occupation, would be nearly 1,300 acres, or 10,000 mow, a magnitude now almost unheard of and always considered extravagant. The proportion is doubtless mentioned because it is exceptional, and landlords who themselves had to pay taxes could not have so many tenants, while there were always freeholders who paid direct to the State. Still this glimpse of a large class of cultivators, who would be omitted from the census, on the ground of their not being taxpayers, is a sufficient reason for doubting whether the population of the empire had ever declined as much as the decline in the number of tax-paying families seems by itself to indicate.

Two classes of families are recognised in the Sung registers-the landowners in their own right and those who farmed the land of others on a kind of métayer system, who were described as "guests." With the prevalence of subletting, laws and customs grew up for the protection of the tenants, which are still in force, and have in effect robbed the rôle of landlord of all its charms. A fair rent and fixity of tenure were secured to the cultivator, and as the central Government again became strong enough to assert its claims as overlord, by the appropriation of all waste or ownerless plots, its influence began to be felt again as formerly in preventing rack-renting.

The custom of letting land had become so common that in 997 A.D. a minister, who was desirous to provide cultivators for the deserted fields round Cai-fong-fu, proposed as the most natural course to let them to willing tenants at a moderate rent, instead of ceding them out of hand, as had been done in earlier times, or granting them for a term, subject to redemption by the Crown, when the tenant ceased to rank as an ablebodied taxpayer. When the land was granted for three years rent free, the rent after that time was at first fixed at 50 per cent. of the produce, following no doubt the precedent of the metayer system in use among private persons; but the Government seems on the whole to have preferred a fixed rent, subject to allowances for bad years. This rent may have included a slight advance upon the ordinary land tax, to which the freehold properties were subject, and indeed otherwise the freeholder who had bought his farm would be at a disadvantage compared with the State tenants who had paid nothing for a lease in perpetuity.

But it is obvious that, as freeholding again became the rule, the distinction between the cultivator paying land tax for the farm of his ancestors, and the cultivator paying rent for a smaller farm, would become more and more unreal: the process by which the tax became a rent would reverse itself, and the rent would recover its normal character of a land tax. The land laws of the present dynasty, which, so far as concerns the agricultural land of China, are substantially borrowed from the Ming, show us the outcome of the agrarian policy of the Sung. The drift of legislation, apart from the Innovator's vagaries, was to assimilate the landlord's position to that of the Emperor, and not conversely. Land could only be let at a fair rent, and when once let, the farmer was not allowed to be disturbed without legal cause.

Nevertheless, large estates continued to grow, and in 1263 it was again proposed to limit the amount of land which might lawfully be held by a single owner to 100 mow; and everything above this amount was liable to be bought or confiscated by the Crown. It was considered a proof of loyalty and liberality voluntarily to surrender such excess lands, without waiting to have them claimed; and a sufficiently absurd abuse is said to have grown up in consequence, officers not possessed of any land, but desiring to pay their court in this way, having obtained possession tyrannically of other men's lands, that they might make a merit of surrendering them.

Choo-hi is said (in a work compiled under the Ming, 1602 A.D.) to have put an end to disorders which prevailed in the district where he resided by restoring the system of public storehouses for grain. It was complained, that every spring and summer the rich closed their granaries and sold corn at a high profit, till the poor forced them open for plunder, riots and murders being multiplied. Choo-hi having taken counsel with the people of the neighbourhood established a public granary from which corn was issued at a steady price, security being taken for its repayment. The satisfactory result was reported to the emperor, and other provinces. commanded to follow the same plan. The record of the grievance and the popular mode of dealing with it is no doubt quite authentic, and the story harmonises with what we are told of the attitude of the modern Chinese towards speculators for a rise in corn. But it is not clear how Choo-hi's action, which is praised, differs in principle from some of the measures of Wang-ngan-shi, which were regarded as practically mischievous as well as speculatively unsound. The fact seems to be that experiments in State socialism in China are judged by their result, which is only satisfactory when they are devised and carried out by exceptionally able and absolutely disinterested officials.

CHAPTER XVI.

TAXATION AND FINANCE UNDER THE SUNG.

Up to the year 1021 the record of the area of cultivated land subject to taxation continued to increase proportionately with the population. By 1052 it had fallen off unaccountably,—more than 50 per cent.,—whilst the number of tax-paying families still multiplied. Officers of the time, according to Ma-twan-lin, reported that as much as 70 per cent. of the cultivated land escaped assessment; but this is probably an exaggeration, as if taken literally, it would imply that a quarter of the whole area of China was under cultivation, which is scarcely possible, considering the scantiness of population in the wilder districts.1 It seems most probable that the large landowners continued to defy the tax collectors, and it is possible that while the State tenants were counted as householders in the census, their lands, for which the rents were paid separately, were not counted as liable for land tax.

It is not conceivable that the average size of holdings should have been reduced by half in the course of thirty years, and the reference to a proposition made in 1062 to tax 20 per cent. of the cultivated lands, seems to imply that then, at all events, no attempt was made to tax, nor therefore to register, all the lands under cultivation. Usage oscillated between two plans, both sanctioned by antiquity, of levying a certain percentage of the whole produce and levying the whole produce of a corresponding fraction of the productive land. Either plan was open to objection, and it is likely that financiers from time to time thought it desirable to meet the abuses that had grown up under one system, by reverting to the other, till that in its turn became unworkable.

2

M. Biot gives, after Ma-twan-lin, tables showing the total revenue from direct taxation in the years 997, 1021, and 1077 A.D. Besides the large contributions of grain, copper money, silk of various qualities, and fodder, mention is also made of hempen cloth, gauze, tea, firewood, coal, feathers, iron, and wood suitable for arrows; in 1021, leather, salt, paper, and undressed hemp are also added. It would be useless to attempt an exact valuation where there are so many necessary elements of uncertainty, but the copper money received in 997 may be taken as worth under 1

1 Fokien, which in 1812 had an average population of 276 to the square mile, had under the Sung only about 85, and the two Kwangs about 20 to the square mile, instead of over 160. 2 Journal Asiatique, 3me série, vol. vi. pp. 315–9.

millions sterling, and the grain for the same year, while worth perhaps twice as much in money, would have served as rations for nearly five million persons for a twelvemonth.

In 1021 there is a slight increase in the grain receipts and a very material increase (nearly 35 per cent.) in the copper money. By 1087, when the area of tax-paying land had fallen off, the receipts in grain had fallen from 21 and 22 million chi to 18 million, and the money receipts were intermediate between those of the two former periods in other words, the growth of the revenue had been arrested, but there was no positive falling off in its amount if the eighty years are taken together.

The increase of the receipts in copper seem to show that the Govern. ment had not been unsuccessful in its first attempts to deal with the standing difficulty of the currency. The first innovation was the introduction of iron money in addition to the familiar copper. Though it was even more cumbrous-a shilling's worth of iron cash weighed over 2 lbs.—yet it was found less laborious to use iron money in remote provinces, where iron was found and copper was not, than to import the necessary quantity of copper from a distance. On this ground the use of iron money was first sanctioned in Fokien and Sz'chuen, where it had been invented. So long as the iron money was in use alone, no fresh inconvenience was felt except from its great weight, and to meet this a first experiment in private banking was tried in Sz'chuen. A certain Tchang-yang invented the idea of notes or coupons exchangeable for metal money at a specified date : "this was not a money; 1 it was only a means of transporting the value of metal money." After 997 this invention received further development.

Sixteen leading firms associated themselves for the purpose of issuing kiao-tsze, or bills of exchange, which were to be repavable at intervals of three years over a total period of sixty-five years. The bills were at first in great favour, and it is perhaps not surprising that some of the associated firms should have been tempted by the confidence reposed in them to indulge in rash speculations; the heirs of some of the sixteen partners were unable to meet their engagements, litigation followed, and as the case was one not contemplated by the common law of China, no satisfactory issue was possible, and in 1017 the Company was wound up by order of the local governor. The kiao-tsze, however, had become popular, and their suppression gave rise to the more inconvenience because engagements could be made by their help, which it became exceedingly costly to fulfil without them. Accordingly, in 1023 the Government itself established a bank for the issue of bills of exchange, and private persons were prohibited from engaging in the same business.

In other parts of the empire the "flying money" of the Tang had already been re-introduced under the name of pien-tsing, or "convenient money."

1 Ma-twan-lin ap. Biot, Journ. As., 3me sér., tom. iv. p. 217.

2 The term tsien-yin="stretching," as of a bow-string, is used for credit in general, and besides the kiao-tsze we meet with hoei-tsze, private agreements or bonds; kwan-tsze, or frontier bills, used in paying army contracts, and promissory notes called tsing-ti="so thing to counter-balance a real possession."

VOL. II.-P.C.

some

These bonds when given in exchange for metallic deposits were of course repayable on demand; and they only served to increase the amount of money in circulation, in so far as they passed from hand to hand oftener than the money they represented would have done. Similar bonds seem to have been given, as previously, on the security of tea, salt. or other recognised values, and these were only repayable in kind, so that they passed from hand to hand as money and could only be "cashed" by some one in want of the commodities specified on the bill. The pien-tsing thus, though a great convenience when money was scarce, were not a real paper money they could not be issued in excess, as the quantity in circulation was regulated by the deposits of private persons; according to Biot, it never exceeded the moderate proportion of one in thirteen of the copper money in use.

To set against this security, however, there was the inconvenience of the original deposit which had to be reclaimed to time, and in the case of the salt and tea bonds the necessity of converting the paper into goods within a limited period. For all these reasons the pien-tsing disappeared from use with the introduction of the Government bills of exchange, a real paper money, which, like that of the first private banks, was to run for sixty-five years, covering twenty-two triennial terms of repayment. In the middle of the century the amount due for repayment at each term is said to have been between £300,000 and £400,000.

The Government, it is obvious, had not realized that the object to be aimed at was a permanent addition to the currency, limited only by the credit of the State; for the triennial cash payments seem to have been made regularly for nineteen terms, so that the paper currency must have contracted considerably before it was proposed (early in the administration of Wang-ngan-shi) to redeem the expiring bonds with a new issue to run for twenty-five terms, or seventy-five years in all.

Meanwhile the introduction of paper had done nothing to lessen the evils arising from the joint use of iron and copper, or from the dearth of both. To save the expense of carriage and to meet the cost of border wars, mines and foundries were opened in the neighbourhood of the armies, and the local metal, whatever it might be, was put into circulation. Unfortunately, however, when copper and iron were used together, the nominal value of the two coinages was not proportioned to their real comparative value in exchange; even that proportion fluctuated, and speculators exaggerated the fluctuation, so that all the disadvantages of a bimetallic currency were introduced with little corresponding benefit.

Some idea of the confusion may be formed from the report of an official, protesting against the instructions he had just received to buy up the superfluous iron money at a fixed price. "For 40 copper coin," he observes, "we get one pound of bad iron money, and 1,000 pieces of small copper money may be exchanged for 1,000 iron pieces of a value of 2,000. A thousand small iron pieces contain 6 lbs. of iron, and if people cast 2,000 pieces of it and exchange them for 1,000 copper pieces, the Government