Statistics of Religions
Concerns itself with religious bodies, the number of their members, and their distribution over various countries
Statistics of Religions.—I. DEFINITION.—This study concerns itself with religious bodies, the number of their members, and their distribution over various countries. In a wider sense the numerical account of the external manifestations of religious life also belongs to the same study, but of late it has been customary to comprise this latter group of facts under the designation of “Ecclesiastical Statistics” and to treat of them separately. As the whole field has only in the last decades been thoroughly worked, language has not as yet afforded a clear distinction between these terms. Practical reasons, however, speak in favor of such a distinction, and therefore we retain it in the present article, and treat ecclesiastical statistics separately (see Ecclesiastical Statistics).
II. HISTORICAL DEVELOPMENT.—The first attempts to determine exactly the number of members of a religious body are found in the records of the Jesuit and Franciscan missionaries of the sixteenth and seventeenth centuries. But they only give the number of the Christians, and not that of the adherents of the indigenous religions in the respective countries. Dating from the eighteenth century, some accounts indeed of the various religious systems and their spread are extant, but they only mention the countries over which the respective religions extended; as to the number of their followers we possess but scattered-data even of that period, and no comprehensive and comparative records. It was only in the nineteenth century that an effort was made to distinguish statistically, according to religion, the entire population of the earth. The accounts given in Table I are the most accurate.
In all these calculations the total of the earth’s population is considerably underrated. The numbers of the non-Christians are evidently only vague estimates without any solid foundation, as is clear from the round numbers and the great differences between the various estimates. Regarding Christians the computation is indeed more accurate, but very far from the exactness requisite for scientific research. Even the attempts made by geographers, such as Hubner, Peterman, Kolb, between 1850 and 1880, do not show any essential progress.
Statistics of religions that should come up to the requirements of science would be possible only if for every country the number of members of the various religious bodies were ascertained from reliable sources, and the totals arrived at from the individual results were tabulated. Average estimates that extend over entire groups of countries without definite indications of the numbers of the population and its distribution with regard to religious denominations are of little use for statistical investigations. Detailed religious statistics, dealing distinctly with all countries of the earth, were for the first time presented by Fournier de Flaix to the second congress of the International Institute of Statistics, held in Paris, in 1889. His example was followed by F. Von Juraschek (1898), H. Zeller, and H. A. Krose, S.J. (1903). The figures given by Fournier de Flaix mostly correspond to the conditions at the beginning or the middle of the eighties; those of Juraschek to the period 1890-97. Zeller has in essentials taken over the statements of Juraschek and made them the basis of his own investigations; he has, however, completed and arranged them more clearly (in Warneck’s “Allgemeine Missionszeitschrift,” 1903), and has added exact references for the various items. The numbers as given by Krose belong to the last decade of the nineteenth century and only in a few cases to 1901. The total results of these four accounts are shown in Table II.
The differences between the first and the last two accounts seem to be considerable. But we must keep in mind that Fournier’s figures refer to a time about ten years previous to that of Juraschek-Zeller; and that the distance in time from Krose’s record is even greater. Within a period like this an increase of from 10 to 15 per cent is by no means extraordinary. Hence, so far as regards the Christians, the statements may easily be made to agree. (The Raskolniks have apparently been counted with the “Greek Orthodox” by Fournier and with “Other Christians” by Juraschek-Zeller.) Neither is the disagreement regarding the Mohammedans and the Brahmins remarkable. The number of the Jews, however, has evidently been underrated by Fournier, and that of the Buddhists overestimated. The latter may easily be accounted for, as in the great Chinese Empire, with its hundreds of millions of inhabitants, Buddhism, Confucianism, and ancestor-worship cannot be sharply separated from one another; they are at times professed and practiced by the same individual. It must be borne in mind, too, that the population of China has hitherto been difficult to estimate precisely—as much so, indeed, as that of the interior of Africa. Regarding the three religions of Eastern Asia, as well as the Fetishism of Africa, the statistical data necessary for a reliable calculation are wanting even now, and therefore fluctuations of the estimates are easily understood. Again, Juraschek-Zeller did not make special categories for Taoism in Japan and ancient cults in India, but added them to the great collective groups just mentioned; and the individuals having no religious denomination seem to have been allotted by Juraschek to other groups on certain principles. Juraschek decidedly underestimated the number of Mohammedans: recent investigations have proved that Mohammedanism is far more widely extended in Africa than was believed. Otherwise the statistical accounts of Juraschek-Zeller and Krose show a far-reaching agreement, considering the different periods of their estimates. Their calculations having been carried out in complete independence of each other, this harmony no doubt confirms the reliability of the results.
III. PRESENT STATUS OF RELIGIOUS BODIES.—The tables of Juraschek-Zeller and Krose given in section II correspond on the whole to the last decade of the nineteenth century. At present, therefore, the first decade of the twentieth century being over, their accounts need complementing and revising. This is especially necessary with the various Christian denominations considering their steady and vigorous increase, while the estimates made ten years ago of the Asiatic and African religions may even now be to a large extent accepted in the absence of more exact computation. The great difficulties of religious statistics have been hinted at above. They are indeed greater than the difficulties of any other branch of statistics bearing on population. Even countries possessing in other respects well-grounded official statistics often lack official accounts regarding religion. The science of statistics has long since come to the conclusion that religion belongs to the essential items of every census. As early as 1872 the Eighth International Congress for Statistics at St. Petersburg expressly emphasized this, pointing out the great importance that must be attributed to a full and clear statement of the individual’s religion, “one of the most essential elements of civilization”. This want is less felt in countries like Belgium, Spain, Portugal, and the majority of the republics of South and Central America, whose populations generally profess one and the same religion, excepting a small minority, whose number can usually be ascertained in other ways with sufficient accuracy. But the difficulty is great with countries of mixed denominations, like Great Britain and the United States, where up to now the distribution of the various religious bodies has not been ascertained by a universal census. In such cases the defect is to some degree remedied by an ecclesiastical census; but this is the case only when all the individuals are counted; and the census is not reliable when only the communicants or those with full right of membership are counted, and a certain ratio is added for the rest, as is commonly done in the Protestant denominations of England and America. The totals arrived at in this way are vague estimates, possessing only approximate value. The same applies to Protestant missionary statistics as far as English and American_ missionary societies are concerned.
Another difficulty in comprehensive statistics of religions lies in the classification of the various creeds. We cannot but combine smaller communities into collective groups. This, however, is a great inconvenience; for thus denominations differing from one another must be connected, and then the large totals produce the impression that one important religion is meant, whereas in fact it is but a combination of a number of religious communities, possessing neither common organization nor identity of belief. In the first place this holds good of that great collective group comprised under the designation of “Protestantism“. This term can, in the statistics of religions, be applied only in the widest and merely negative sense, i.e., as meaning all those Christians who are neither Roman Catholic nor members of a Greek or Oriental schismatical Church. As soon as we try to point out a note proper to this whole group and to it exclusively, we find ourselves at a loss. In the following list, therefore, we have reckoned the group, designated as “Other Christians” in some official statistics, under the heading “Protestants”. On principle, only those are to be counted as Catholics who are in communion with the Church of Rome; it is evident that differences in rite or liturgical language, which do not constitute any diversity of creed, are to be neglected. The self-styled “Old Catholics” do not belong to the Catholic Church, even though the official statistics of some countries reckon them as Catholics; this, however, is of no importance, as their number is insignificant. The designation “Schismatic Greeks” comprises all the Russian or Greek Orthodox, whether they acknowledge the Patriarch of Constantinople as their head or belong to independent Churches. The schismatic Armenians, Syrians, Chaldeans, Copts, and Thomas Christians may be collectively grouped under “Schismatic Orientals”. The Russian Raskolniks, on the contrary, must be regarded as a separate group, distinct from the Christian denominations mentioned above.
With all religious bodies only external membership comes under statistical reckoning. Thereby it is not denied that, e.g., among the 38 millions of France belonging exteriorly to the Catholic Church many, perhaps even many millions, are interiorly altogether separated from the Church, just as in Germany and other Teutonic nations we have the analogous fact regarding Protestantism. In the Christian religions which are, after all, the most important, membership, ever since the days of primitive Christianity, is founded on baptism; this membership, from the point of view of statistics, must be considered as severed only by a formal withdrawal or excommunication from the particular religious body. In official census of religions nothing but the individual’s own declaration comes into consideration.
A census represents the religious status of a country at a given date. Of course, when hundreds of states are to be taken into account, there cannot be one fixed date, but at least a limited period ought to be assigned, so that the calculations for the different countries may not lie too far apart. Otherwise the general impression conveyed would not be correct.
On these principles the following tables are made up, the data being taken as a rule from the years 1905 to 1910, in most cases 1907 or 1908. The results of official census taken in 1910 and 1911 have not yet been published, and although a few more recent figures have become known since these lists were put together in 1910 for the “Staatslexikon der Gorresgesellschaft”, they have not been incorporated, in order not to impair the uniform character of the tables. In the first place, the official government census of religions has been followed in each case; but with regard to those countries in which since 1900-1901 no Government census of religions has been taken, though the numerical status of the population is officially ascertained every year, the ratio of the various religious bodies established by the preceding census of religions has been applied to the present number of inhabitants; for, excepting the “immigration countries”, the ratio of denominational membership shows little change within ten years. Where a government census is wanting, the data of the religious bodies themselves are made use of. Our sources are given in full in the bibliography at the end of this article. In Table III (second column) the results of the government census of religions are marked C, along with the year in which the census was taken; the computations founded upon the ratio derived from previous official records, are marked R; the non-official figures and estimates are marked E. (See Table III.)
Of the nearly 430 millions living in Europe at present, almost 411 millions (95.5 per cent) are Christians. The number of Jews (2.3 per cent) may in reality be a little less than appears in the table, as the considerable emigration of Jews from Russia during the last decade could not be taken into account. On the other hand, the natural increase of the Jewish population of Russia, in contrast to that of the Jews in Germany and Western Europe, was exceptionally large within the period in question, so that the total number of Jews living at present (1911) in Europe should be at least 9 millions. Not quite so large is the number of Mohammedans (2 per cent). Finally there remain 1 million (2 per thousand) of other non-Christians, of individuals without religious denomination, etc. Among the Christians, Catholics form by far the most numerous group. They make up 43.8 per cent of the total population of Europe. Formerly the percentage was even higher. The extraordinary increase of the Slavic races, chiefly Greek Orthodox, and the great exodus of emigrants from Austria-Hungary, Italy, Spain, and Ireland are the principal causes of the relative decrease of Catholics.
The Greek Orthodox have, on account of their high birth-rate, outnumbered the Protestants. The former are now 26.4 per cent of the total, the latter only 24.7 per cent, while, according to the earlier computation by Krose, the Greek Orthodox (omitting the Schismatic Orientals added to the “Greek-Catholics” by Juraschek-Zeller and others) were a little below the Protestants. In the total of Christians are included 2,056,000 Raskolniks in Russia (the real number probably is much greater), 232,000 Gregorian Armenians in Turkey, Bulgaria, and Rumania, 24,000 Old Catholics in Austria, and about 9000 Jansenists in Holland.
In Asia (see Table IV) government censuses of religions have been taken only within Russian and British territories. Regarding the other countries only the number of Christians and Jews can be ascertained with any degree of certainty. Of the wide-spread religions of Eastern Asia we have nothing but estimates of very doubtful value. The Christians of the various creeds amount in all to about 32,270,000, only 3.9 per cent of the total population of Asia, which may be reckoned as 829 millions. Among the Christians the Greek Orthodox (in round numbers, 13,800,000) are the best represented; yet the Catholics (12,660,000) come fairly close to them. The Protestants (2,350,000) are far fewer, even if the high estimates of Warneck regarding China and Korea be accepted. The remainder (3,500,000) are Armenians, Raskolniks, Thomas Christians in India, and what is still left of the old Christian communities in Japan.
Of about 6,634,000 inhabitants of Australia and Oceania (see Table V), about 5,240,000 are Christians. The Protestant denominations take the lead (almost 77 per cent of the total). The Australian continent, Tasmania, New Zealand, New Caledonia, Fiji, the Tonga and Navigator Islands are almost completely Christianized; whereas the populations of New Guinea, the Bismarck Archipelago, the Solomon Islands, and most of the smaller groups of islands are for the most part pagan. Jews are few on the Australian continent and New Zealand; Buddhists and Mohammedans are found among the immigrants in Hawaii and on the continent. An official census of religions was taken in New Zealand in 1906 and by the Common-wealth in 1900. As, however, the population has grown very considerably since the last census, we have applied to the Catholics of the Australian Commonwealth the results of the ecclesiastical census of 1909 and raised in due proportion the number of Protestants ascertained in 1900. With regard to the other countries and islands, the Catholic and Protestant missionary statistics have served as our chief sources of information. Thus a fairly high degree of accuracy is attained concerning the Christians, while for the pagans mere estimates have had to suffice.
In Africa (see Table VI) there are, in a total of about 126 millions, more than 11 millions of Christians, of whom more than half belong to the Monophysites of Abyssinia and Egypt. Catholics and Protestants are in almost equal numbers, if we add to Africa the Canary Islands and Madeira, which administratively belong to the European possessions of Spain and Portugal. The main stock of Protestants live in British South Africa, the numerous immigrants being for the most part of English and Dutch extraction; but the Protestants have won many converts among the natives. Of Catholics the greater number reside in the French, Spanish, and Portuguese colonies. With regard to the last-named, especially Angola, much higher figures were formerly given, but without sufficient foundation; hence we have inserted in our table the lowest estimate. Jews are somewhat more numerous in Abyssinia, Tunis, Algeria, and Morocco, their total being about half a million. More than one-third of Africa professes Mohammedanism, which is ever gaining ground and encroaching on paganism; yet the latter remains the religion of the majority. A more accurate determination of the number of Mohammedans and pagans in Africa is not possible, as the population has not yet been ascertained in many districts of the interior.
TABLE IV.—ASIA COUNTRIES.
Commonwealth of Australia
Other British Poss. (I)
German Possessions (I)
America (see Table VII) roughly counts 169 millions of inhabitants. Of these more than half (87,614,635, or 51.8 per cent) are Catholics; 70,868,923 (41.9 per cent) Protestants. In all 93.7 per cent are Christians. The number of Jews, very small up to a few decades ago, has increased considerably of late on account of the immigration from Russia. There are nearly 2 millions at present. The pagan Indians and Negroes may be put down at from 2% to 2% millions; in their case a more accurate estimate is out of the question. The great variety of denominations in the United States makes it very difficult to determine their creeds; an official census of religions has never yet been taken. The American statistician, Dr. Carroll, has tried to find a substitute by inquiries addressed to the church authorities, but in this way he has ascertained only the number of communicants (i.e., according to English and American usage, partakers of the Lord’s Supper, or full members), not the total number of adherents to the different denominations. These data, however, do not carry us very far for the purposes of general statistics of religions. The proportion of communicants to non-communicants differs widely in the various denominations. Calculation of membership in the denominations from these data results only in vague estimates of very doubtful value. Still, as Carroll’s list is of some interest, his figures for the more important denominations are given below (table: “Number of Communicants, United States”) as they appear in “The Christian Advocate” for January 26, 1911, omitting only the ordinal numbers indicating the relative numerical importance of each denomination. From this table it is evident that the Catholic Church is by far the largest religious denomination in the United States, and that, excepting the Mormons, no other body shows as high a rate of increase within the last twenty years, the number of communicants having almost doubled. Further, the totals of the official Catholic Directory (for 1909: 14,347,027; for 1910: 14,618,761) are by far too low. For, although the proportion of non-communicants is much smaller among Catholics than among Protestants, yet, even with Catholics, the number of communicants was, up to 1910, hardly more than two-thirds of the total. Moreover, the statistics furnished by the parochial clergy for Wiltzius’ Directory can, from the nature of the case, comprise only those Catholics who have for some time resided in a parish and are known to the clergy; such records, therefore, fall far short of the reality, on account of the great Catholic immigration and the great fluctuation in the population. Hence the present writer believes that, of the 5% millions whose denomination is marked in our tables as unknown, the majority are Catholics. In those tables, however, Wiltzius’ figures for 1909 have been used in default of any accurate data for another estimate.
The total number of Protestant communicants in the United States, according to Carroll, is 21,663,248. As, on the average, there is one communicant to every 2.6 inhabitants, this would mean at most 56 millions of Protestants, and it is quite clear from Carroll’s statistics that there are millions in the United States unconnected with any denomination, a fact which ought to be taken into consideration in calculating the total number of adherents from the number of communicants. Taking, however, the term “Protestant” in the wider sense explained above (Christians who are neither Catholics nor connected with the Greek or Oriental schismatical Churches), we have put down the number as 65 millions. The number of Jews in full membership given by Carroll is evidently far too low, nor is it clear what Carroll understands by this term in the case of Jews. We have therefore given preference to the number of “The Jewish Year Book” for 1910 (1,777,000). Catholics
Adventists TABLE VI.—AFRICA COUNTRIES
Algeria & Tunis
French N. & W. Africa
Other French Poss.
Spanish Poss. (2)
Portuguese Poss. (3)
British N. & W. Africa
British S. Africa
Other British Poss.
Africa (I) Inclusive of United Oriental Catholics, who were put down separately in the official census of 1907.
(2) Inclusive of Canary Islands
(3) Inclusive of Madeira
(4) With regard to the Catholics of Angola the data vary considerably; we have taken the lowest estimate, 250,000.
In Southern and Central America the determination of religious profession is easier, as the entire population may be regarded as Catholic, making allowance for the few Protestants and the uncivilized Indians not included in the census. The same may be said of Cuba, Porto Rico, Haiti, San Domingo, and the French West Indies, while in the British, Danish, and Dutch colonies there are partly official, partly ecclesiastical data. In Mexico, too, a census of religions was taken by the Government in 1901.
According to the synopsis presented in Table VIII, the entire population of the Earth at present (i.e. the average for the years 1906-08) amounts to about 1561 millions. The various figures show a notable difference when compared with the previous accounts of Krose and Zeller-Juraschek. In the first place, the latest figures are considerably higher, at least as far as the Christian denominations are concerned. The reason of this is that more than a decade has passed since the last calculations. Considering the high birth-rate of the Christian nations, an increase of 10 to 15 per cent is not improbable. Besides, the recent and more accurate census in Southern and Central America brought in far higher figures than the older and rougher estimates. As these territories are almost exclusively Catholic, it is clear that the increase of Catholics apparently surpasses that of the Protestants. On the other hand, the column of Fetish-worshippers and other pagans of lower civilization shows a very considerable decrease, which is explained by the recent estimate of the population of Central Africa. While in 1898 Juraschek supposed the population of Africa to be 178 millions, in 1908 he reckoned the population as 129 millions. Thus in these regions religious statistics are subject to great fluctuations. The total number of Christians amount to 618 millions, or 39.6 per cent of the entire population of the earth. Of the Christians, not quite one-half—292 ¬æ millions, or 47.4 per cent—belong to the Catholic Church; 186 millions, or 30.1 per cent are Protestants; 127¬? millions, or 20.6 per cent, Greek Orthodox; the rest are Oriental Schismatics or belong to sects not separately mentioned in the table—Raskolniks, Jansenists, Old Catholics etc. The Roman Catholic Church alone comprises almost one fifth of mankind, and has more followers than any other form of religion. Buddhism, Ancestor-worship, and Confucianism, which, taken together, would indeed possess a larger number of adherents, are not distinct religious bodies, but forms of religion and systems of religious customs all of which, as mentioned above, are at times observed by the same individual. With reference to the number of adherents, Brahminism and Mohammedanism, of all religious denominations, approach nearest to Catholicism: they each have more than 200 millions of followers. But their extension is not so universal as that of Catholicism; locally and ethnographically they are much more limited. In Table VIII the Jews probably appear more numerous than they are in reality, as the great emigration from Russia could not be determined numerically for want of reliable official statistics, while in the most recent records from countries to which they migrate, the Jews seem to be included. Nevertheless the total number of Jews can scarcely fail to reach 12 millions. (See also Migration : Immigration to the United States.) The remaining 7 1/3 millions not classified are individuals without any religious denomination and, still more, those whose creeds could not be ascertained.
H. A. KROSE