vol. i. p. 481, vol. ii. p. 542, with much greater accuracy calls such cases ‘apparent exceptions’ to laws; but it would be better to say, ‘exceptions to the operations of laws.’ The context clearly proves that Mr. Paget distinctly apprehends the difference; but a slight alteration of this kind would prevent confusion in the minds of ordinary readers.
33
Mr. Rawson, in his Inquiry into the Statistics of Crime in England and Wales (published in the Journal of the Statistical Society, vol. ii. pp. 316–344), says, p. 327, ‘No greater proof can be given of the possibility of arriving at certain constants with regard to crime, than the fact which appears in the following table, that the greatest variation which has taken place during the last three years, in the proportion of any class of criminals at the same period of life, has not exceeded a half per cent.’ See also Report of British Association for 1839, Transac. of Sec., p. 118. Indeed, all writers who have examined the evidence are forced to admit this regularity, however they may wish to explain it. M. Dufau (Traité de Statistique, p. 144) says, ‘Les faits de l'ordre moral sont, aussi bien que ceux de l'ordre naturel, le produit de causes constantes et régulières,’ &c.; and at p. 367, ‘C'est ainsi que le monde moral se présente à nous, de ce point de vue, comme offrant, de même que le monde physique, un ensemble continu d'effets dus à des causes constantes et régulières, dont il appartient surtout à la statistique de constater l'action.’ See to the same effect Moreau-Christophe des Prisons en France, Paris, 1838, pp. 53, 189.
34
‘It is curious to observe how intimate a relation exists between the price of food and the number of marriages.’ … ‘The relation that subsists between the price of food and the number of marriages is not confined to our own country; and it is not improbable that, had we the means of ascertaining the facts, we should see the like result in every civilized community. We possess the necessary returns from France; and these fully bear out the view that has been given.’ Porter's Progress of the Nation, vol. ii. pp. 244, 245, London, 1838.
35
‘The marriage returns of 1850 and 1851 exhibit the excess which since 1750 has been invariably observed when the substantial earnings of the people are above the average.’ Journal of Statistical Society, vol. xv. p. 185.
36
See Somerville's Physical Geography, vol. ii. pp. 409–411, which, says this able writer, proves that ‘forgetfulness as well as free will is under constant laws.’ But this is using the word ‘free will’ in a sense different from that commonly employed.
37
Achenwall, in the middle of the eighteenth century, is usually considered to be the first systematic writer on statistics, and is said to have given them their present name. See Lewis, Methods of Observation and Reasoning in Politics, 1852, vol. i. p. 72; Biographie Universelle, vol. i. p. 140; Dufau, Traité de Statistique, pp. 9, 10. Even so late as 1800, the Bishop of Llandaff wrote to Sir John Sinclair, ‘I must think the kingdom is highly indebted to you for bringing forward a species of knowledge (statistics) wholly new in this country, though not new in other parts of Europe.’ Sinclair's Correspondence, vol. i. p. 230. Sinclair, notwithstanding his industry, was a man of slender powers, and did not at all understand the real importance of statistics, of which, indeed, he took a mere practical view. Since then statistics have been applied extensively to medicine; and still more recently, and on a smaller scale, to philology and to jurisprudence. Compare Bouillaud, Philosophie Médicale, pp. 96, 186; Renouard, Hist. de la Médecine, vol. ii. pp. 474, 475; Esquirol, Maladies Mentales, vol. ii. pp. 665–667; Holland's Medical Notes, pp. 5, 472; Vogel's Pathological Anatomy, pp. 15–17; Simon's Pathology, p. 180; Phillips on Scrofula, pp. 70, 118, &c.; Prichard's Physical Hist. of Mankind, vol. iv. p. 414; Eschbach, Etude du Droit, pp. 392–394.
38
I cordially subscribe to the remark of one of the greatest thinkers of our time, who says of the supposed differences of race, ‘of all vulgar modes of escaping from the consideration of the effect of social and moral influences on the human mind, the most vulgar is that of attributing the diversities of conduct and character to inherent natural differences.’ Mill's Principles of Political Economy, vol. i. p. 390. Ordinary writers are constantly falling into the error of assuming the existence of this difference, which may or may not exist but which most assuredly has never been proved. Some singular instances of this will be found in Alison's History of Europe, vol. ii. p. 336, vol. vi. p. 136, vol. viii. pp. 525, 526, vol. xiii. p. 347; where the historian thinks that by a few strokes of his pen he can settle a question of the greatest difficulty, connected with some of the most intricate problems in physiology. On the supposed relation between race and temperament, see Comte, Philosophie Positive, vol. iii. p. 355.
39
As to the proper limits of physical geography, see Prichard on Ethnology, in Report of the British Association for 1847, p. 235. The word ‘climate’ I always use in the narrow and popular sense. Dr. Forry and many previous writers make it nearly coincide with ‘physical geography:’ ‘Climate constitutes the aggregate of all the external physical circumstances appertaining to each locality in its relation to organic nature.’ Forry's Climate of the United States and its Endemic Influences, New York, 1842, p. 127.
40
By unemployed classes, I mean what Adam Smith calls the unproductive classes; and though both expressions are strictly speaking inaccurate, the word ‘unemployed’ seems to convey more clearly than any other the idea in the text.
41
This has been entirely neglected by the three most philosophical writers on climate: Montesquieu, Hume, and M. Charles Comte in his Traité de Législation. It is also omitted in the remarks of M. Guizot on the influence of climate, Civilisation en Europe, p. 97.
42
See the admirable remarks in Laing's Denmark, 1852, pp. 204, 366, 367; though Norway appears to be a better illustration than Denmark. In Rey's Science Sociale, vol. i. pp. 195, 196, there are some calculations respecting the average loss to agricultural industry caused by changes in the weather; but no notice is taken of the connexion between these changes, when abrupt, and the tone of the national character.
43
This expression has been used by different geographers in different senses; but I take it in its common acceptation, without reference to the more strictly physical view of Ritter and his followers in regard to Central Asia. See Prichard's Physical History of Mankind, vol. iv. p. 278, edit. 1844. At p. 92, Prichard makes the Himalaya the southern boundary of Central Asia.
44
There is reason to believe that the Tartars of Thibet received even their alphabet from India. See the interesting Essay on Tartarian Coins in Journal of Asiatic Society, vol. iv. pp. 276, 277; and on the Scythian Alphabet, see vol. xii. p. 336.
45
In Somerville's Physical Geography, vol. i. p. 132, it is said that in Arabia there are ‘no rivers;’ but Mr. Wellsted (Travels in Arabia, vol. ii. p. 409) mentions one which empties itself into the sea five miles west of Aden. On the streams in Arabia, see Meiners über die Fruchtbarkeit der Länder, vol. i. pp. 149, 150. That the sole deficiency is want of irrigation appears from Burckhardt, who says (Travels in Arabia, vol. i. p. 240), ‘In Arabia, wherever the ground can be irrigated by wells, the sands may be soon made productive.’ And for a striking description of one of the oases of Oman, which shows what Arabia might have been with a good river system, see Journal of Geographical Society, vol. vii. pp. 106, 107.
46
Mr. Morier (Journal of Geog. Soc. vol. vii. p. 230) says, ‘the conquest of Persia by the Saracens a.
