180 BUREAU OF AMERICAN ETHNOLOGY [Bull. 197 



Two arguments can be adduced in support of this contention. In 

 the first place, the basic mortality rates from which the several United 

 Nations model life tables were developed are heavily weighted by 

 the age-specific mortality levels reported among European countries 

 since 1920. One can certainly question the degree to which these 

 largely European rates would pertain to the members of a population 

 such as the Navaho, vrhose entire mode and condition of life are so 

 differentia 



Secondly, the selection of the most appropriate model or group of 

 models to represent a specific population at a particular time in its 

 development is confronted with great difficulties, when we lack reliable 

 information on precisely those values which are needed in guiding our 

 selection. The 80 life table models which were originally developed 

 by the United Nations staff purport to cover, in their wide range of 

 mortality levels, the known or expected conditions to be found among 

 human populations anywhere. However, the selection of the single 

 most appropriate model for a specific population is contingent upon a 

 fairly precise knowledge of the infant or early childhood mortality 

 experienced by that population. Unfortunately, in most situations 

 where resort to such models would be helpful, the level of infant or 

 child mortality is not precisely ascertainable. Conversely, where the 

 level of infant mortality is known with a fair degree of accuracy, 

 related information concerning mortality at other ages is likely to be 

 sufficiently reliable as to make it unnecessary to resort to the models.-" 



CONCLUSION 



The major purpose of this study was to describe and analyze data 

 on selected demographic characteristics of the Navaho population, 

 in order to indicate the nature and limitations of the two major 

 sources of these data ; the Bureau of Indian Affairs and the Bureau of 

 the Census. The main conclusion which has been reached on the basis 

 of this investigation is that most of the data provided by these sources, 

 at least until the recent past, are of insufficient quality to support 

 precise inferences with respect to the basic demographic characteristics 

 of the Navaho population. This conclusion is supported by the many 

 vagaries and inconsistencies contained in these data, as revealed in 

 the foregoing analysis. However, the results of this investigation do 

 not appear to be entirely negative. On the basis of the available data, 



^3 For example, the apparent "undercount" of adult Navahos, as reflected in the low 

 median ages for most of the Navaho age distributions, may in fact be an accurate indication 

 of higher levels of mortality among young adult Navahos of both sexes. Their largely 

 European counterparts, whose mortality levels are expressed in the model life table values, 

 may simply have experienced lower mortality (see Carrier, 1958, p. 155). 



^"For a general criticism of the United Nations model life table scheme, together with a 

 brief exposition of an alternative method, consult Gabriel and Ronen, 1958. 



