ported with greater chronological detail than, for instance, minor shifts in policy stances. Also, there simply tended to be more variation across time for indicators of governmental status than for indicators in other clusters. It was not uncommon for the outlawed parties of the first half to emerge as the ruling parties of the second. By limiting our scoring of parties to a simple dichotomy of first half and second half for the governmental status indicators, we inevitably iron out some roller-coaster patterns in party fortunes that deserve to be treated with more finesse. But our inability to support more than a dichotomization of coding for the other clusters imposes this simple dichotomy as the least common denominator for relating governmental status to the other concepts at the data analysis stage.
The precise "halves" of our time periods for the 53 countries in the study are given in Figure 2.1. The division points for certain countries were adjusted to take account of political events of extreme importance, like military coups or alterations in a country's constitutional structure. In the absence of an "obvious" dividing line caused by some key event in a country's political history, the dividing line was drawn arbitrarily between 1956 and 1957. In this and subsequent chapters, the coding results are presented separately for the "first" and "second" halves of our period, with the number of parties in each analysis somewhat smaller than the 158 that provide the base for the institutionalization indicators. Only 135 parties were operating in our 53 countries during the first half of our period and 147 during the second half. A total of 124 parties were active in both halves.
Dividing our time period into halves presented a special problem for former colonies that achieved their independence during or at the end of our time period. For many of these countries, the struggle for independence extended throughout our thirteen-year time span. Especially in Africa, the common practice of both France and Britain as colonial powers was to grant increasing powers of internal government to the various colonies in response to demands--albeit more readily in the case of France than Britain. Elections of Africans to legislative councils or territorial assemblies sometimes came rather late in our time period, but somewhat earlier in French than British colonies. Depending on the timing of grants of governmental responsibility or the calling of elections before independence, the researcher encountered several possible dates within a given country to divide the time period. His choice of a dividing date had implications for establishing governmental status of the preindependence parties, especially in scoring parties on specific variables in this cluster, namely, 2.02, "governmental leadership"; 2.03, "cabinet participation"; and 2.05, "legislative strength."
To help establish a strategy for drawing dividing lines between the first and second halves of our time period for these emerging nations, we might again consider the value of simultaneous or near simultaneous cross-national comparison, a subject broached in Chapter 2. There it is argued that comparatively little is gained from adherence to a fixed time period to compare party activity in Turkey with party activity during exactly the same period in Cuba. These country examples were deliberately chosen in the absence of any established theoretical interest for making precise time comparisons of internal party activity between Turkey and Cuba. We might tolerate a considerable amount of slippage in matching time periods among countries to support such gross cross-national comparisons. All we might ask is that the activity occur roughly in the same decade around the middle of the twentieth century.
But, in moving away from gross cross-national comparisons to the comparative analysis of party activity in nations within a given cultural geographic region of the world, for example, Africa, we may properly insist on a better fit between time periods in our analysis. For instance, if one wanted to compare the governmental statuses of parties during the preindependence period in countries that achieved independence nearly together, he would want similar time periods for comparison and certainly not want the parties' governmental status scores to depend heavily on the particular dividing lines chosen by the researcher on an ad hoc basis for each country.
One approach to establishing general comparability in time period subdivisions across ex-colonial countries, and the approach used in the ICPP project, is to view the preindependence period as a crude two-stage process. The first stage in this process consists of the native people's struggle to win recognition of their right to participate in their own government; the second stage consists of the granting of some degree of internal self-government, perhaps in increasing amounts, with the effect, if not exactly the intent, of preparing the country for full independence. Insofar as is possible, we attempt to capture this two-stage process in our establishment of time periods and divisions therein.
Not all the ex-colonial countries in our study conform to this process, of course. In Ghana, for example, the first stage occurred before our time period began, and the Constitution of 1951 is taken to fix the beginning of the second stage, which ended conveniently in 1956, followed by independence granted in 1957. But, for most of the former French colonies, the process fits rather well for our purposes, with the Overseas Reform Act (or loi cadre) passed in 1956 marking the end of the first stage and the implementation of that law thereafter providing for the second stage. In some cases, autonomy if