Alba Regia. Annales Musei Stephani Regis. – Alba Regia. Az István Király Múzeum Évkönyve. 10. 1969 – Szent István Király Múzeum közleményei: C sorozat (1969)
Tanulmányok – Abhandlungen - Éry Kinga, K.: Investigations on the Demographic Source Value of Tomb Stones Originating from the Roman Period. – Vizsgálatok a római kori sírfeliratok demográfiai forrásértékéről. X, 1969. p. 51–67.
W i 11 с о x 6 very definitely pointed out already in 1937 that the sex disparity is a consequence of exclusively the custom of tomb inscriptions. Russel's opinion lies between the former two, that is, he partly admits that there might have been a certain bias to the advantage of the males in the commemorating practice, but he also holds it probable that the greater number of the male inscriptions may reflect actual population conditions owing to diverse social factors 7 . However, Rüssel found a sex ratio of merely about an average 130 in the commemorative material originating from the Empire period examined by him; whereas 5 z i 1 á g y i' s material, of greater numbers and treated per settlements, displays more extreme values. According to the data of Table 3, the lowest sex ratio occurs in Africa and the Noricum, though it is still between 123 and 134 even in these territories. The largest disparity appears in the material of the first and second centuries originating from Germania, with more than 1000 male inscriptions for every 100 female ones, that is, almost only males had been thus commemorated during the early period of this province. With regard to the provincial averages, the sex ratio, is generally higher in the early period, than in the later one. It is only in North Africa that some slight increase can be observed with the pass of time, and a more significant one in Hispánia, Britannia and Raetia. (These two latter ones are smaller series and thus their values less reliable). Even if R u s s e Г s contention be accepted, namely that a sex ratio of about 130 may yet reflect the actual though slightly distorted sex ratio of the population, a sex ratio higher — and frequently considerably higher — than 130 are found for 73 per cent of the published series of the early period and for 48 per cent of the later one. This fact unequivocally substantiates the statement that the observed disparity on the sex ratio is due exclusively to the custom of making tomb inscriptions. It would also be most instructive to analyse per cities or settlements the causes of the observed disparities, but this is the task of historians. It is probable that in territories where the ratio of the males is very high, the commemorative usage might primarily have been in custom among the military. The example of Germania supports this assumption, and also the fact that the proportion of the male inscriptions decreases in the later period. This latter phenomenon may probably explained by the fact that the custom of commemoration by inscriptions spread at that time to wider strata of the society. * In the followings, let us attempt the examination of mortality according to age groups. For a start, I have reducet the combined mortality data of the several provinces, and the individual ones of the towns or settlements, into age groups per 10 years for both sexes. The advantage of reducing to ten years instead of five, the more frequent usage in demography, lies in the present case in the fact that by this the disparity per individual case, observable in ages ending in 0 and 5 (owing to 6 W. F. WILLCOX, o. c.,p. II. 17. 'J. С RÜSSEL, o.e., pp. 13-14,22. a hazy knowledge of the actual age of a person at that time), will thus disappear 8 . Similarly, I have reduced into one age group all those above 80 years, in order to equalize the tendency of deliberately or involuntarily rouding up the age figures of elderly people. Subsequently, I have calculated the percentage of dead in the given age group (d x value), and then plotted it on graphs 9 . To make an objective study and interpretation of the distribution per age group possible, I have also entered by a thicker line onto the graph of every series the age group distribution of the model life-tables as published be the UNO, so that there was plotted on the graph the one life expectancy modell distribution which was similar to the life expectancy (actually the average age at death) calculated by Szilágyi. The model life-tables have been constructed by the demographers of the UNO from 158 life-tables made between 1900 and 1950 and of 50 countries, with due attention to the requirement that the basic data derive from highly divers areas, populations, and countries of different economical development 10 . These models do not claim, therefore, absolute validity — nor do they subserve this purpose. They completely afford, however, the drawing of pictures on the laws concerning the specific age frequency of deaths in the case of divers life expectancy values. The question may still arise as to what extent we might expect the alignment of distributions calculated from historical source data and those constructed from recent populations. For a control, mortality distribution data, calculated from skeletal series, are available. Figure 1 shows mortality distributions calculated on the basis of the skeletal remains of two populations deriving from two different historical periods in the Central Danubian Basin, with the plotting of also the distribution of the model life tables corresponding to the life expectancy values. The data of the two series are statistically not equivalent, since the group from the Roman Period (III — IV c.) is represented by merely 172 skeletal findings of two smaller cemeteries (Keszthely — Dobogó 11 and Majs 12 ), whereas the group from the Arpadian Age (X — XII c.) comprises several hundreds of skeletal remains of a number of cemeteries 13 . In the former case a more disproportionate and in the latter case a more equalized age group distribution will obviously result, as demonstrated also by the lines of the model distributions plotted on the graphs. But beyond that, the Figure also shows that even though the age group distributions of the skeletal remains fail to concur entirely with the model distributions, they agree in the trend of distribution. It is well known that the order of magnitude of the figure representing the life expectancy of a population is primarily and fundamentally influenced by the rate of infantile and juvenile mortality, which again depends on the development of the public hygiene conditions of the given popu» A. MÓCSY, AntTan, 10, 1963, pp. 201-211; ID., Acta Ant., Hung., 14, 1966, pp. 387-421. 9 The broken, perpendicular lines introduced in the Figures merely serve to emphasize the distribution trends within the age groups Infans-Juvenis, Adultus, Maturus and Senilis. 10 Age and Sex Patterns of Mortality. Model Life-Tables for Under-developed Countries, Population Studies, No. 22, 1955, United Nations, New York. 11 I am indebted to Gy. Acsádi and J. Nemeskéri for the unpublished demographic data relating to the series of Keszthely —Dobogó. 12 К. К. ÉRY, Anthrop. Hung., 8, 1968, pp. 31-58. 13 GY. ACSÁDI, Tört. Stat. Évkönyv, 1963-64 (1965), pp. 21 -22, 53
