After this it is hard to believe what he tells us elsewhere, that “the only criterion of a real and permanent increase in the population of any country is the increase of the means of subsistence.”[270] It would be at best a negative criterion and _sine quâ non_,—there can be no increase of numbers without increase of food,—though even then it is not true of a “forced population,” living down to a lower food.[271] But there clearly may be an increase of food without an increase of numbers, unless the character of the people is such that they do nothing with the food except increase by it. Therefore, though, within certain wide limits fixed for us by invariable qualities of human nature, predictions are justifiable on the ground of the law of population[272] or any other economical laws, none that specify a particular course of action as a result of a particular event are trustworthy, till we know the character of the people concerned.[273] Malthus always tries to bear this in mind; and, when he tells us that the lists of births, marriages, and deaths in Mid Europe give more information about its internal economy than the observations of the wisest travellers,[274] he is at once interpreting those figures in the light of a principle, and interpreting the principle by means of the figures. This appears when we look at the four chief conclusions of the general chapter in question. The first is the proposition that in the present state of our industrial civilization the marriages depend very closely on the deaths, and the births on the marriages.[275] Montesquieu says that wherever there is room for two persons to live comfortably a marriage will certainly take place.[276] In old countries experience is usually against any sure expectation of the means of supporting a family; the place for a new marriage is only made by the dissolution of an old. As a rule therefore the number of annual marriages is regulated by the number of annual deaths. “Death is the most powerful of all the encouragements to marriage,”[277] while on the other hand the marriages are a frequent cause of the deaths. In almost every country there is too great a frequency of marriages, which causes as it were a forced mortality. Which of these two mutual influences is the more powerful depends on circumstances. In last century the proportion of annual marriages to inhabitants was in Holland generally as 1 in 107 or 108. But in twenty-two Dutch villages it was as 1 in 64. Süssmilch explained this anomaly by the number of new trades in Holland and the new openings for workmen. Malthus would not have denied this possibility, his startling paradox about death being only a particular case of the general principle that “the high price of labour is the real encouragement to marriage.”[278] But in this case the explanation ought to have applied to all Holland if to any part of it. The real reason came out when Malthus observed that the mortality, which