development, and money-making powers of different localities. But the conqseuent inequalities in the prices of agricultural labour in England, it is important to notice, were formerly compensated for in a good measure by corresponding inequalities in the prices of commodities; food was cheap where wages were low, food dear where wages were high. Prices rose around the great centres of mining, manufacture, and trade to a scale greatly above that prevailing in purely agricultural districts, just as they have risen in the industrial districts of West Germany above the scale in Pomerania, Silesia, and Posen. A new inequality in agricultural wages in England took place with the equalization of the prices of food by railways and roads. In not a few parishes in the southern counties money wages remained almost stationary, at 7s. to 8s., from 1770 to 1870, while meat rose in the interval from 2d., and in some places actually only 1d. a pound, to nearly the same price, say 10d., as in Yorkshire and Northumberland, and milk and butter in proportion. When Mr. M'Culloch was laying down in successive editions of Adam Smith's treatise, that canals, roads, railways, &c., had brought the prices of produce much nearer to a common level than at the period of the publication of the Wealth of Nations,' he was right enough about the prices of produce, but so wrong about wages that they varied in England at the time of his last edition from 9s. to 22s., not unfrequently with a cottage and garden in the latter case, and without either in the former. Moreover, on the rise in the prices of food in the southern counties, brought about by their equalization through the kingdom, supervened a succession of further rises caused by the general increase of population, the increased wages and consumption in the manufacturing districts, and the increase of both the metallic and the credit circulation of the country. The recent rise of money wages in the southern counties had in fact been preceded by a fourfold fall in real wages measured in the price of animal food, and a great fall measured in cottage rents. The rise, therefore, during the last three or four years in the wages of the southern farm labourer-though in some cases from 7s. and 8s. to 12s. and 13s.-is, compared with his real wages a century ago, almost a nominal one only. Another inequality which the application of steam to manufactures and locomotion brought about, was a difference between the wages of mechanical and agricultural labour (unless in the manufacturing districts) out of all proportion to any differences in the severity of the employment and the skill and knowledge. required. Adam Smith thought the ploughman in his time beyond comparison superior in skill, judgment, and discretion to the town workman or mechanic, and he certainly is not now inferior in the proportion of his pay. In so far as he is ignorant and inefficient, it is the effect, not the cause, of his low wages; and the groundlessness of a recent suggestion that the lower wages of the southern counties are attributable to an inferiority of race, is proved by the fact, which might have been learned from Mr. Caird, that a hundred years ago wages were higher in the southern than in the northern counties, in the proportion of 7s. 6d. to 6s. 9d. a week. The truth is that the sources of wages are unequally distributed by nature, but a more equal local development and a more equal distribution of labour might long ago have been brought about, had economists, in place of assuming equality, examined the facts, and ascertained the actual inequalities and their causes. It remains briefly to indicate some of the conclusions to which the foregoing review of facts, and the present conditions of the situation, seem to point. A rise in the money wages of agricultural labour is a general European phenomenon, showing the operation of general causes. It is everywhere in part merely a nominal rise, but especially so in the southern counties of England, where a great fall in real wages, by reason of a continuous rise in prices, preceded it. In all countries in Europe there are great inequalities in money wages, but in most of them, Germany in particular, those inequalities are partially compensated by local differences in the prices of food, so that no such real inequalities in real wages exist there as in England; nor is it possible that such inequalities can continue here, now that the economic fictions which concealed them and their causes have been exposed. The future rate of money wages is indeed everywhere partly a question relating to the fertility of the gold mines and the abundance of money; but whatever may be the amount of money in circulation, it may be predicted that its distribution will be such as to secure a larger share than heretofore to the agricultural labourer in the southern counties of England. Temporary and local causes have lately checked the rise of agricultural wages in some countries, but all the permanent and general conditions tend to maintain it, especially in the lower levels. The recent decline in Germany is, in fact, a mark of the increasing influence of manufactures and trade on the price of agricultural labour, and of the disappearance of customary and stationary rates of agricultural wages, and of the excessive disproportion between them and the rates in other employments.* Everywhere in Europe farm labourers are breaking the bonds of tradition and habit, everywhere acquiring new powers; in England they have, moreover, gained those potent forces of union which have long been the monopoly of town workmen, and farmers in the depth of the country are beginning to feel at once the effects of the competition of distant employers in towns and in the new world, and of the combination of their own labourers on the spot. When one considers the number and strength of the causes tending to the elevation of the agricultural labourer, and the breadth of the area over which they are operating, one can hardly err in affirming the ultimate futility of a local effort to put a stop to a movement which, on the one hand, has what may be called universal and permanent forces on its side, and, on the other hand, is fortified by special local conditions. The dark part of the prospect is, that England, in the outcome, may lose the chief part of its rural population, for it seems too plain that nothing will awaken its legislature or its landowners in time to the importance of making it the well-grounded hope of the industrious and thrifty farm labourer to acquire a little farm of his own. * Since this Essay was written, nearly five years ago, the diminished activity of manufactures and trade in Germany has been attended with a further decline in agricultural wages, which are not much higher in many places now than in 1867. Formerly the movements of trade did not affect German agricultural wages. January, 1879. ECONOMIC SCIENCE was not formally included within the Statistical Section of the British Association until 1856; and even since then, addresses of distinguished Presidents of the Section have turned mainly on its statistical functions, and have been devoted principally to an inquiry into the nature and province of statistics. That the inquiry is neither so superfluous nor so easy as might at first appear, is sufficiently shown by the fact that there are, according to the great German statist, Dr. Engel, no less than 180 definitions of the term to be met with in the works of different authors. These various definitions may, however, be said to group themselves round one or other of three conceptions, of which one follows the popular view of statistics; the etymological and original meaning almost disappearing in the notion merely of tables of figures, or numbers of facts, of which the chief significance lies in their numerical statement. According to another conception, statistics, following etymology and the signification given to Statistik by the famous Göttingen school, should be regarded as equivalent to the science of States, or political science, but, nevertheless, as confining itself to the ascertainment and collection of facts indicative of the condition and prospects of society, without inquiring into causes or reasoning on probable effects, and carefully discarding hypothesis, theory, and speculation in its investigations. A third conception is, that statistical science aims at the discovery, not only of the phenomena of society, * Written on the meeting of Section F of the British Association for the Advancement of Science, September, 1873. but also of their laws, and by no means discards either inquiry into causes and effects or theoretical reasoning. It is curious that some who give to statistics the first of these three meanings, and who regard the numerical statement of facts, and the marshalling of tables of figures, as the proper business of the statistician, nevertheless speak of statistics as a science. But, as the eminent economist, Roscher, has observed, numbering or numerical statement is only an instrument of which any branch of science may avail itself, and can never, in itself, constitute a science. No one, as he says, would dream of making a science of microscopics, or observations made through the microscope. The distinguished English statistician and economist, Mr. Jevons, has likewise condemned the misconception of statistics and the misuse of the term we refer to in language worth recalling :-Many persons now use the word statistical as if it were synonymous with numerical; but it is a mere accident of the information with which we deal that it is often expressed in a numerical or tabular form. As other sciences progress, they become more and more a matter of quantity and number, and so does statistical science; but we must not suppose that the occurrence of numerical statement is the mark of statistical information.' The doctrine that the consideration of causes and probable reasoning are excluded from the province of statistics, and that statisticians should confine themselves to the ascertainment of facts, is hardly more satisfactory. No branch of science, no scientific body, confines itself to the observation of phenomena without seeking to interpret them or to ascertain their laws. It is not, indeed, possible, at present, to explain all the phenomena which come within the observation of the statist, or to connect them with any law of causation; and even naked collections of statistical facts may be useful as aids to further inquiry, or as supplying links in the chain of observed effects. But serious error, and even practical mischief, have followed from attention merely to the recurrence of statistical facts without inquiry into their causes. A theory of a decennial recurrence of commercial crises, for example, was based on the occurrence of crises in 1837, 1847, and 1857. Had the causes |