Until recently Colombia has been a mainly rural and agricultural country. At the turn of the century probably 65% to 75% of the labor force was in agriculture; by 1951 this share had fallen to about 55% and in 1975 to about 35%. The share of population in rural areas was a little higher. In 1900 perhaps 50% to 55% of output came from agriculture; though it remains the main producing sector, its share had fallen to 25% by 1974.

Colombia has always had a highly unequal distribution of income, the major historical cause being an extremely unequal distribution of land. Over the years economic policy could be fairly described as “urban biased,” so that the group to whose welfare public policy has contributed the least are the landless agricultural workers, including a good number of migrant workers.

History is written by the winners,” wrote Orwell (1970: 110); he went on to suggest that a totalitarian government “claims to control the past as well as the future.” Yet such a concern with controlling the past is not a distinguishing feature of totalitarianism; many examples of an ideological or legendary version of the past receiving government sanction can be found in nontotalitarian countries. While few governments or political parties can afford to ignore the past completely, it is undoubtedly the case that in some countries a “correct” interpretation of the past is of more pressing concern to contemporary politicians than it is in others. Since the early days of independence, Haitians have shown an absorbing interest in their past—not a dead “historical” past, but a living ideological past. The position which a Haitian has taken with respect to the past has been closely connected to his political commitments in the present, and rival parties have elaborated competing legends which function as legitimations of sheir current policies and interests.

The analysis of voting behavior patterns from electoral statistics aggregated by regions and urban and rural districts has a long and rich tradition (Lipset and Rokkan, 1967). This mode of electoral analysis has, however, declined in usage due to the emergence of the sample survey method, which allows for a more detailed analysis of individual voting choices and the related social and attitudinal factors which influence them.

It is often forgotten, however, that aggregate analysis of area or geographically defined voting statistics is a low-cost research method when compared to the exclusive and much more expensive sample survey research method. Aggregate analysis can be applied effectively with careful adaptation to most of the areas of electoral analysis, particularly so in societies in which community and residential patterns are structured according to clear socioeconomic and stratification variables such as income, status, and occupation. This applies as well to societies in which regional and cerritorial divisions are more than just administrative units but represent focal points of local identities and loyalties.

In 1929 the British scholar, Cecil Jane, published his classic, Liberty and Despotism in Spanish America. In it he wrote:

Instead, Jane (1966: 79) argued that the War of Independence was an attempt to realize the most deeply felt ideals, “which were derived not from any external sources, but from the very hearts of the people.” Similarly he argued that the political conceptions of Spanish Americans were (and continued to be) Spanish, not Anglo-Saxon, and he strongly suggested that the example of the British North Americans exerted little if any influence on the Spanish Americans of the nineteenth century. Jane (1966: 111-112, 168-172) also attacked with thinly veiled contempt the idea that the salvation of Latin America lay in emulating England or the United States.