently located on the centralization/parochialism axis.
(h) Indigenismo

Indigenismo is best treated alongside agrarianism, although its origins and role are, in fact, very different. Like agrarianism, indigenismo became a key item of revolutionary ideology after 1917, and of revolutionary practice during the 1930s: in other words, it was not until the 1930s that ejidal distribution and federal education began to have decisive results in Mexico's `regions of refuge' -Chiapas being a classic example.[97] As a practical policy to help, educate, protect, and `revalorize' Indian communities, indigenismo was therefore a product of the Cárdenas era. The case of the Yaquis was, perhaps, atypical, but nevertheless suggestive: where Obregón and Calles resorted to counter-insurgency and repression, Cárdenas (`Tata Lázaro') brought schools and ejidos.[98] As a means of effective mobilization, therefore, indigenismo depended heavily on the state's commitment to agrarian reform and rural education, without which it would have remained what it had been during the 1920s: a somewhat cerebral and rhetorical phenomenon, the work of intellectuals, políticos, artists, and `applied anthropologists' -not of Indians themselves.

Indeed, it is scarcely an exaggeration to see indigenismo as a `top-down' programme, conceived by mestizo ideologues who -whether sincerely or cynically-saw it as an appropriate instrument of the new revolutionary regime as it set about bolstering its fragile legitimacy. During the 1920s, indigenista symbols were certainly not lacking: they were to be found on the walls of the Ministry of Education, on CROM letterheads, and in the pages of Gamio's pioneering anthropological work. But neither the makers nor the consumers of these symbols were Indians; unlike agrarianism -which the insurgent peasantry forced on to the national agenda -indigenismo infiltrated from above and trickled down to its supposed beneficiaries, sparingly in the 1920s, more torrentially in the 1930s. As such, it was, from the start, a centralizing and assimilating ideology, attractive to ideologues and políticos who were keen to forjar patria. Where agrarianism was ambivalent -its Zapatista form was decentralizing, its later Cardenista form centralizing-, indigenismo was at all times a vehicle of political and cultural integration, an instrument of state- building. Practically speaking, it combined with agrarismo (since it often shared the same means, e.g., federal schools, and the same targets, i.e., the Indian campesinos); but in terms of goals and motives, it had more in common with anticlericalism, for it sought to change mentalities to the benefit of state and nation, and was channelled from above, not below.

What is more, it had a long lineage. Revolutionary indigenismo was new in terms of its mass application in the 1930s; but the notion of assimilatiog Indian communities and rehabilitating Indian culture was old. Creole patriotism, as David Brading has shown, embodied proto-indigenista symbols; `Mexico' -as an `imagined community'- traced back to pre-Conquest, especially Aztec, foundations.[99] Nineteenth-century ideologues -including positivistic Porfirians- displayed a patriotic reverence for the country's august Indian past, even if they deplored its degenerate Indian present.[100] Revolutionary ideologues certainly gave this old tradition a new twist: they spurned the more overt racism of their Porfirian predecessors and credited Indian culture with greater value and resilience. But they had no doubt that Indians ought to assimilate to the mestizo nation state and that the state was justified in taking strenuous measures to achieve assimilation. Indigenismo therefore enjoyed an oddly incestuous relationship with the revolutionary cult of mestizaje, pioneered by the arch-ideologue of the Revolution, Andrés Molina Enríquez.[101]

What finally prevented revolutionary indigenismo from becoming a purely elitist imposition was the practical experience of the 1930s: for, as reformers - teachers, anthropologists, union and party organizers- `went to' the Indian during that decade, they initiated processes which they could not fully control; hence -in Chiapas, for example- a new cadre of Indian activists arose, who partially took charge of the state's indigenista programme.[102] As a result, real flesh was put upon the skeletal bones of indigenismo, making some sort of social reality out of old symbols and slogans. Again, therefore, the distinctive feature of the Revolution was less the creation of radically new ideas than the practical implementation -and, as a result, the partial transformation- of ideas that enjoyed a long history in Mexico, and/or clear parallels elsewhere in Latin America.[103]

This, indeed, must be the clearest conclusion of my analysis. The Mexican Revolution -as Tannenbaum saw long ago- was notable less for its ideological novelty than for its successful implementation of ideas that were, in general, well-known, even commonplace. Political liberalism, patriotism, anticlericalism, developmentalism and indigenismo were all familiar elements of the pre-revolutionary -in some cases even the colonial- ideological landscape. Economic nationalism and labour reform were relatively new, being products of late nineteenth-century desarrollo hacia afuera; as such, however, they were by no means confined to Mexico and were to be found throughout Latin America. The distintinctiveness of Mexican revolutionary ideology, therefore, lay in its practical application, the result of an unprecedented social revolution -a revolution which, pace Keynes, was the result of objective social conditions- and random political vicissitudes -rather than of subjective ideological perceptions. Thus, the trayectoria ideológica of the revolution tended to follow events and processes rather than determine them: Maderista liberalism failed because of both structural weaknesses and conjunctural obstacles; anticlericalism -dormant during the early Revolu- tion- burst into life following the Huerta coup; economic nationalism was decisively stimulated by the economic collapse of 1915-17 and the economic depression of 1930. What I have identified as highly `ideological programmes - i.e., those most driven by the force of ideas and least endowed with supportive material interests and social constituencies (e.g., anticlericalism, middle-class political liberalism)- were less capable of achieving their goals, and of transforming Mexican society than those which coupled ideology (often a fairly bland, commonplace or derivative ideology) to powerful interests and constituencies (e.g., agrarianism).

The result was a Revolution which followed a pragmatic course and did not adhere to any strict blueprint. Mexican revolutionaries killed each other in the pursuit of power, but they did not parade their ideological virtue in political show trials like their Soviet counterparts. Nor did the Mexican Revolution seek to export itself as a world-historical model, even if aspects of the Revolution certainly influenced some Latin American leftists -Haya de la Torre and Sandino, for example. The genetic material of the Revolution was therefore Mexican, hardly suitable for extensive cross-breeding in foreign climes. This ideological particularism was both a strength and a weakness. It served as a deterrent to dogmatism, but also enabled a host of competing interests to claim the mantle of the Revolution -while, in somewhat insular, even chauvinist fashion, rejecting `exotic doctrines' imported from abroad.[104]This peculiarly national revolution, possessed of a vague, eclectic ideology, was unusually resistant to external upheavals (the rise and fall of Communism had less impact in Mexico than perhaps any other Latin American country); but it was also capable of endless Jesuitical re-invention by opportunistic politicians and their pet intellectuals. Long after the revolutionary generation of 1910-40 had left the scene, Mexico's leaders continued to legitimize themselves -and every twist and turn of their policy- in the name of the Revolution. Even Salinista neo-liberalism, which made a bonfire of so many revolutionary heirlooms, claimed to do so in the name of the Revolution. To the extent that such legitimizing claims were believed (and often, no doubt, they got a mixed reception), we may speak of the enduring success of Mexico's revolutionary ideology; a success which had little to do with intellectual novelty or consistency, but a lot to do with political structures, affective allegiances, and cultural diffusion. As Slavoj Zizek observes, with only a touch of hyperbole, "an ideology really succeeds when even the facts which at first sight contradict it start to function as arguments in its favour".[105]

Now, as the millenium draws lo a close and the centennial of the Revolution stands only thirteen years away, we are tempted -and perhaps justifed- in answering the old question -'ha muerto la revolución mexicana?'- with a confident affirmative. But, bearing in mind my genetic metaphor, we should perhaps be more cautious: recessive genes can slumber through generations before fortuitous recombination, coupled with changing environ- ment, brings about their reawakening. The result, of course, is not a carbon copy of some long-dead person, but rather a revival of specific ideological traits, which, in the current circumstances, possess peculiar relevance or utility. Díaz appeared to have laid Church-State conflict to rest, but it flared up with even greater vehemence after 1913, responding lo political events; Calles pronounced the end of the agrarian reform in 1930 yet Cárdenas, pushed by the peasantry, soon enacted the greatest reparto of all. On the day that NAFTA carne into being, seeming to crown Salinas's neo-liberal reforms, the Chiapas rebels startled the world, evoking the old revolutionary symbol of Zapata. They have not been silenced. It would be rash lo assume that the genetic material of the 1910 Revolution is definitively dead, rather than quietly dormant.


[1]Ian MacKenzie, Introduction: The Arena of Ideology', in Robert Eccleshall et al., Political Ideologies: An Introduction (London, 2nd ed., 1994), p. 1, citing Malcolm Hamilton.

[2] Victoria Lerner, Historia de la Revolución Mexicana: Período 1934-40. La educación socialista (México, 1979), p. 83.

[3] Clifford Geertz, The Interpretation of Cultures (London, 1993, first pubd. 1973), p. 94.

[4] Alvin Gouldner, The Two Marxisms (London, 1980), ch. 2.

[5] Donald N. McCloskey, `The Economics of Choice: Neoclassical Supply and Demand', in Thomas G. Rawski, et al., Economics and the Historian (Berkeley, 1996), pp. 122-58.

[6] Ernest Gruening, Mexico and its Heritage (London, 1928), p. 342.

[7] On `fictitious ejidos': Frans J. Schryer, Ethnicity and Class Conflict in Rural Mexico (Princeton, 1990), pp. 134-7.

[8] James C. Scott, Domínation and the Arts of Resistance: Hídden Transcripts (New Haven, 1990). It may be useful to add a third dimension: the `international public transcript' -i.e., what the government wants foreign audiences to hear and believe, which may be different from both its domestic public transcript and (a fortiorí) its domestic hidden transcript. For an interesting analysis of United States' perceptions of the Revolution, see John A. Britton, Revolution and Ideology. Images of the Mexican Revolution in the United States (Lexington, 1995).

[9] Alicia Hernández Chávez, La tradición republicana del buen gobierno (México, 1993 ) is a good (i.e., a bad) example of formalism.

[10] Gruening, Mexico and its Heritage, pp. 399-493, offers some graphic examples.

[11] Alan Knight, The Mexican Revolution (2 vols., Cambridge, 1986), v. 1, ch. 6. Zapatista `liberalism' was a somewhat different species - more parochial, decentralizing, informal. Nevertheless, if Womack is right in stressing the Zapatistas' attachment to civilian politics and electoral representation, it would be wrong to exclude Zapatismo from consideration as a `politically liberal' movement: see John Womack Jr, Zapata and the Mexican Revolution (New York, 1969), pp. 3-9, 225-8. As I suggest elsewhere, Zapatismo was also `liberal' in a somewhat different, historical and subjective sense, that of seeing itself as part of the liberal- patriotic tradition of Juárez.

[12] Eyler N. Simpson, The Ejido: Mexicós Way Out (Chapel Hill, 1937), pp. 440-3.

[13] Robert Darnton, The Kiss of Lamourette (New York, 1990), p. 206.

[14] Ibid., pp. 206-7.

[15] Frank Tannenbaum, Peace By Revolution (New York, 1966; first pubd. 1933), pp. 115-6.

[16] This is, of course, a'subjective judgement; but cf. Sheldon B. Liss, Marxist Thought in Latin America (Berkeley, 1984), pp. 129-37 and 219-26, which seems implicitly to corroborate it.

[17]There is a story of Obregón, accompanied by his idealistic Education Minister Vasconcelos, visiting a poor, remote pueblo, where the campesinos can barely eke out an existente amid arid fields; "ah", Obregón exclaims to Vasconcelos, "what these people clearly need are a few more volumes of Goethe and Plato".

[18] This is clearly connected to the fact that the French Revolution, by repudiating monarchy and feudalism, dramatically broke new ideological ground, in a way that the Mexican Revolution - with its invocation of Juarista liberalism - did not; a difference which must caution us against the blanket application of `French revolutionary' models to Mexico (compare Frangois-Xavier Guerra, Le Mexique: De l'ancien régime á la révolution [Paris, 2 vols., 1985]). The English Revolution of the seventeenth century offers a comparable example of a major revolution which remained largely national in scope and embodied `traditional' prescriptive elements (radical Protestantism and `Anglo-Saxonism'). This is not to say that the Mexican Revolution lacked intemational impact or foreign sympathizers; simply that the Mexican revolutionaries did not make the export of the revolution a central task.

[19] Mary Kay Vaughan, Cultural Politics in Revolution: Teachers, Peasants, and Schools in Mexico (unpubd. ms., p. 100; publication forthcoming by the University of Arizona Press, Tucson).

[20] John Maynard Keynes, The General Theory of Employment, Interest and Money (London, 1936), pp. 383-4.

[21] Idealistic' in the strict sense of stressing the autonomy and importante of ideas in history; not necessarily in the - broader and more common - sense of `principled and optimistic' (though Keynes was probably that too). I am proposing, instead, a closer - though far from simple - relations between ideas and social groups: what Geertz, The Interpretation of Cultures, p. 201, refers to - and criticizes - as the `Interest theory'.

[22] Knight, Mexican Revolution, I, p. 295.

[23] David Hackett Fischer, Historians' Fallacies (New York, 1970), pp. 167-9, terms this the cum hoc propter hoc fallacy.

[24] Robert C. Millon, Mexican Marxist: Vicente Lombardo Toledano (Chapel Hill, 1966) is fairly `ideological' in approach; so, too, is a good deal of Mexican labour history. For all its ideological outpourings, Lombardo's CTM - like Mexican parties and governments - worked on the basis of caciquismo, patronage, and even coercion; ideology was not irrelevant, but any study which places ideology on a pedestal is bound to mislead.

[25] Alan Knight, `Land and Society in Revolutionary Mexico: The Destruction of the Great Estates', Mexican Studies/Estudios Mexicanos, 7/1 (winter 1991), pp. 82-8.

[26] Rodney D. Anderson, `Mexican Workers and the Politics of Revolution, 1906-11', Hispanic American Historical Review, 54/1 (1974), pp. 94-113.

[27] Knight, Mexican Revolution, II, pp. 9-11.

[28] Ernesto Laclau, Politics and Ideology in Marxist Theory (London, 1977), especially pp. 97- 9, whence the quote. Of course, this comes as a blinding revelation only for those who assumed - usually on the basis of crass economic reductionism masquerading as Marxism - that ideological stance invariably correlated with material/class interest; an assumption more common then than now. Note that the rejection of this reductionist assumption does not imply that ideology consists of free-floating `signifiers', detached from material or class interest; an implication which some, following Laclau and other postmodern theorists, seem to have drawn. `Elective affmities' captures, I think, the right balance between reductionist determinism on the one hand and serendipitous idealism on the other.

[29] This is a particularly knotty question in regard to Villismo and Carrancismo: Knight, Mexican Revolution, II, pp. 262-303. For me, `factions' and `coalitions' are, respectively, smaller and larger versions of the same phenomenon, as described in the text.

[30] On Zapatismo: Knight, Mexican Revolution, I, pp. 309-15.

[31] Womack, Zapata, pp. 235-6, 396-8, 405.

[32] Knight, Mexican Revolution, II, pp. 342-4. As this example suggests, `anti-imperialism' may be less an ideological constant than a political expedient; as such, it is not a good litmus test of ideological identity. This point links to the discussion of patriotism and economic nationalism below.

[33] Nora Hamilton, The Limits of State Autonomy: Post-Revolutionary Mexico (Princeton, 1982), ch. 8.

[34] Laurens Ballard Perry, Juárez and Díaz: Machine Politics in Mexico (De Kalb, 1978).

[35] Knight, Mexican Revolution, I, ch. 6. Although Madero departed from liberal principle in several respects, as mentioned, this did not prevent his administration being the most liberal of the period under discussion; hence, I accord that administration `politically liberal' credentials in the accompanying diagram. Of course, had Madero survived, he might have grown more illiberal with time; or, we might even reverse the relationship: had Madero grown more illiberal with time, he might have survived. That, at least, was Carranza's considered conclusion.

[36] Knight, Mexican Revolution, II, pp. 379-80.

[37] See the accompanying diagram, which embodies the graphic skills of Henry Knight. `Social radicalism' could no doubt be measured in several different ways. Given the context of a capitalist Mexico, 1 tend to take (private) property rights as a key indicator: policies which infringe such rights (e.g., agrarian reform) are `socially radical'; protection of such rights (e.g., the agrarian amparo) are `socially conservative'.

[38] Madero ruled out agrarian reform; Carranza `devolved' confiscated land to landlords; Calles, after a modest burst of reform, declared the process terminated: Knight, Mexican Revolution, 1, pp. 416-23, II, pp. 465-6, 489-90; Simpson, The Ejido, pp. 440-3. Calles also moderated his initial (1925-7) `anti-imperialism' and, responding to the blandishments of Ambassador Morrow, became more friendly to the United States. In general, I think, the US could more easily pressure the constituted governments of the 1920s (Obregón's or Calles's) than the shifting factions of the 1910s (e.g., Villismo, Carrancismo); again, therefore, incumbency tended to produce more `responsible', conservative, amenable authorities.

[39] Hamilton, Limits of State Autonomy, especially ch. 8.

[40] I subscribe to the conventional, though not unanimous, view that c. 1940 the Mexican regime experienced a significant shift in both policy and personnel, characterized by the passing of a political generation, of its particular concems and conflicts. Thereafter, both the domestic and intemational contexts changed rapidly and radically. As I argue in this paper, however, it would be rashly anthropomorphic to speak of the definitive `death' of the Revolution.

[41] "The dominant elite's flattering self-portrait... is not without its political costs since such disguises can become a political resource for subordinates. Ruling groups can be called upon... to live up to their own idealized presentation of thmselves to their subordinates": Scott, Domination and the Arts of Resistance, p. 54; see also pp. 10-11, 76, 103, on the Tace- grows-to-fit-the-mask' syndrome.

[42] Adam Przeworski, Democracy and the Market (Cambridge, 1991), p. 3.

[43] Knight, Mexican Revolution, II, p. 518, points to the relative consistency of the Sonoran revolutionaries' goals, policies, and accomplishments, and criticizes allegations of the Sonorans' 'betrayal' of the Revolution as 'normative nonsense'. Marjorie Becker, Setting the Virgin On Fire (Berkeley, 1995), p. 3, n. 8, derides this comment as normative nonsense itself. Like many of the throwaway lines which litter this bizarre book, Becker offers no clarification or justification. Presumably, she things the Sonorans did betray the Revolution. If so, what was the Platonic universal THE REVOLUTION - which they betrayed? Is it normative good sense - as against nonsense - to blame historical actors for doing things they said they would do, or not doing things they never said they would do?

[44] In the accompanying diagram I have tried to depict the genetic material of four major factions/coalitions: Maderismo, Zapatismo, Callismo and Cardenismo. Other groups could, of course, be included - even those of non-revolutionary provenance (Porfirians, Huertistas, clericals, social Catholics).

[45] One Treakish' factor was Cárdenas's own character - which Calles clearly misread. In addition, the circumstances of the depression favoured popular mobilization, while calling the preceding project of desarrollo hacia afuera into question and lessening the opportunity cost of reform, notably agrarian reform.

[46] Armando Bartra, Los herederos de Zapata (Mexico, 1985).

[47] As suggested above (n. 40), I see the revolutionary period, c. 1910-40, as possessing a certain historical unity, premised on the rise and fall of the revolutionary generation, its goals, struggles and accomplishments. For this reason, 1910 and 1940 mark useful boundaries; 1920, though it marked the end of the armed revolution (i.e., the last successful armed revolt), is not, in my view, ideologically significant.

[48] As mentioned above (n. 44), this is meant to be an illustrative, not an exhaustive, depiction of ideological alignments in the period. 'Callismo', in particular, represents a broad current of opinion which antedates the Calles presidency, which might alternatively be labelled Sonoran ideology', or suchlike, and which both resembles and traces back to Carrancismo/ Constitutionalism: Knight, Mexican Revolution, II, pp. 508-9, 517-8, 527-8.

[49] For an interesting analysis of the key indicator of property rights, see Carlos Elizondo, `property Rights in Mexico: Government and Business After the 1982 Bank Nationaliza- tion', Oxford University D. Phil. diss. (1992), ch. 2.

[50] Tulio Halperin Donghi, 'Argentina: Liberalism in a Country Born Liberal', in Joseph Love and Nils Jacobsen, eds., Guiding the Invisible Hand: Economic Liberalism and the State in Latin America, (New York, 1988), pp. 99-116.

[51] Research into nineteenth-century popular, patriotic, liberalism is growing fast: major contributions are: Florencia Mallon, Peasant and Nation: The Making of Postcolonial Mexico and Peru (Berkeley, 1995); and several works by Guy P.C. Thomson, notably, 'Bulwarks of Patriotic Liberalism. The National Guard, Philharmonic Corps and Patriotic Juntas in Mexico, 1847-8', Journal of Latín American Studies, 22 (1990), pp. 31-68.

[52] Arnaldo Córdova, La ideología de la revolución mexicana (Mexico, 1973), p. 87.

[53] A caveat is in order: Díaz, Limantour, and the Científicos certainly subscribed to economic liberalism, in that they favoured free markets and prívate property; however, in order that these beneficial arrangements might flourish, the regime had to create the right conditions - for example, by subsiding infrastructure, protecting industry, even coercing labour. This meant that the Porfirian state could not be content with the passive role of a nightwatchman. We should note, however: (1) that state intervention was seen as temporary and targeted; (2) that, by serving economic liberal goals, it clearly differed from, say, the state intervention of the 1930s; and (3) that plenty of developing market societies went through a phase of state-promoted protection and even `primitive accumulation' - the United States, for example.

[54] Knight, Mexican Revolution, 1, pp. 69-70.

[55] On Maderista social concern, Knight, Mexican Revolution, 1, pp. 58, 64, 138-9.

[56] Córdova, La ideología de la revolución mexicana, p. 110.

[57] Knight, Mexican Revolution, 1, p. 58. Recent research on the early national period - chiefy the 1820s - depicts a more vigorous, participatory, electoral politics than previously imagined. Nevertheless, political closure and praetorianism soon supervened, and even the liberal triumphs of the 1850s and '60s failed to introduce a functioning representative system; hence the novelty and radicalism of Madero's programme. For some good examples of the new electoral history, see Eduardo Posada Carbó, ed., Elections Before Democracy (London, 1996).

[58] Knight, Mexican Revolution, II, pp. 9-10, 91-3, 102-4, 446-7.

[59] John Skirius, Vasconcelos y la cruzada de 1929 (Mexico, 1978).

[60] Compare the relevant volumes on Madero, Calles and Cárdenas in Enrique Krauze, Biografías del poder (Mexico, 1987).

[61] Hence the recent vogue for Tocquevillean' theories of the Mexican Revolution (not to mention other revolutions): theories which stress the revolution's state-building accom- plishments rather than, say, its class or social reformist character: Alan Knight, `Viewpoint: Revisionism and Revolution: Mexico Compared to England and France', Past and Present, 134 (Feb. 1992), pp. 166-79.

[62] I see xenophobia as a more broadly-based, popular, sentiment, directed against foreign residents - e.g., Chinese and Spaniards (gachupines), who, unfke the big Anglo-American companies, were viewed as cultural interlopers, price-gougers, or rivals in competitive labour markets: Alan Knight, U.S.-Mexican Relations, 1910-40: An Interpretation (San Diego, 1987), pp. 53-70. To the extent that xenophobia is relevan[ to the present discussion, it belongs less with economic nationalism than with indigenismo (see below).

[63] Lorenzo Meyer, Mexico and the United States in the Oil Controversy, 1917-42 (Austin, 1977), pp. 30-1.

[64] Knight, U.S.-Mexican Relations, pp. 24, 58-62.

[65] It was particularly evident in the oil sector; but oil was far from typical: Knight, U.S.- Mexican Relations, pp. 35-6, 81-7.

[66] Ibid., pp. 82-3.

[67] Although this needs to be qualified: Mexican reactions to the oil expropriation were more mixed, and more manipulated, than many accounts suggest: Alan Knight, `The Politics of the Expropriation', in Jonathan C. Brown and Alan Knight, eds., The Mexican Petroleum Industry in the Twentieth Century (Austin, 1992), pp. 90-128.

[68] Gruening, Mexico and its Heritage, p. 358.

[69] It is my general impression that the role of the supposed `national bourgeoisie' in the genesis of modern Latin American economic nationalism has been exaggerated; that of government officials and técnicos underestimated. Certainly in Mexico it is surprisingly hard to detect the clamorous nationalist bourgeoisie, at least before the 1940s, by which time economic nationalist provisions (e.g., article 27) were well established.

[70] Jonathan C. Brown, Oil and Revolution in Mexico (Berkeley, 1993), ch. 4 is the best study.

[71] Knight, Mexican Revolution, I, pp. 44-7. Other historians attribute greater importance to the Magonistas: for example, Ward Albro, Always A Rebela Ricardo Flores Magón and the Mexican Revolution (Forth Worth, 1992).

[72] Knight, Mexican Revolution, 1, pp. 392, 398, 400-2.

[73] It is worth stressing that, whereas Mexico's nineteenth-century liberals sought to seperate Church and State (e.g., by instituing civil marriage), revolutionary anticlericals went a good deal further, demanding not separation but - the Catholics complained - subjugation: e.g., the state's right to control Catholic education, to register priests, hence to control their numbers.

[74] Alan Knight, 'Popular Culture and the Revoluthionary State in Mexico, 1910-1940' Hispanic American Historical Review, 74:3 (1994), pp. 415-9.

[75] On the role of the northern proconsuls, see Knight, Mexican Revolution, II, pp. 240-51.In some cases- such as Calles- anticlericalism assumed obsessive proportions and, perhaps, responded to deep psychological causes: e.g., Calles's resntment at the stigma of illegitimacy. This is about the only example I can find of Mexican revolutionary ideology conforming to Geertz's 'strain' theory: that ideology as part of a "chronic effort to correct sociopsychological disequilibrium": The Interpretation of Cultures, p. 201. It is also about the only instance where Freudian psychohistory (possibly) rises above the level of the silly or the purely speculative.

[76] I take this anarchist influence as a given; by way of - very sketchy - explanation, five factos might be mentioned: (1) the prevalence pf small artisanal workshop, which favoured anarchist over socialist/collectivist solutions; (2) the existence of a corrupt, autoritarian (Porfirian) state, which stultified parliamentary politics; (3) the presence of a powerfull Catholic Church, which provoked rationalist, anticlerical opposition; (4) the indigenous liberal/patriotic tradition, which could mutate into libertarian anarchism more easily than into statist socialism; and (5) the role of foreign, especially Spanish, ideologues and literature.

[77] Knight, 'Popular Culture', pp 394-5, 403-4; and Alan Knight, 'El liberalismo mexicano desde la Reforma hasta la Revolución (una interpretación)', Historia Mexicana, 35/1 (julio-septiembre 1985), pp. 67-9.

[78] Manuel Gemio,Forjar Patria (Mexico, 1916); David Brading, 'Manuel Gamio and Official Indigenismo in Mexico',Bulletin of Latin American Research, 7/1 (1988), pp. 75-90.}

[79] Behind this thinking lay a set of fairly racist and social-Darwinist assumptions, which coloured much of Mexican revolutionary ideology. We tend to focus on the more positive (indigenista) side of revoluthionary ethnic/racist thinking; but in th 1920s Sinophobia flourished, and in the 1930s, a measure of anti-Semitism: Alan Knight, 'Racism, Revolution and Indigenismo: Mexico, 1910-1940', in Richard Graham, ed., The Idea of Race in Latin America (Austin, 1990), PP/ 86-95.

[80] Cf. Juan Pedro Viqueira Alban, ¿Relajados o reprimidos? Diversiones públicas y vida social en la ciudad de México durante el Siglo de las Luces (Mexico, 1987). Of course, Bourbon 'developmentalism' was not - indeed, could not be - nationalistic.

[81] Knight, 'El liberalismo', pp. 60-1.

[82] Gordon Wright, The Ordeal of Total War (New York, 1968), p. 235.

[83] Mary Kay Vaughan, The State, Education and Social Class in Mexico, 1880-1928 (De Kalb, 1982).

[84] Knight 'Popular culture', pp. 407-10; Becker, Setting the Virgin on Fire; Adrian A. Bantjes, 'Burning Saints, Molding Minds: Iconoclasm, Civic Ritual and the Failed Cultural Revolution' in William H. Beezley, Cheryl English Martin and William E. French, eds., Rituals of Rule, Rituals of Resistence (Wilmington, 1994), pp. 260-84. It is worth stressing popular Catholicism, since, in a good many cases, the Church hierarchy was also concerned to extirpate bibulous fiestas and bizarre superstitions; despite their bitter enmity, clerical and secular elites entertained surprisingly similar notions about popular degeneracy.

[85] Alan Knight, 'The Working Class and the Mexican Revolution c. 1900-1920' ', Journal of Latin American Studies, 16 (1984), pp. 51-79.

[86] Carlos Martinez Assad. El laboratorio de la revolución. El Tabasco garridista (México, l979).

[87] Ibid., pp. 47, 148 on Garrido's 'anticlerical'and 'anti-alcohol'radio programmes (which were not confines to Tabasco).

[88] Consider, by way of comparsion, the postwar rebuilding of Japan; or, the collective commitment which went into the establishment of the state of Israel.

[89] For example, Gruening, Mexico and its Heritage, pp. 10, 36.

[90] Córdova, La ideologia revolucion mexicana, pp. 268-75; Plutarco Elías Calles, Mexico Before the World (New York, 1927), pp. 10, 36.

[91] Collier and Collier, Shaping the Political Arena, pp. 208, 237.

[92] James O. Morris Intellectuals, and Consensus: A Study of the Social Question and the Industrial Relations System in Chile (Ithaca, N.Y., 1966), ch. 4.

[93] On export sector militancy, see Charles Bergquist, Labor in Latin America: Comparative Essays on Chile, Argentina, Venezuela and Colombia (Stanford, 1986).

[94] Calles, Mexico Before the World, p.8.

[95] This needs to be stressed, given that revisionist interpretations of the Revolution have tended to downplay its radicalism, not least in regard to the agrarian reform, which they depict as a calculating policy of 'top-down' manipulation: Knight, 'Land and Society', pp. 82-3.

[96] Bolivia Nationalized Standart Oil in 1937, a year before Mexico's expropriation. In contrast, the only country to experiment - pretty tentatively- with the agrarian reform in the 1930s was Colombia. The major Latin America land reform came a generation later: in the 1950s in Bolivia, in the 1960s elsewhere.

[97] See the very interesting article by Jan Rus, 'The "Comunidad Revolucionaria Institucional": The Subversion of Native Government in Highland Chiapas', in Gilbert M. Joseph and Daniel Nugent, eds., Everyday Form of Stat Formation (Duke, 1994). pp. 265-300.

[98] Fernando Benítez, Lázaro Cárdenas y la revolución mexicana (3 vols., Mexico 1993, first pubd. 1978), III, pp. 108-10.

[99] D. A. Brading, The First America (Cambridge, 1991), PP. 307, 362-3, 389, 458-9.

[100] Knight, 'Racism, Revolution and Indigenismo', p. 79 and p. 105 n. 39 for further references.

[101] Ibid., pp. 85-6 on the cult of Mestizo; see also Agustín Basave Benítez, Méxicomestizo(Mexico, 1992).

[102] Rus, 'The "Comunidad Revolucionaria Institucional".'

[103] E.g., the indigenismo of Peru's President Leguía in the 1920s.

[104] This became particulary evident with the redefinition of official political philosophy in the 1940s, when revolutionary nacionalism shed its Cardenista association with radical social reform, anti-fascism and popular frontism, and became discursively aligned with private enterprise and the Cold War: Tzvi Medin, 'La mexicanidad política y filosófica en el sexenio de Miguel Alemán. 1946-1952', Estudios Interdisciplinarios de América Latina y el Caribe, 1/1 (enero-junio 1990), pp. 5-22.

[105] Quoted by MacKenzie, 'Introduction: The Arena of Ideology', p.1.