There are two ways to get an idea of the number of children born within a certain group of the population:
first: a regular registration of all these births;
second: through computation of other data obtained in a reliable way.
Data obtained from registration:
The figure resulting from birth-registration is just as reliable as this registration itself. In countries of rapid social change the registration of births and deaths is mostly unreliable, and the figures obtained through it must first be investigated to find out whether they are complete.
In Java a first start was made with the registration of the inhabitants at the time of the governorship of Raffles (1811-1816). The figures from that period relate chiefly to the number of inhabitants. The oldest figures data from after 1815. After 1816 when the British handed over to the Dutch government this registration was continued. In 1830 it was restricted for financial reasons; this lasted until 1850. These old figures were published by Bleeker, and some of them for 1845 were referred to in chapter 3.
The general idea was the registration of all the villagers liable to unpaid services, and to acquire a basis for levying taxes. These intentions and the fact that things were decreed by foreign authorities, were none of them factors to further completeness of data. This became apparent in a checking-up examination made in 1852 in the areas of Tjirebon and Banjumas. The desa-registers there were far from complete and not nearly up to date. In 1864 a new arrangement for the regular registration of inhabitants of Java was introduced and given up again in 1879, as the results were unsatisfactory. In 1880 a new method was introduced again by the then Home Department which apart form some alterations is still used in Java and Madura to-day and which method was also applied after independence had been acquired in Sumatera, Banka, North-Sulawesi, Ambon, Kalimantan and Lombok.
The method of registration in which births and deaths were reported orally to a village official appeared to be unsatisfactory. Besides, this method had as its drawback that the lack of any certificate made every control impossible, for instance of the reports sent to the medical officer.
On the instigation of J.L.Hydrick, active in Indonesia form 1924-1939 in the development of his public health programme in Central-Java, a test area was chosen viz. The kabupaten Banjumas. In judging about general health conditions the disposal of figures relating to births and deaths was a minimum requirement, and much attention was there fore paid to their registration.
A new system was devised and introduced here. Under this system on certificate for the family concerned, one for the statistics department of the kabupaten office, and one for the desa archives was drawn up, so that control later on was possible. The data about the number of births obtained after the introduction of the new system of registration and after the trial execution of Hydrick’s public health programme are the only trust-worthy pro-war figures as far as the islands of Java is concerned. That is to say they are the most reliable or least unreliable as far as completeness is concerned. After the introduction of this new system in the dabupaten Banjumas in 1933 it was in 1935 introduced in the whole residency of Banjumas. The area consisted of 5 Kabupatens in 1935 among which also kabupaten Banjumas. By an administrative alteration this was later limited to four. The residency of Banjumas covers about one sixth of the province of Central-Java. In the 1930-census the population of kabupaten Banjumas was 396,878. In 1930 the residency of Banjumas had a population of 2,474,447 (inclusive of the kabupaten Purwokerto). In the 1930-census 41½% of the population of the residency of Banjumas fell into the first two age-groups. The increase in the number of registedred births, after the introduction of the new system, appears very clearly form the stave-diagrams taken form van Os’s treatise. See fig. 11 and 12.
The birth-rate went up from 25-30% to 41-43% in only a few years time. These figures for the residency of Banjumas were arrived at via a figure for the total population which was not quite accurate. It was probably estimated too low. If it is assumed to be 10% too low, the birth-rate for 1937 is 38-39% and obviously higher than the rate for 1931-1935, which amounted to 29-30%. A rise which was almost certainly caused by an improved registration as a result of the better system, in which the possibility of control will have had a stimulating influence. Cooperation on the part of the population and the stimulating effect of the presence of the public-health workers active in this test-area, who supervised the registration of births and deaths, together with the goodwill of the lurahs desa, will certainly have led to this improvement. Before the second world war this method of registration had already been introduced in the whole province of Central-Java, in the D.I. Jogjakarta, and besides in some smaller regions of the provinces West- and East-Java. Here, however
Birth-rate of the population Birth-rate of the population
of the kabupaten Banjumas. Of the Residency of Banjumas
in 1931 – 1937. in 1931 - 1937.
Fig. 11. Fig. 12.
the data obtained were not so convincing as in Banjumas. Of some ketjamatana of the D.I. Jogjakarta some data about births for the period 1927-1934 are still available. In 1930 the new registration on the proof-system was introduced first in a number of ketjamatans in Jogjakarata and, successively, up till 1936, in practically all of them. In places this happened therefore before the test made in the kabupaten Banjumas in 1933.
In some ketjamatans results were satisfactory. Notifications of births were regularly received, and in checking-up examinations there appeared to be no reasons to believe that notifications were very incomplete. The ketjamatans concerned were Mlati and Kasihan bordering on the city of Jogjakarta on the north-west and south-west sides. The data are shown in table IV-1.
Table IV – 1.
Birth-rates of ketjamatan Mlati and Kasihan in the
period 1927-1934
Number of inhabitants of Mlati according to census 8-10-1930: 24,883.
Number of inhabitants of Kasihan according to census 8-10-1930: 24,218.
The population-figure the birth-rate was based on was a little too low, but in the 1930-census it appeared that the administrative data came up very nearly to those of the census itself. Data about births in the whole of the D.I. Jogjakarta for the period 1913-1936 are shown in table IV – 2.
Table IV – 2.
Birth-rates of the population of D.I. Jogjakarta
in the period 1913-1936.
The birth-rates based on the probable number of inhabitants according to the census of 1920 and to that of 1930. It is almost certain that registration was incomplete and particularly in the city of Jogjakarta and in the ketjamatans farthest from it. The best registration took place in the ketjamatans surrounding the town, of which Mlati and Kasihan were instanced before. In the city of Jogjakarta the birth-rates for the period 1920-1936 were 15.1% - 29.4%. In the 1930-census administrative data appeared to differ considerably from the census ones, so that there was every reason to believe there had been no proper notification of births.
In the days of the Japanese occupation (1942-1945) and of the revolution registration was seriously disturbed. Not until independence had been acquired, and peace and quiet had returned did it function better again.
Since 30-June 1949 a new civil registration ordinance came into force in the D.I. Jogjakarta. Every inhabitant was in the period just after 30 June 1949 obliged to report himself for a new registration, ad far as possible with the complete family at the same time. Negligence in the registration of birth and death in the family within three days after the occurrence could be punished with 15 days detention or a fine with a maximum of Rp 2,500.-. It was originally the intention to control this registration by sample taken at random in the different areas inside and outside the town of Jogjakarta. This registration of all the inhabitants in the month July and August 1949 was meant as an improvised census. The result should be used as a starting point for the civil registration. The circumstances during the revolution made the realisation impossible. This new registration was a new start of the registration of the population in the D.I. Jogjakarta, but on the opening of the conference at Malang (11-5-1959) for the officials of the vital statistics offices Mr R. Sudarjono declared that the vital statistics of the provinces of Java and the D.I. Jogjakarta were after ten years still in a stage of the first development and orientation. Mr r. Sudarjono was the spokesman of the Ministry of Health.
In the D.I. Jogjakarta a special circumstance concerning the registration of births occurred similar to that which happened in Banjumas in 1933. In 1952 two of the three kelurahans of the ketjamatan Depok were selected for a demonstration- and test-area for mothe- and child-health work under the supervision of Dr Mardjaban. In 1954 the third kelurahan was also included in this test-area, so that this comprised the whole of the ketjamatan Depok. The attention of M.C.H. Workers being concentrated mainly on pregnant women, infants and toddlers, great pains were taken to collect data about births, Birth-rates of the ketjamatan Depok for 1950-1956 collected in a stave-diagram result in the following fig. 13.
Birth-rate of the Ketjamatan
Depok 1950 – 1957
(D.I. Jogjakarta)
1952: Two kelurahans became test-area
for the M.C.H-work
1954: Also the third kelurahan was in-
corporated in the M.C.H.-test-area.
Fig. 13.
In the period 1953-1957 the ketjamatan Depok appeared to have the highest birth-rate among the ketjamatans of Sleman.
The birth-figures registered in the D.I. Jogjakarta for the years 1952–1956 were collected kabupaten-wise in table IV-3.
Table IV – 3.
Summary of the birth-rate of the kabupatens of the D.I. Jogakarta
and the town of Jogjakarta during the period 1952 – 1956.
As a basis for computation served the mid-year population, the figures for which were supplied us by the chief Civil Registration office D.I. Jogjakarta. These figures were composed from statements by the lurah desa, and cannot be looked upon as entirely complete as long as we are not convinced of a full registration of births and deaths. The theoretical minimum figure we find in Wertheim’s article about the 40% test may serve as a measure to judge on the rates found. This figure was computed in the following manner:
In a population 40% of which is 14 years old or under 14, as is very likely in most |Asiatic countries, all fifteen-year groups would consist of about an equal number of persons, if no deaths occurred among in fats, toddlers, and (pre)-school children. The population-pyramid would then assume the shape of a shell-case. Then the size of all the year-groups 0-14 would be equal that is 40% : 15 = 2.66% of 26.6%,.
The group of infants would in this theoretical case at the same time be the measure of the number of births As a high death-rate among infants in many countries in rapid social change, all statements of a number of births of 27% and under are highly improbable. The real percentage must be much higher.
When approximating the theoretical maximum birth-rate in a particular population we arrive at 60% for a population with a texture like the population of the D.I. Jogjakarta.
This figure was calculated as follows:
The number of women of 15-45 years of age comes from the data about the population of Minahassa, Sangir, Talaud, and Ambon, This group was 20-24% of the total of those populations. In our calculation we used the highest percentage: 24. The adult women of the population in the D.I. Jogjakarta that were married comprised 67% in 1930. As this percentage relates to the group of adult of 15 and older, the percentage of married women of 15-45 years of age will be higher and we estimate 75%. this leads to the assumption that 18% of the population consists of married women in the period of fertility of 15-45 years of age. When every woman has some ten children in this period of 30 years, this corresponds to a birth-rate of 60%
As the number of barren women was left out, and as an average of the children per married women in the period fertility is very high, the real birth-rate will obviously be lower than 60%. We return to the registered birth-rates of the 4 kabupatens and the town of Jogjakarta in the period 1952-56.
In 1952 we find in 30 of the 60 ketjamatans of the D.I. Jogjakarta a birth-rate under 30%. the kabupaten with the lowest figures of all is Kulon Progo, where 75% of all the figures are undr 30%. In Gunumg Kidul 69% of all the figures are under 30%, in Sleman only 17%. In 1954 all 60 Ketjamatans have higher birth-rates than in 1952 which it would seem to me is a result of an improvement in the notification of births. In this year rates lower than 30% are found in only 3 ketjamatans, and in 21 they are 40% and over.
The ketjamatan Sleman has the best that is the least incomplete birth-registration. In 1954 a number of 9 out of the 17 ketjamatans of Sleman have a birth-rate of 40% and over. The years 1955 and 1956 generally show lower figures. In 1956 the kabupaten Sleman, too appears to be the least incomplete.
None of the four kabupatens as a whole ever reaches a birth-rate of 40%, in the period 1952-1956. When we consider the birth-rates of other parts of Java, we find for the period 1952-1956 a number of kabupatens having higher birth-rates, which range from 40-50% especially in the province of Central-Java, Only in 1954 did three kabupatens of the province of East-Java out of a total of twenty-nine together with five towns reach a birth-rate of over 40%. The birth-rates of a dozen kabupatens in the province of Central-Java are shown in table IV-4.
Table IV – 4.
Summary of the birth-rates of some kabupaten of
Central-Java during the period 1952 – 1956
In 1952 only one kabupaten out of the 28 kabupatens and 4 towns of the province of Central-Java had a birth-rate above 40%. (Wonoboso). In 1853 , 1954 , 1955 and 1956 this was the case in 10 , 17 , 17 and 4 kabupaten respectively. All the kabupaten which ever reached a birth-rate of 40% or higher in the period 1952-1956 are compiled in table IV – 4.
Also the kabupaten North-Sulawesi reached birth-rates of 40.4% and 40.1% in 1954 and 1955.
Of other parts of Indonesia only a few prewar data are available. Those data which were collected in the exhaustive investigations made by the government physician of Menado (North-Sulawesi) Dr A. Kϋndig, are the most important.
From 1930 to 1932 he collected data about the number of births in some five ketjamatans. Dr Kϋndig had his won staff, and the data which are very reliable are shown in table IV-5.
Table IV - 5.
The birth-rates of five ketjamatans of the Minahassa-region
in the period 1930 – 1932 (Dr A. Kϋndig).
As the data of the 1930-census are available, those figures are based on a very reliable total number of inhabitants. Nearly all these birth-rates are above 40%.
Bosma in his article about deaths among infants in Medan mentioned that as far as he knew, the only reliable birth-figures for Sumatera are those of the Kajoeagoeng region near Palembang, which date form before 1930, where a birth-rate of 39% was found at the time.
A figure we meet several times in the literature about the birth-rate in Java is professor J. van Gelderen’s estimate of 38%. More recent data were obtained by means of a 25% sample-survey in the Senen quarter in Djakarta in July 1956. in it the number of births in the twelve month preceding the survey was investigated. For the Indonesian population group a birth-rate of 45.2% was found. According to the ordinary registration the birth-rate was 32.3% in 1955. In the general health survey made in Tanah Tinggi (Djakarta) in March 1957, the birth-rate of the Indonesian population in this urban region appeared 38.6%, concerning the year preceding this survey. According to the civil registration the birth-rate was 32.4% in that same period (12-3-’56 till 12-3-’57).
In the Tjikarang 100% survey made in the villages Simpanganand Waluja in West-Java in March 1957, a birth-rate of 46% was found, while according to the official registration it varied in 1953-1956 from 13 to 19%. In the survey in the village Andir (July ‘56) a birth-rate of 50% was obtained. According to the civil registration of this areas the registered birth-rate was 32%.
Another possibility to establish the approximate birth-rate is the substitution-method.
If we dispose of reliable observations made in countries with:
a about the same texture of population,
b about the same density of population in agricultural regions,
c an equal level of education,
d and equal general level of prosperity,
e an equal method of production backed by little capital,
f an equal attachment to the traditions of the desa,
g a great increase of population,
then in using this “borrowed” figure, the margin of mistakes in computations will be smaller than that occasioned by using unreliable figures for the area under consideration.
When drawing up the life-tables of India for 1911-1921 and 1931-1941, Kingsley Davis used the gradation of ages of people of seventy and over as found in Chili in 1930, because these figures were not available of India. To compute the reproduction-rates of the Indian population he used the age-specific fertility rates of the Chilean population for 1930-1931 (appendix F. Population of India and Pakistan). Countries comparable to Indonesia we find only in South-East Asia; and about other Asiatic countries we have very few data.
Of Malaya Smith published the figures of the former Federal Malayan States.
Table IV – 6.
Crude birth-rate for the states Perak, Selangor, Negri Sembilan
and Pahang of the year: 1921-23; 1931-33; 1946-48.
The figures are from a regular birth-registration in these countries with the exception of the figure for 1946. Smith computed this by means of the data of the census of September 1947. The figures for the last three years clearly have a higher average than the previous ones. This increase is also due to an improved registration. In 1947 the population had a high percentage of literates, particularly in Selangor, Perak, and in Negeri Sembilan. In Pahang it was smaller. All the figures, except that for women of 25-29, are above the average for all Malaya.
The men in the Malayan population-group in the above states are chiefly employed in rice-growing and further in the rubber plantations.
India which is also an agricultural country might do for a possible comparison. There has been birth-registration in India since 1864. The figures published about these areas where the greater part of the population consists of illiterates, did not convince us by their completeness. The basis for the computation of the birth-rate appeared to be the number of inhabitants stated in the last census. From 1872 a census was taken every decade ending in 1, viz.: 1881, 1891, 1901, 1911, 1921, 1931, 1941, 1951. So the birth-rates were based on the same figure for ten years. The highest figure was 39.6% in 1914.
From 1961 – 1940 the average for every five-year period appeared to vary from 33.5% to 34.7%. In the war years 1640-1945 the birth-rate arrived at was 28.3%, which is very low. So the registration figures of India cannot be of much use to us. Kingsley Davis computed the more acceptable figures for the size of the population of India in all the years 1911-1945, and by means of these the birth-rates of these years. The figures of Davis are given in table IV-9.
The registration birth-figures of some Asian and African countries are collected in table IV-7. Nearly all these figures are above 40%. Also the birth-rate of the Latin-American countries are nearly all above 40%.
Table IV – 7
Summary of birth-rate figures of some Asian and African
countries in the period 1950 – 1954
1) A number of towns only: in 1951 en 1953: 60 towns, in 1952: 62 towns,
having a population of approximately 1.3 million.
2) For the both cities the birth-rate concerns the present of de facto population.
3) Excluding children born alive, but dead before registration of their birth.
Data acquired by computation
Another way to approximate the number of births is that of computation from other reliable figures about a population.
For Indonesia this was done by N.Keyfitz. His starting point was the number of children not yet able to walk counted in the 1930-census.
In Java and also in the D.I. Jogjakarta this group of children appeared to make up over 5% of the total population. The central census office in 1930 gave to understand that according to the experts consulted, the ages of these children ranged from 0 to 18 months. When two thirds of this group is taken for younger than one year, this results for the D.I. Jogjakarta in 2/3 x 5.4% = 3.6% i.e. Infants per 1,000 members of the total population.
Infants of this group are the survivors of all those born in the year preceding the census. When we consider the death-rate among infants is great and suppose it to be 20%, the birth-rate should be much greater viz. (100 : 80) x 36% = 45%
Wertheim had this criticism of Keyfitz’s method of computation:
- Two thirds of the group of children is too rough a factor to be used. It is only correct either when death is equable under 18 months of age, or in the theoretical case that the moment the oldest child of the group concerned becomes 1½ year old, 20% of all of them are killed, if chances to have this happen to them are equal for all children under 18 months. Because the infant mortality rate id higher than the mortality in the second year of life, the number of infants will be greater than 2/3 of the age-group 0-18 months.
- It is not correct to apply the whole death-rate to infants of this group. In Jogjakarta for instance about 75%of the infant mortality occurs in the first six months of life, so that three fourths of the deaths at the utmost must be added to this group to approach the birth-figure.
The remaining one fourth applies to children born longer than one year before the census, and were therefore not taken into account. The first inaccuracy resulted in too low an estimate of the number of births, while the second partly compensated the first. Professor van Wijngaarden of the Mathematical Centre, Amsterdam, who was consulted by Wertheim, set up a more comprehensive calculation for a more accurate approach of the number of births in 1929, the year previous to that of the census (printed in Appendix B. pp. 184-187 of Ekonomi dan Keuangan Indonesia, March 1955).
On the basis of these computations we calculated the birth-rate of the population of the D.I. Jogjakarta for the period October 1929- October 1930. As the death-rate among infants approximated that of Djakarta for 1935-1936, the figures of table IV-8 will come nearest the truth for the D.I. Jogjakarta.
The figures are slightly lower than those Keifitz computed. The factor “under enumeration of children” was probably very small according to the census office. This in contrast to the counts in India where the groups of children of 0-4 years old age were far too small in practically every census.
Table IV – 8.
Birth-rate of the Indonesian population in the D.I. Jogjakarta during
October 1929 – October 1930, computed from the census 1930
figures according the formulae of van Wijngaarden.
Kingsley Davis found only incomplete birth-registration figures in India. A sort of trial-area like the province of Central-Java before the second world-war India never had. To approximate the birth-rates, he used as a starting point the number of children of 0-9 years of age counted in the census according to the improved age-gradation scale for India.
The year-group 0-4 years being to small on account of “under enumeration”, and the gradation of some year groups being irregular, he wanted a correction.
Corrected life-tables were drawn up for age-groups of then years for nearly all the decades a census was taken (1871-1941). The two lacking life-tables were drawn up by K. Davies. On the basis of the census and the life-tables he calculated the average birth-rate for every decade (see K. Davies appendix C. p. 238 and E. p.244).
Table IV – 9
Computed birth-rate of the Indian population
during the decades 1881-91 till 1931-41
according to K. Davis
K. Davies expected there would be a slight fall in the high birth-rate of the Indian population.
A third way to approximate the number of births per year was devised by Liem Tjay Tie and de Haas in their paper about infant-mortality in the Tasikamalaya area in 1939. They computed the number of births that had died before vaccination. Vaccination was carried out by the vaccination controller when the babies were 3-4 months old. They arrived at a birth-rate of 39%, which figure must be regarded as a minimum, as it is not know whether there was a post-mortem examination of all the infants that had died, or whether all the survivors had indeed been vaccinated.
This method of computation had already been used by de Haas to calculate the number of births in the Karo countries.
The data about the number of vaccinations in the D.I.Jogjakarta were not available, so that it was not possible for us to approximated the number of births in this way.
A fourth method to approximate the birth-rate was described by Wertheim. Starting out from the fact that in most countries in rapid social change about 40% of the population is younger than 15, he calculated the correlation with the birth-rate. Using his three tables, based respectively on a infant death-rate of 6%, 15% and 25%, the minimum birth-rate to occasion a natural annual growth of 2% in a population 40% of which was younger than 15, appeared to be the following:
For an infant death-rate of 6% and a corresponding death-rate of preschool children and school going children a birth-rate of 34% is required.
For an infant mortality of 15% and a corresponding death-rate of preschool children and school going children a birth-rate of well over 39% is required.
For an infant mortality of 25% and a corresponding death-rate of preschool children and school going children a birth-rate of over 48% is required.
The number of births among a particular population depends on a number of factors which are for a large part rather constant.
The data about the population of the D.I. Jogjakarta of the 1930 – census supply these most important factors for us to follow out:
- The age at marriage was very low as compared to European countries, although slightly higher than in West-Java.
- In the adult group there was a considerable surplus of women. The ratio between the sexes in the D.I. Jogjakarta was such that after the age of 15 1158.4 women were counted for every 1000 men.
- In the group of adult women 677 out of every 1000 were married. This figure was the highest as compared to the other provinces of Java, except for West-Java. There 711 adult women were married out of every 1000.
- Polygamy appeared to occur to some extent in the D.I. Jogjakarta in 1930. Of the married men 2.3% had more than one wife.
- A great number of children are positively appreciated at birth. Every child is supposed to bring its own happiness. (Saben anak nggawa bedja dewé – dewé.)
- About the possible presence of taboos in sexual intercourse nothing is known.
- Of the most important factor: an effective conscious birth-control, which so much decreased the birth-rate in many European countries after the Industrial Revolution, little is known in the D.I. Jogjakarta, but the occurrence of it seems probable in view of the data published by Verdoorn and Stratz about birth-control among the Javanese population. That the methods used (placing the uterus in retro flexion, carried out by the dukun, native “obstetrician”) and the measures taken will influence the birth-rate seems much less probable.
These factors, originated from the adat and age old customs of the Javanese population seems to be rather constant. The birth-rate of the population in the D.I. Jogjakarta is very probably constant high, Just as it appeared in India during 1881-1941 according to the results of Davis’ examination.
We considered:
- that the most reliable birth-figures of Central-Java, both of before and after the second world-war, showed values of 40% and above per annum,
- that the most reliable birth-figures of the ketjamatans of the D.I. Jogjakarta also showed values of 40% and above per annum.
- that the birth-figures found via two methods of computation also showed values of 40% and above per annum.
- that these birth-figures were about the same as those for comparable countries.
Therefore we assume the birth-rate for the D.I. Jogjakarta to have a value of at least 40% For the computation of the infant death-rate we shall use this figure in chapter 6.