Encyclopædia Britannica, Ninth Edition/Population
POPULATION. The phenomena of population are the product of physical forces the nature of which it will be necessary to investigate. It will, however, be convenient to consider population, in the first place, as a statical phenomenon, that is, to observe and classify the principal features it presents, without attempting to investigate the system of causes of which they are the effects. Thereafter the dynamical aspects of the subject, namely, the general laws governing the forces whose joint action has produced population, will receive attention.
I. Population, statically considered, may be defined as “the totality of human beings existing within a given area at a given moment of time.” This definition is identical with that adopted by Haushofer (p. 87), except that that eminent authority thought it unnecessary to add the clause relating to time. The totality just mentioned is ascertained in modern times and by civilized nations by the statistical operation known as the Census (q.v.). It is usual to obtain by means of a census a good deal of information beyond the bare fact of the number of persons whose existence is, for the purposes of the census, taken cognizance of. Part of this information is obtained for purposes connected with the administration of the state, such as that contained in replies to questions as to the religion, profession, &c., of the individuals numbered. But these facts, though highly important, are not facts of population strictly speaking. There are two very important characteristics common to all considerable populations—namely, the approximate constancy of the distribution of the population as regards sex and age. A census which did not distinguish between the number of male and the number of female persons composing the population of which it takes cognizance would be seriously defective. Inquiries as to the height and the girth round the chest of individuals are usually made in countries where military service is compulsory, and the degree of prevalence of bodily defects, such as blindness and deafness, is also noted for similar reasons; but such inquiries are the work of specialists, official and other, and in any case are not included in the information obtained from a census. The age of each individual is, however, easily obtained in the course of the operations of the census. We shall now briefly set forth the general characteristics of a population, examined at a particular point of time and without reference to similar phenomena at previous points of time.
Population of the World.—The total population of the world is, to a large extent, an estimate, inasmuch as in some countries a proper census has never been taken, while in many the interval that has elapsed since the last operation is so long as to reduce it to the level of serving as a basis for a calculation in which estimates play a large part. So great, indeed, is the uncertainty in which all such calculations are involved that an eminent French statistician, M. Block, abandons all attempt to deal with the problem, dismissing the subject in the following note (Traité, &c., p. 401),—“Nous abstenons de donner le chiffre de l’ensemble de la population de la terre; personne ne connait ce chiffre.” With this view of the matter we entirely agree, without, however, any disparagement to the valuable work done by Behm and Wagner, who have made the population of the earth their special study, and are under no illusions as to the accuracy of the results they have to offer. The work of these two eminent men of science has at any rate drawn attention to the lacunæ, in our present knowledge, besides arranging and co-ordinating the great multiplicity of well-ascertained facts at our disposal. As civilization advances the area of the unknown or partially known, which is at present large, will gradually diminish.
|Author of Estimate||Year.||Number (in Millions).|
|Graberg v. Hemsö||1813||686|
|Behm and Wagner||1880||1,456|
Table I. (p. 513 supra), taken from Haushofer’s work (Lehr- u. Handbuch, p. 90, note 1), will show how greatly the estimates of the world’s population have varied since people first began to make them. We venture to say that any person of fair intelligence and ordinary education would, even without any statistical training, come to the conclusion that there was nothing certain to be known on the subject which these figures profess to illustrate. The fact that Behm and Wagner’s latest estimate is less than that published by them two years previously shows how difficult the subject is. We should add that the reasons given by them for this discrepancy, for even a tyro would have expected a slight increase, are quite satisfactory, and add to our confidence in that part of the investigation for which they profess to give figures approximating to accuracy.
|Area in Square
|Number.||Per Sq. Kilo.||Per Sq. Mile.|
Sex.—The obstacles which make it difficult to attain even an approximate statement of the population of the world prevent us from obtaining any accurate knowledge whatever as to the sexual constitution of that population. We have, however, tolerably accurate information on this subject for most of the countries of Europe, for the United States, and for Canada. From the figures available it is evident that no general proposition can be laid down on the subject of the normal proportion of females to males, except that in so-called “old” countries there is usually a slight excess of the former.
|Year.||Females to each|
|England and Wales||1871||1,054|
The census of England and Wales for 1881 gave 1055 females to 1000 males. A slight tendency to an increase in the proportion is perceptible in some countries, and to a decrease in others, as the following table (IV.) given by Wappäus and quoted by Haushofer (p. 217) will show. The reader will observe that Wappäus’s figures are the proportions to 100, not to 1000, as in Table III.
The 1880 census of the United States states the proportion of females to males at 96·54 per cent., which is rather smaller than that shown in 1870 (97·2 per cent.); but immigration is still a potent factor in the growth of the population of that country.
With regard to the causes of the excess of females, as in most other social phenomena, our knowledge is very small at present. The reason for the broad distinction between Europe and North America is pretty obvious. New countries are continually receiving many male and fewer female immigrants. Probably also, life being very rough in the more unsettled portions of such countries, the rate of mortality among females is a little higher than in places where women can receive more protection from hardship. On the other hand, even in Europe men run many risks to which women are not exposed. The subject is a very interesting one, but cannot be adequately treated except at much greater length than is possible here, and we must refer our readers to special works for further information.
|Average for Europe||121||108||100||92||87||78||134||112||85||55||21||5||0·4|
Age.—The characteristics of a population from the point of view of age, which German writers term “Altersaufbau” can only be treated very generally. Table V. on p. 514 is quoted by Haushofer (p. 213) from Von Scheel’s Handbuch der Statistik.
This “age scale” shows us the proportion in which persons of various categories of age are found combined to form populations. The general characteristics of the groups are tolerably obvious. It must be remembered that after thirty years the periods are decennial. The difference between the age scale of Europe and that of North America is considerable. In the latter, owing mainly to the fact that emigrants are usually young, a much larger proportion of the population than in Europe are under thirty years of age. On the other hand the age scale of France presents a feature of an opposite kind, namely, a deficiency of persons under fifteen years of age, and an excess of those over forty, as compared with the average of Europe. This conformation of the age scale may be compared with that of Hungary, where the number of children is larger and the number of persons over forty less than the average. It is probable that the smaller number of children in the one case and the larger in the other directly lead respectively to a smaller infant mortality in France than in Hungary. As M. Block observes (Traité, p. 409), “Nous avons moins d’enfants; mais, grâce à une moindre mortalité dans le jeune age, nous avons plus d’adults.” It is obvious that cæteris paribus it is easier to pay the requisite attention to the rearing of a small number of children than to do the same for a larger number.
Careful inquiries into age scales are of very recent origin, the data required for evaluating those relating to earlier periods being absent. Moreover, erroneous statements as to their age are made by a much larger number of persons than might be supposed, sometimes from carelessness or ignorance, but also intentionally. The tendency of women over twenty-five to understate their age, combined with overstatements of age by girls and young women under twenty, always tends to make the twenty to twenty-five section of the age scale unduly large (see Census of England and Wales, 1881, vol. iv., “General Report”). We must regard even the age scales now in existence as merely first approximations, for it is evident that observations obtained from several censuses must be reduced and combined before we can feel certain that accidental causes of error have been eliminated. This is all the more necessary as the age scale of any given population cannot be regarded as fixed, any more than the magnitude of the population itself, both being liable to modifications arising out of the varying dynamical conditions existing at different periods. And this brings us to the second portion of our inquiry, in which we shall indicate in the most general way the nature of the proximate causes which underlie the phenomena of population considered as a fact existing at a particular moment of time.
II. Population, dynamically considered, is the result of two pairs of opposing forces, whose combined action may, for convenience, be theoretically conceived of as balancing each other, but which never do so balance as a matter of fact. A comparison of two successive censuses invariably shows some “movement of population.” In nearly all civilized countries the movement shown is one of growth when the body of population examined is large. The population of a village or a small town may, quite conceivably, show a reduction in number for the period between two censuses, but this can hardly be the case with a large town, and still less with a nation, unless as the consequence of some great calamity such as an earthquake or a pestilence or a change in the climatic or economic conditions of the country inhabited. A great war, of course, produces a certain retardation of the rate of increase. Although some of the uncivilized peoples of the world are rapidly disappearing, the tendency of the population of the whole world is evidently to increase—at what rate it is impossible to say, for reasons already mentioned; and our inquiry will, therefore, be confined to peoples regarding whose population we have comparatively accurate information for an adequate number of years.
The causes of the movement of population are internal and external. The internal arise out of the numerical relation between the births and deaths of a given period, there being an increase when there are more births than deaths, a decrease in the contrary case. Haushofer expresses this by a formula which is sometimes convenient:—“There is an increase where the intervals between successive births are smaller than those between successive deaths” (p. 115). The external are immigration and emigration. The intensity of these two forces operating on population depends on a variety of causes, into which we do not propose to enter. Generally speaking, it may be said that “new” countries, where the density of population is small, attract immigrants from countries in which the density of population is great. The density of population is expressed by the figure denoting the number of inhabitants per square mile (or square kilometre) of the territory they occupy. For a discussion of the various political, social, and economic causes which determine density of population, we must refer our readers to the works of Haushofer (p. 173) and Block (p. 456). Before analysing the components of the movement of population it will be useful to examine briefly that movement itself, and ascertain what is its normal rate in civilized countries. The mode of expressing this rate which is most commonly adopted in the exposition of statistics of population is to state the number of years in which a given population “doubles itself.” It is not a very scientific method of expressing the facts, since it assumes that the rate of a few years will continue for a period of many years, but, in deference to custom, we give a table constructed in accordance with it.
|Basis of Calculation.||Approximate|
It must be noted that, while the table may be relied on so far as Signior Bodio’s treatment of the data goes, the data for the earlier part of the century are very defective, and the results deduced from them must be regarded as less trustworthy than those for the more recent of the two periods.
The above tables of increase of population include the effects of immigration and emigration, regarding which we have nothing further to say in this article, as the causes of these phenomena are too heterogeneous for general treatment. Moreover, except in comparatively unimportant cases—unimportant, that is, from our point of view, but by no means so from the standpoint of the statesman—the effects of these two causes are small, the main cause of the growth of population being the internal forces already mentioned, namely, the birth-rate and the death-rate.
During the earlier half of the century the rate of increase in the United States ranged from 2 to 3 per cent. per annum in the successive decades from census to census. The increase in the population of the United States has hitherto depended so much on immigration that at present inquiries into the normal birth and death rates of that country are very difficult, except in the eastern States. Of the total population, 50,442,060, as shown in the census of 1880, no less than 6,619,943, or over 13 per cent., were foreigners. The fact already mentioned, that the proportion of women to men is unusually low, serves to remind us that normal phenomena of population must not as yet be looked for in the American Union.
The Birth-Rate.—The birth-rate of a population is the proportion borne by the number of births in a year to the number of the population. It might seem that it is easy to obtain this rate, but as a matter of fact it is practically impossible to do so. It is not difficult to ascertain, with sufficient accuracy, the number of births; the difficulty is to ascertain what is the number of the population, for that number is never the same for two days together. It is obvious that it would never do to evaluate the birth rate of the United Kingdom, say for 1885, by means of the figures obtained in the census taken on April 4, 1881, and the error would be greater next year, and greater still the year after. The growth of the population since the last census must, therefore, be taken into account; but, even when it has been decided to adopt this plan, there is the difficulty of fixing on the date up to which the additions are to be made. The usual practice is to take the population of a date as near as possible to the middle of the year for which the birth-rate is required as the basis for the calculation. We mention these difficulties as a caution to students of statistics. The following table (VIII.) quoted by Haushofer, p. 123, is taken from Bodio’s Movimento dello Stato Civile (Rome, 1880); the figures for the minor countries have been omitted, and still-births are excluded:—
|Period Observed.||Average Yearly|
Number of Births to
|England and Wales||1865–78||3·56|
|Russia in Europe ¹||1867–75||4·95|
¹ Excluding Poland.
The birth-rate in different countries is influenced by various circumstances into which it is not possible to enter at length. The most important circumstance is the proportion borne by the number of women of child-bearing age to the whole population. There are other circumstances which must be kept in mind in comparing the birth-rates of different countries, such as the character of the age scale as a whole, and the density of population, besides climatic and other physical characteristics of the environment of the populations examined. The birth-rate is high in new countries, where there is always a larger proportion of young men than in old states, and where the proportion of women of child-bearing age is also large. The latter circumstance is, we may point out, quite consistent with the statement already made, that in new countries the proportion of women to men is smaller than in old ones. For an unusually large proportion of the total number of women in new countries are young.
Some facts relating to the absolute number of births may here be briefly referred to. The most important of these is its composition as regards sex. We have already seen that in most populations there are more women than men. This is not a consequence of there being more girls born than boys, for the fact is just the contrary. The following table (IX.) shows the number of male births to every 100 female births which took place in the undermentioned countries during the periods stated (Movimento, &c., p. 126; Haushofer, p. 218):—
for 100 Girls.
|England and Wales||1865–78||104|
|Russia in Europe||1867–74||105|
On the somewhat anomalous figures we must observe that those relating to Greece and Servia are possibly to be explained by the hypothesis of inaccurate returns. We may add that, if a distinction is made between legitimate and illegitimate children, it is usually found that the excess of male births is greater among the latter. In countries, therefore, where the proportion of illegitimate to legitimate births is high there will usually be a higher proportion of male to female births than in countries where there are not relatively so many illegitimate births (Block, p. 429). Interesting inquiries have been made into the facts regarding the distribution of births during the year, showing that there are, as a rule, more births in some months than in others, and also as to the influence high prices for the primary necessaries of life have on the number of births (Mayr, p. 235).
The Death-Rate.—The death-rate of a population is the proportion borne by the number of deaths in a year to the number of the population. The population is to be reckoned as has been already described in dealing with the birth-rate. This very important statistical quantity is sometimes confused with another relating to the same phenomenon, namely, the mean duration of life. The difficulties in obtaining an accurate death-rate are, if any thing, greater than in the case of the birth-rate.
|Countries.||Period Observed.||Average Yearly|
Number of Deaths
to 100 Inhabitants.
|England and Wales||1865–78||2·20|
|Russia in Europe||1867–75||3·67|
This table is sufficient for our purpose, which is to give a general idea as to the death-rate of these countries. Much more accurate approximations are, however, needed for actuarial purposes, and very elaborate valuations of the death-rate will be found in G. F. Knapp’s work Ueber die Ermittelung der Sterblichkeit (Leipsic, 1868).
We must now show how the death-rate is usually composed as regards age. The following table (X.) shows the number of persons out of every hundred deaths who died at the undermentioned ages in each of the countries named (Haushofer, p. 143; quoted from the Movimento):—
It will be seen that from nearly one-fifth to nearly one-third of the deaths were those of children less than twelve months old. The very high proportion of deaths at this age in Bavaria was some years ago made the subject of a special inquiry by Dr Mayr, and it was found to be largely due to the bad mode of bringing up infants peculiar to certain localities (Mayr, pp. 91, 319).
The composition of the death-rate in regard to sex must be touched on briefly. As we have seen, more boys are born than girls. Owing, however, to the greater mortality among the former their number is rapidly reduced during the first few years of life, so that at any given moment the population is composed as stated in the age scales. The exact mode in which a given number of persons born in the same year disappears by death is shown in the elaborate tables of mortality used by actuaries. These tables are different for different countries and for males and females. Very elaborate tables of survival were prepared for the British Government in 1883–84 for calculating annuities.
We cannot here deal with what is known as the “population question.” Any adequate discussion of that highly important subject would involve considerations outside the limits of this article. The “population question” is a question of conduct, while the present article seeks only to point out certain well-ascertained facts regarding the phenomenon of superorganic evolution called population. The facts in question are general, and, though sufficient to indicate the nature of the phenomenon, and the broad divisions which are most convenient for its further investigation, are quite insufficient as the basis for the formation of any ethical judgment regarding the actions of the individuals composing the population.
Among the works that may be consulted to the greatest advantage by the student are the following:—the numerous works of the late Dr William Farr, F.R.S., formerly registrar-general; various works by Dr W. A. Guy, F.R.S.; those of Adolphe Quetelet; various monographs by Dr Ernst Engel, and other eminent statists in the official publications of the Prussian Statistical Office; various publications of the Italian Statistical Bureau; the publications of the lnternational Congress of Statistics. Systematic treatment of the whole subject of population will be found in the following works:—Bertillon, Mouvements de la Population dans divers états de l’Europe, Paris, 1877; Maurice Block, Traité théorique et pratique de statistique, Paris, 1878; L. Bodio, Movimento dello stato civile, Rome, 1878 (publication of the Italian Statistical Bureau); A. Gabaglio, Storia e teoria della statistica, Milan, 1880; M. Haushofer, Lehr- und Handbuch der Statistik, Vienna, 1882.