Olnrn^ll IGam ^rl|nnl library CORNELL UNIVERSITY LIBRARY 3 1924 070 152 859 M- ^ Cornell University ^' 'J Library The original of this book is in the Cornell University Library. There are no known copyright restrictions in the United States on the use of the text. http://www.archive.org/details/cu31 9240701 52859 CSe Hanti HahJ^ THE ENGLISH CITIZEN: HIS RIGHTS AND RESPONSIBILITIES THE LAND LAWS Sir FEEDEEICK PpLLOCK, Baet. EABKISTEK-AT-LAW M.A., HOX. LL.D. EDIN., DUBLIN AND HARVARD CORPUS CHRISTI PROFESSOR OF JURISPRUDENCE IN THE UNIVERSITY OF OXFORD LATE FELLOW OF TRINITY COLLEGE, CAMBRIDGE CORRESPONDING MEMBER OF THE INSTITUTE OF FRANCE THIRD EDITION Honian MACMILLAN AND CO. AND NEW YORK 1896 All rights reserved The First Edition of this Book was published 1883. The Second Edition., 1887. Addenda, to a Re-issue in 1893. The Third Edition, i8g6. WILLELMO .^STVBBS COLLE&AE . COLLEGA MAGISTRO . DISCIPVLVS HAEC . QVANTVLACVNQVE D.D.D. FRIDERICVS . POLLOCK PEEFACE TO THE THIED EDITION The object of this book, as I formerly defined, it, is to make the principles and the leading features of the English law of real property intelligible to a reader who is without legal training, but is willing to take some little pains to understand. It does not aim at being a practical manual, and therefore many things that take an important place in counsel's chambers and the solicitor's office, such as the Conveyancing Acts and the Settled Land Acts, are purposely treated with great brevity. Collateral topics such as local taxation, death duties, and game laws, were and are omitted simply because there was no room for them. Eesearch has done much to the history of the law, and legislation to the modern law itself, in the eight years that have passed since the second edition was issued. The present revision is, for divers reasons, less thoroughgoing than I could have wished ; but I trust that few serious errors remain, and that no recent statutes of general importance have been overlooked. I have found it necessary to rewrite a good deal of the chapter on early customary law, to alter much of the Appendix, and to add a wholly new note on the Origins Vlll THE LAND LAWS of the Manor, in order to keep the antiquarian part abreast of the present state of scholarship. Perhaps we do not know much more about the Anglo-Saxon period than we did a dozen years ago (except that on some points we really know less than we supposed) ; but of the Anglo-Norman and Angevin periods — thanks to Mr. Seebohm, Mr. VinogradofF, Mr. Maitland, Mr. Round, and Dr. Liebermann — we do know a good deal more both positively and negatively. At some points I have now ventured to refer for proofs or details to the History of English Law before the Time of Edward I., lately published by Mr. Mait- land and myself. The dedication to the Bishop of Oxford remains in its original form in memory of the too short time for which I had the honour to be his colleague as a professor in the University. F. P. Lincoln's Inn, Deceiriber 1895. CONTENTS CHAPTER I PAGE Introductory . . i CHAPTEE II The Old English Customary Laws . 19 CHAPTER III The Medieval System . .53 CHAPTER IV Legislation and Transformation . 80 CHAPTER V Development of the Modern Law 110 X THE LAND LAWS- CHAPTER VI PAGE Landlord and Tenant . 139 CHAPTER VII Modern Reforms and Prospects . .164 APPENDIX Note A. The Germanic Land System . .197 Note B. Symbolic Transfer in Early English Customs . 199 Note C. The Origins of the Manor . 200 Note D. Villenage, Villein Tenure, and Copy- holds . ... 208 Note E. Primogeniture in Socage Lands . 219 Note F. "Cestui que use" at Common Law . 222 Note G. Settlements and Perpetuity . 223 Note H. Recent Publications on Land Transfer 228 Index . . 231 CHAPTER I INTRODUCTORY The laws and usages which govern the tenure of land in England are, as a whole, unique. Our land system is commonly called feudal, sometimes by persons who use the word as a disparaging epithet without any clear notion of what it means. This is not in itself wrong, but it conveys a most imperfect notion of the number and variety of the influences that have made our land laws what they are. The statement and the belief implied in it are so inadequate as to be misleading. Almost every possible kind of ownership, and almost every possible relation of owners and occupiers of land to the State and to one another, have at one time or another existed in England, and left a more or less con- spicuous mark in the composite structure of the English law of real property. We have to follow out a long story before we can understand how much and how little feudalism abides in the present state of things. There are still in force many local customs and rights which are now known to date — in their nature in all cases, and in their actual origin in many — from a time B 2 THE LAND LAWS chap. when the feudal system was unheard of, and indeed when private property in land, as we now understand it, was a struggling novelty. The main body of the technical expressions of the law, and of the technical habit of thought which they preserve, is derived from feudalism ; but this feudalism has been deeply modified by circumstances peculiar to England. In Scotland the feudal system grew to its full development with little interference, if any, from legislation ; and the forms of Scottish land law still preserve the system in great com- parative purity. Here legislation has constantly inter- fered, and its effects have been to produce radical changes. These changes have not always been such as the Legislature intended ; in one or two material in- stances the effect has been the very opposite of that which was aimed at. One celebrated measure of Henry VHI.'s reign, the Statute of Uses, was passed in order to restore the ancient simplicity and notoriety of titles to land, though more in the interest of the Crown and other great lords than in that of the public. The object of the statute was before long defeated by judicial construction. But it did not remain inopera- tive ; it had other and quite unexpected results. The first was to make the transfer of land, without any act or ceremony for securing publicity, far easier than it had ever been before. The second, worked out in the days of the Commonwealth and the Eestoration by the ingenuity of two or three lawyers, was to introduce the method of strict settlement of landed property which is practised by a great proportion of landowners to this day. Thus, as we shall see more fully hereafter, a measure intended to compel notoriety and simplicity I INTRODUCTOEY 3 became the chief instrument of secrecy and complica- tion. Turning, on the other hand, from the titles and tenures of owners to the relation between owners and occupiers, we find this resting to a great extent on something thoroughly opposed to feudal ideas — namely, the modern economical conception of land as an article of commerce which, like any other commodity, is bought, sold, and hired for prices regulated by com- petition. In the case of town dwelling-houses, and buildings used for trade or manufacture, this view is carried out to its full extent, and the relation between landlord and tenant is a purely commercial one. As to farm -holdings, it is still otherwise in most cases, owing to the survival of usages and habits which we may in a loose way call feudal if we please. Thus our system of landed property is a structure of the most complex and heterogeneous kind. So great is the technical complication and difficulty of our laws on the subject, that within the special studies of the legal profession the study of them is a speciality of itself. Even among accomplished lawyers the number of those who are well versed in real property law is but small ; the number of those who know the history of the law is a smaller one still. Forty or fifty years ago learned persons might be found, such as the late Lord St. Leonards, who seriously maintained that this complica- tion was inevitable, and was indeed only a mark of per- fection in the machinery. Few persons, if any, can be found to maintain it now. But this very complication which calls for amendment is one of the chief obstacles in the way of amendment being made. The whole subject is such a mystery to laymen that, though they 4 THE LAND LAWS chap. may know something is amiss, they cannot tell where the remedy should begin, and do not know what to ask for. Among lawyers a considerable number are hostile to change, and a greater number indifferent. Those who make the shoe do not feel it pinch, and those who feel it pinch do not know how shoes are made. It has often been said that in no country are landowners so ignorant of their legal position or so dependent on legal advice as in England ; and I believe it cannot be con- tradicted. A century or two ago country gentlemen commonly had some little knowledge of the law of pro- perty ; three centuries ago they had a good deal, as we may learn from Shakespeare ; now they seldom have any unless they have practised at the Bar. It would seem, therefore, worth a serious effort to overcome or break down in some fashion the barrier between the minds of lawyers and laymen which is apt to make dis- cussion between them a game of cross purposes. How this can best be brought about is a knotty question. Statements made in the proper technical terms are in danger of not being understood, or, what is worse, being misunderstood ; while for those who are once accus- tomed to the use of such|[terms it is far from easy to state the same facts, even in the most general outlines, in language to be understood by all men, and yet accu- rate as far as it goes. One common method of popular exposition, not only in law but in other special sciences, is to give loose or insufficient explanations of the terms of art, and then use the terms as if the reader had been enabled really to understand them. This is the most dangerous way of all, being by so much worse than those which lead to mere bewilderment as false know- I INTRODUCTORY 5 ledge is worse than ignorance. Instruction of this kind is answerable, presumably, for the loose talk about primogeniture and entail which still abounds in the mouths of people who might easily know better. Serious and capable writers, Mr. George Brodrick and some others, have done something to provide a remedy. But the discussion of economical questions has left them, as a rule, hardly room enough to show clearly and plainly the legal composition of our land system. Now the whole structure, as we have just said, and as Mr. Brodrick points out in a notable passage of his book, is a result of many successive accretions, and those not of a natural but of a casual sort. It has not been produced by deliberate legislation, nor yet by the spontaneous growth of custom. Hence it cannot be understood by itself. It has no intrinsic coherence, and no organic principles. It is a series of historical accidents, and becomes intelligible only in the light of its historical conditions. And this is not less the case, but rather the more so, when the persons desiring to understand it are viewing it from the outside and are unfamiliar with its details. Our aim therefore will be, so far as our skill reaches and the thing can be done on so small a scale, to disentangle the several historical elements that go to make up our modern English real property law. Let us imagine ourselves placed on some commanding point within the boundaries of a great English estate, looking over its mansion-house and its park, its fields and pastures, its woods and wastes. Over against us there rises an open hill, covered, it may be, vidth bril- liant gorse and heather in their season, and fringed and 6 THE LAND LAWS chap. crested with wild woods. These are open and common lands, over which many persons have rights of putting so many beasts to graze, of cutting turf and underwood for the use of their habitations, and the like, according" to the custom of the country and place. Such rights were explained by our law-books, until quite recently, so as to make them fit into a complete feudal theory of land-holding. They were supposed to have been granted by the lord of the manor to his tenants, or to have grown up within his domain by way of sufferance and usage, till the long-continued approval of successive lords passed from a matter of favour into a matter of right. This, or something like it, may sometimes have happened. But in general the true history is just the other way. The people who exercise rights of common exercise them by a title which, if we could only trace it all the way back, is far more ancient than the lord's. Their rights are such as belonged to the members of a township or agricultural community long before the legal theory of manors and lords of the manor was heard of. Perhaps there are also parcels of Lammas land in the neighbourhood — fields which are enclosed and cultivated part of the year, and during the other part thrown open for the common use of the several occupiers, or (as is more likely to be the case) of a larger class of persons. Such arrangements are relics of the time when separate ownership of land was in its infancy. These and other ancient communal rights are often vested in the inhabitants of the parish, which may well be thought to represent a still older community, and to preserve in such lingering usages some traces of its original constitution. Examples of them have been I INTRODUCTORY 7 much diminished in the last few generations by the steady progress of enclosures, but they are still not uncommon. Between our imagined post and the waste land there lies a stretch of cultivated ground, occupied by one or more farmers. They may hold under leases for a considerable term of years, or only from year to year, but in the latter case they may, under favourable circumstances, enjoy a good deal of practical security in their tenure.^ In any case their legal condition is of a relatively modern and simple kind. It was barely pro- vided for in the economy of the feudal system, and is exempt from the mysteries of the law of real property. A leaseholder's interest in his farm is dealt with, in case of his death without disposing of it by will, in exactly the same way as his interest in the stock on the farm, or money in the funds, or any other movable property. In the language of the law it is personal, ' and not real estate. The complexities of which we have spoken affect him only through his landlord, in so far as they tie the landlord's hands in dealing with the tenant and improving the property. Nearer to us, again, at the foot of the park, is a little home-farm, kept in hand and managed by an agent on the lord's immediate behalf. This method of cultivation was the prevailing one for a considerable part of the Middle Ages. We now meet with it in England only as an exception ; sometimes it is a luxury, sometimes an ex- periment, sometimes the necessity of a bad season. Now let us turn to the park and the manor-house ^ During the recent and unhappily continuing period of depres- sion the difficulty has heen much oftener for landlords to keep their tenants. 8 THE LAND LAWS chap. itself, which may fill the foreground of our imaginary landscape. The lord of all this is himself a tenant, though not in the popular sense, or to much practical effect. His lands are held of the Crown, or perhaps of some other superior who himself holds of the Crown. His predecessors before the Commonwealth time owed rent or services, or both, and were subject to a variety of occasional dues and payments, some of them of a vexatious kind. They were bound to follow the king or other over-lord when he went forth to war, and bring with them a specified armed force, or pay for the maintenance of its equivalent. The feudal dues and services have been abolished ; but ancient money-rents, technically known by various names, and reduced to a nominal amount by the changes that have taken place in the standard of the coinage and the value of the precious metals, often survive to this day. Some ancient rents are not in money but in kind. The city of London stiU pays to the Crown certain horse-shoes and nails as the rent of a piece of land in the parish of St. Clement Danes, once granted by the king to a farrier, and a faggot as the rent of some waste lands in Shropshire.^ In some cases the Crown is entitled to receive some weapon of war, or part of warlike equipment — a sword, a banner, a pair of gloves, or spurs. Once or twice these ancient tenures, which were esteemed peculiarly honourable, have been imitated in modern times on the occasion of public grants for distinguished military services. But ^ These rents are now received by the Queen's Remembrancer a few days before the beginning of Michaelmas term. Tlie payment has for more tlian 160 years become merely ceremonial, the same horse-shoes and nails, preserved in the Queen's Remembrancer's office, doing duty on each occasion. I INTRODUCTORY 9 if the remnant of feudal relations to a superior is at this day no burden to the English landowner, and at most, adds a picturesque circumstance to his title, he is apt to be restrained in other ways of more modern and subtle invention. The lord of this mansion is named by all men its owner ; it is said to belong to him : the park, the demesne, the farms, are called his. But we shall be almost safe in assuming that he is not the full and free owner of any part of it. He is a " limited owner," hav- ing an interest only for his own life. He might have become the full owner, though still under a greater or less burden of encumbrances created by his predecessors, if he had possessed the means of waiting, the independ- ence of thought and will to break with the tradition of his order and the bias of his education, and the energy to persevere in his dissent against the counsels and feel- ings of his family. But he has had every inducement to let things go their accustomed way. Those whom he had always trusted told him, and probably with sincere belief, that the accustomed way was the best for the family, for the land, for the tenants, and for the country. And there could be no doubt that it was at the time the most agreeable to himself. As soon, or almost as soon, as he was of age to bind himself, he entered into a new settlement, by which his own interest was reduced, like his father's before him, to that of a life-tenant, and the succession of his offspring secured in advance down to the furthest limits allowed by the law. The legal machinery by which this is done is little more than two centuries old, and, though refined and improved in details by the ingenuity of generations of conveyancers, has not been much altered in substance since its first invention. 10 THE LAND LAWS chap. It owes nothing to legislation, except by accident. Thus, then, the apparent owner of the domain is no more abso- lute as to its actual disposition and management than the king is absolute in a limited monarchy. He can do but little of his own motion, and what he does is for the benefit of successors not of his own choice. Likely enough, he has no clear notion of his own powers and their limits. An English family settlement is on the whole less intelligible, and certainly less understood by most English citizens, than the English Constitution. In practice the limited owner has to put himself a good deal in the hands of experts, and oftentimes he is fain to make the family solicitor his prime minister. The advice he gets is pretty sure to be on the safe side — that is, on the side of not trying experiments. A family solicitor, un- like those who administer affairs of State, has no motive whatever for being enterprising in his client's affairs, and many to the contrary. He cannot hurt himself by over- caution, and may hurt himself much by rashness. So he takes, as a rule, the line of doing as little as possible, which is one much commended in all walks of life to those whose first object is their own peace and quietness. Peradventure the lord of this estate is lord in a strict legal sense — that is, as lord of a manor. We have, in- deed, assumed as much in our description. In this capa- city he is a kind of small sovereign prince, possessed of his own courts, and doing justice according to his own procedure and customs. But his powers and jurisdiction are shrivelled by the changes and chances of centuries into next to nothingness, and only the names of them remain. In the voluminous settlement which confers title to these lands on him and his issue, the manorial I INTRODUCTORY 11 franchises are enumerated in a roll of strange-looking terms, many of which are now obscure even to the lawyer, unless he is also a historical student ; there are English words among them of immemorial antiquity, which had their technical meaning centuries before the Conquest, and which the Norman lawyers, only half understanding them, thought it prudent to leave un- translated. They held their own through the invasion of Norman-French and Latin, and their native English hardly knows them again when it meets them. As for the lord himself, he knows neither the words nor their meaning unless he happens to be a scholar and an antiquary. Nor is he concerned to know them for any purpose of business. The ancient franchises and profits are obsolete, and have been so for many generations. The manorial courts exist in form, and their records are kept in the ancient fashion. But the fine by which a thief caught within the boundaries redeemed his life is no longer a source of revenue to the manor, neither does the lordship of " view of frankpledge and all that to view of frankpledge doth belong " convey any sensible increase to the wealth or the dignity of the modern landowner. To be lord of a manor is to be the lord of a secular ruin, in which he that knows the secret of the crabbed spell- book may call up the ghosts of a vanished order of the world. Thus we have taken a hasty view of the legal aspects of an English landed estate, which will presently come before us one by one for a more detailed survey. It is an unparalleled accumulation of layer upon layer of diverse materials. Tenure and convention, custom and competition, legislation and usage, the rude common 12 THE LAND LAWS chap. life of the free Teutonic warrior tribes, an aristocratic military system sprung from sheer necessities of mutual defence, and disguised in the terms and reasons of a Eomanised law, the subtle deductions of a legal profes- sion trained in scholastic disputes, the attempts of an impatient Parliament to make their crooked things straight, the not less subtle and more flexible inventions of modern lawyers, the partial clearances and half- hearted amendments of modern law-reformers : all these have gone to the making of the vast and inextricable mass, and all must be considered in their turn by the seeker who is bold enough to search out the history and the meaning of the land laws of England. It may be not amiss, meanwhile, to point out one or two of the general features in which the legal concep- tions of ownership and rights over land are at variance with the popular ones. It is commonly supposed that land belongs to its owner in the same sense as money or a watch. This has not been the theory of English law since the Norman Conquest, nor has it been so, in its full significance, at any time. No absolute ownership of land is recognised by our law-books except in the Crown. All lands are supposed to be held, imme'^ ' -^ly, or medi- ately, of the Crown, though no rent or services may be payable, and no grant from the Crown on record. The feudal lawyers forestalled to some extent in substance, and to a large extent in form, the modern Socialist dream of the State as the universal landlord. On the other hand, the law is equally far from countenancing the belief that there is land \ to the intermediary person or persons (there were generally several of them) to the use of the beneficial owner, and these persons, taking as they did the legal title, by the regular ceremony of feoffment (explained above, pp. 75, 76), were said to be enfeoffed, or to be feoffees, to uses. As to the beneficial owner, he was described by the uncouth phrase, cestui que use, he for ^ Spence, Equitable Jurisdictimi of the Court of Chaiicery, i. 440. 2 15 Rich. IL c. 5 (a.d. 1391). IV LEGISLATION AND TRANSFORMATION 95 whose sake the use is : no single word was ever found to take the place of this, and at the present day English lawyers have to speak of cestui que trust} Such were the terms most commonly employed ; but the word use was not (as for creating estates at common law certain appropriate words were) a matter of necessity. "Trust" or "confidence" would serve as well, or indeed anything clearly showing the intention. Assuming the feoffees to uses to be willing and faithful instruments of the beneficial owner, his advantages were great. Though he were involved in the civil strife of York and Lancaster, and dealt with as a traitor by victorious enemies, the land would be secured for his children ; for it legally belonged not to him but to the feoffees to uses, and therefore was not forfeited by his attainder. For the same reason nothing was payable to the over-lord on his death ; there could be no legal succession while any of the feoffees remained alive, and herein^ was the convenience of naming several in the first instance. The numbers might be kept up from time to time by new conveyances, as is the common practice to this day with bodies of trustees established for charitable and public purposes. Again, there was by law no power to give lands by will (except in some cities and towns by special custom) ; but the possessor of lands in use could without any formality at all give directions to the feoffees in his lifetime, or by testamentary declaration as to the 1 In the case of an estate for the term of a life or lives other than the tenant's -(as a lease for lives), the person whose life is named is called the cestui que vie. I know no other example of this construction. 96 THE LAND LAWS chap. enjoyment of them after his death.^ In this case the feoffees were said to hold the lands to the uses of his will. It was also possible to employ the method of conveyance to uses, and it was not unfrequently employed for the less laudable purpose of evading creditors. The debtor made over his lands to some friend on the understanding that he should still have the profits, and betook himself to one of the many sanctuaries or liberties where personal process could not be executed against him. " There were two inventors of uses, fear and fraud,'' said Coke in a summary phrase; Bacon adds, and rightly, the desire of larger powers of disposition than were known to the common law ; but fraud would, in Coke's eyes, perhaps include that as well. Down to the end of the foiu'teenth century or later the interests of the possessor in use were protected only by the honour of the feoffees, though he could ensure a certain measure of safety by being one of them himself, which was a very common practice. The ecclesiastical courts dared not interpose in a matter of conscience which so nearly concerned the title to land ; the king's ordinary courts could not recognise interests which those who created them had studiously put outside the scope of the coinmon law, for the very purpose of avoid- ing the risks and burdens of legal ownership. So com- pletely was an use of lands ignored that if the feoffee chose to treat the " cestui que use " as a mere trespasser, there was no legal defence to his action. Such a state ' Many examples of suoh declarations may be seen in The Fifty Earliest English Wills in the Court of Proiate, London, edited by Mr. Furnivall for the Early English Text Society. IV LEGISLATION AND TRANSFORMATION 97 of things could not go on indefinitely ; the quantity of land held by feoffees to uses increased year by year till it comprised the greater part of the realm, and the un- certainty of titles depending on mere private understand- ing became intolerable. The judges might in more than one conceivable way have given some indirect protection to the beneficiary ; '^ but recourse was had to the extra- ordinary jurisdiction of the Chancellor, a power still fresh, flexible, and ambitious. It was then held that the king was in a real sense the fountain of justice. His justice was not exhausted by the functions of the regular tribunals which guided it, so to speak, in the channels of its common application. There remained a supreme executive discretion which might be used on special occasions, a discretion which (according to modern doctrine at least) could not directly abrogate the common law, but might supply its defects or temper its application. This discretion was exercised through the Chancellor, and was the foundation of the whole system of the Court of Chancery ; the isolated petitions by which aggrieved subjects begged for its exercise were gradually moulded into regular pleadings, and the tran- scendent action of the king in his attribute of justice became the settled process of the Court. Thus the possessor of land in use complaining of want of faith on the part of the feoffees, would beseech the Chancellor to relieve him "for the love of God, and in the way of charity." The king's charity, when the Chancellor decided to exercise it, was masterful. Although the feoffee in trust could not be disturbed in his legal title and possession, yet, if he refused to dispose of the 1 Note F. H 98 THE LAND LAWS chap. enjoyment and profits in the manner that seemed to the Chancellor agreeable to good faith and conscience, he might be sent to prison for contempt of the king. Once habitually put in use, this power became by rapid steps a normal and systematic jurisdiction. The use of lands, as distinguished from the apparent feudal tenancy or "legal estate," became hardly less secure and a more convenient form of ownership. Broadly speaking, it resembled legal ownership in everything but the burdensome incidents ; and not only the original feoffees were bound by the confidence of their grantor, but so were persons taking the legal estate from them by purchase with knowledge of their fiduciary office, or by succession or other gratuitous title in any case.^ In other words, the Chancellors extended the conscien- tious obligation on which their jurisdiction was founded from the actual feoffee to uses to every one claiming title through him who had not honestly given value for the land in ignorance that his vendor was commit- ting a breach of faith. As Bacon said, " The Chancery looketh further than the common law, namely, to the corrupt conscience of him that will deal with the land knowing it in equity to be another's." These prin- ciples, though shifted in their application by the means and in the manner to be forthwith described, are not ^ I have thought it convenient, as the learned reader will see, to use " purchase " in the text in its popular sense. The man of business (if I have such a reader) may be helped by the analogy of the special rights and immunities allowed to the bona fide holder of a negotiable instrument ; the technical principles in the two branches of law are different, but the policy is the same. An ex- cellent summary of the law of Uses before the statute may be seen in Blackstone, ii. 330, 331. IV LEGISLATION AND TRANSFORMATION 99 obsolete ; the -working out of them has produced the whole structure of that part of the modern law of the Court of Chancery, technically called Equity, which regulates the vast amount of property, movable and immovable, held in trust throughout England. Thus the beneficial titles to a large proportion of the land in the kingdom were being drawn into the exclusive jurisdiction of the Chancellor. Notwithstanding the difficulty and expense that must have attended an application to the Chancery as com- pared with proceedings at the assizes, this jurisdiction was apparently popular. But the Crown and the great lords once more took alarm. In the fourteenth and fifteenth centuries a series of statutes made the " cestui que use " subject to certain liabilities, as if he were legal owner of the estates, and in one or two points gave him corresponding powers, but more for the benefit and security of purchasers than for his own.-' At last, in 1535, the Parliament of Henry VIII. passed "an Act concerning uses and wills," which has ever since been known as the Statute of Uses, and is one of the funda- mental and peculiar points of our modern law of real property. The intention was to abolish the system of uses altogether, and reunite the beneficial enjoyment of land to the legal estate ; it was likewise intended to abolish the power of disposing of lands by will, which had been introduced by the machinery of uses, and was 1 Blaokstone, ii. 332. ^ Our chief authorities for the intention of the Statute of Uses are the prea-mble of the statute itself, and the generally similar statements made by Coke in his report of Chudleigh's case (1 Rep. 113a), and 100 THE LAND LAWS chap. The statute ^ declared that " by the common laws of this realm lands, tenements, and hereditaments be not devisable by testament, nor ought to be transferred from one to another but by solemn livery and seisin " (feoff- ment), " matter of record " (judicial process entered on the records of the court, that is, fine or recovery), "writing sufficient made lona fide without covin or fraud " (this clause lacks construction, but is, I think, to be understood as explaining or summing up the two previous heads, unless possibly it means leases for years) ; nevertheless, the preamble proceeds, " divers and sundry imaginations, subtil inventions, and practices have been used, whereby the hereditaments of this realm have been conveyed from one to another by fraudulent feoff- ments, fines, recoveries, and other assurances craftily made to secret uses, intents, and trusts," and also by wills, formal or informal, " for the most part made by such persons as be visited with sickness, in their extreme agonies and pains, or at such time as they have scantly Bacon in his unfinished Reading on the Statute. I do not find it easy to make out in detail what either of these authors really means by the restoration of the common law which is supposed to have been the object. Bacon seems to take a more benignant view of Uses than Coke, who would have liked to treat the whole system as a nuisance to be rigorously abated. Without attempting to fathom all the learning of Chudleigh's case, one may pretty safely say that Coke would have looked with unmixed disgust on such inventions as the "name and arms clause" of a modern settle- ment ; and it is probable that such, among other things, were the dispositions which the statute was intended to frustrate. ' Revised Statutes, vol. i. p. 452. I have modernised the spell- ing, which throughout the statutes of Henry VIIL is in a stage of confused transition. Its "dyverse and sundry ymaginacions, subtile invencions, and practises " are amusing, but would disti'act a reader not familiar with the matter. IV LEGISLATION AND TRANSFORMATION 101 had any good memory or remembrance " ; all which tended to the loss of feudal dues, services, and for- feitures, to uncertainty of title such that " scantly any person can be certainly assured of any lands by them purchased, nor knowen surely against vi^hom they shall use their actions or executions for their rights, titles, and duties," and moreover to " manifest perjuries by trial of such secret wills and uses," and finally to " the utter subversion of the ancient common laws of this realm." And therefore, " for the extirping and extin- guishing of all such subtil practised feoffments, fines, recoveries,^ abuses, and errors heretofore used and accus- tomed in this realm," it was enacted that thereafter whoever should have an " use, confidence, or trust " in any hereditaments should be " deemed and adjudged in lawful seisin, estate, and possession " for the same estate that he had in use ; that is, that he should become, instead of the feoffees or trustees, the full legal owner (or, more exactly, feudal tenant) for all purposes, and that the absence of any actual delivery of possession to him should make no difference. Here this brief state- ment must suffice ; there are, in fact, subsidiary provi- sions of some length to work out and safeguard the general idea. The result was as follows. If Stan- ford made a feoffment to More, Fisher, and Brooke, to hold to the use of himself, the statute made this ineffectual. Before the statute. More, Fisher, and Brooke would have become the only tenants whom the 1 Observe that fines and recoveries are put on a level with feoff- ments as common and recognised forms of conveying land. It is only the ' ' subtil practised " conveyance to secret uses that is aimed at. 102 THE LAND LAWS chap. common law coui-ts could notice, and the only persons liable for the feudal dues (which, however, would mostly never become demandable), while the Court of Chancery would compel them to allow Stanford all the benefit of the estate. By the operation of the statute. More, Fisher, and Brooke would not become owners at all ; Stanford, by being named to take the use, would at once come into their place. He would be as much the legal tenant as before, and liable to all the legal burdens and incidents. A feoffment to John, or to John and William or to John and William and Peter, to the use of Peter, or in trust or confidence for Peter, was made by the statute equivalent to a feoffment to Peter. The use carried with it the legal estate ; in the curious technical phrase which has ever since been current in the books, the use was said to be executed in Peter by the statute. And the law thus made by the Statute of Uses is law to this day. Yet the statute ultimately failed in every one of its chief objects. It abolished wills of land (though, oddly enough, not in direct terms) ; but so unpopular was the restriction^ that a few years later (1540) an Act was passed expressly enabling tenants in fee simple to dis- pose by will of two-thirds of the land held by them in military tenure, and the whole of that held in socage. The statute was not improbably meant to put an end to new-fangled modifications of ownership by allowing no conveyance to take effect which would not have been ^ The repeal of the Statute of Uses was one of the articles of redress of grievances put forward by the leaders of the Pilgrimage of Grace in 1536. See the documents ap. Froude, Hist, Eng. iii. 91, 105, 158 IV LEGISLATION AND TRANSFORMATION 103 good at common law; ^ but it was interpreted by practice and in time by the courts as giving full legal validity to all interests which could formerly have been created by way of use, and would have been protected by the Chancellor. All that the jealousy of the common lawyers could effect was to saddle this new licence of disposition with certain qualifications of a highly techni- cal and irrational kind. Again, it was intended that there should be no use or confidence without legal ownership. But some ingenious person bethought him that the statute had provided only for one transfer or " execution " of the legal possession to couple it with the use. There was no denying that a conveyance to John to the use of Peter gave Peter the legal estate. But what if the conveyance was to John, to the use of Peter, to the use of Paul 1 It was not only argued, but decided and settled, that the bidding of the statute was satisfied, or as was then said, its operation was ex- hausted, in making Peter the legal owner. Paul had " a use upon a use '' ; this interest, it was held, was beyond the scope of the statute, and must be left, like uses before the statute, to the protection and manage- ment of the Court of Chancery. "An use,'' said the judges, " cannot be engendered of an use." ^ In like manner if Thomas, tenant in fee simple, conveyed to John and his heirs to the use of Thomas for his life, ^ Both Coke and Bacon appear to have thought so. The fact that the statute wholly omits to say anything which could fairly be so construed goes, at that time, for very little. ^ It is difficult to appreciate this reason. Perhaps it was made plausible by its analogy to the familiar (and then commonly received) argument against usury, that it is against nature for money to beget money. 104 THE LAND LAWS chap. and after his death to the use of Peter and his heirs, in trust to perform Thomas's last will, and then by his will declared that Robert and his heirs should have the land, — here after Thomas's death Peter had an estate in fee simple, and Robert had no title at all in a common law court ; he could only compel Peter in the Court of Chancery to let him enjoy the land, or deal with it according to his direction. In technical terms, Robert had an equitable estate, but no legal estate. Thus the Chancellor's jurisdiction, so far from being cut short, was fortified and enlarged ; and uses and trusts, instead of being " extirped," flourished all the more in a new form. The word trust became appropriated, as it still is, for these uses of the second order which the statute leaves untouched. We commonly hear and speak of lands being conveyed to the use of A in trust for B ; the dis- tinction is so fixed in practice and so convenient that nobody would think of neglecting it, but still it is matter of convenience only. " In trust for A " or " in confid- ence for A," followed by " to the use of B," would have exactly the same effect. Once more, among the mischiefs to be remedied by the statute was the secrecy of conveyances, which had been introduced through uses. The open ceremony of " livery of seisin " was to be restored to its full value, as pointing out the real owner ; and in those days of sparse population and little migration, when every man knew his neighbours and their doings, it was no doubt as good a means of securing publicity as any register could have been. But how did the statute operate 1 By one master- ful stroke, against all principle formerly recognised, it turned the possessor in use into a legal possessor ; IV LEGISLATION AND TRANSFORMATION 105 and there were other ways besides actual conveyance to feoffees in trust by which uses might be created before the statute. In particular, if Brooke agreed with Fitz- herbert to sell him a piece of land, and Fitzherbert paid the agreed price, it was the rule of the Court of Chancery that Brooke was bound in conscience not only to make a proper legal conveyance to Fitzherbert, but forthwith to let him have all the advantages of ownership. By the " bargain and sale " (such was the accustomed term) Fitzherbert acquired the use or equitable interest in the land. If, then, the bargain and sale were made after the statute came into force, Fitzherbert would get the legal interest also ; ho must be " deemed and adjudged in lawful seisin, estate, and possession." That is, the full legal ownership of land would be transferred from one man to another by acts which might be strictly private, need not be recorded in writing, and might be incapable of legal proof.^ So manifest a danger did not escape notice. It was seen that without fresh legislative precaution lands might pass from one to another " on payment of a little money in an alehouse," as one or two of the old books say. In the same year with the Statute of Uses an Act was passed to the effect that no estate of in- heritance or freehold, or any use thereof, should be conveyed by bargain and sale, unless the bargain and sale was made by deed, and the deed enrolled ^ within ^ At that time and long afterwards parties could not be witnesses in a court of common law. ^ An enrolment is an official copy made on a roll of parchment. Before 1849 the authentic text of Acts of Parliament was the en- grossment on the Parliament Roll : now a copy printed on vellum has taken the place of this. 106 THE LAND LAWS ohap. six months either in one of the courts at Westminster or in the county where the land lay. This was in- tended to provide, and did provide for some time, that land should be dealt with either by feoffment, by the still more solemn "matter of record," or by deeds publicly registered. But the invention of lawyers was at length too much for the precautions of Parliament. The Statute of Enrolments had nothing to say of estates less than freehold, such as a term of years. Interests for a term of years (of which more in their place under the head of Landlord and Tenant) had become familiar ; and probably the makers of the statute thought it un- necessary and impracticable to impose the formality and expense of an enrolled deed on farmers' tenancies. Here, however, was an unsuspected loophole. It was discovered that since the Statute of Uses a bargain and sale by Dyer (having the freehold), to Anderson, sup- pose, for the term of one year, put Anderson in the same position as if he had actually entered on the land ; for by the words of the statute he was to be " deemed in lawful possession." Now a tenant in possession could acquire the freehold by a simple deed (called a release) from the owner of the reversion. Livery of seisin could not be given to one already in possession, and the fact of possession (which by the old law implied an actual entry) was thought to supply the notoriety of a feoffment. In the present case Dyer, the day after making the bargain and sale for one year, would grant by release to Anderson, who was now in constructive possession by the Statute of Uses, his remaining interest or reversion in the freehold. Thus Anderson might become tenant in fee simple without any publicity at tv LEGISLATION AND TRANSFORMATION 107 all, and the Statute of Enrolments was evaded. This process was called a conveyance by lease and release, and was the common method of transferring freehold lands for more than two centuries. Doubts were for some time entertained as to its validity, but by 1620 the point was considered no longer open to discussion. Thus was the secrecy of modern English conveyanc- ing established. Its lines were fixed by the results of the Statute of Uses, and what legislation has done since amounts to little more than the simplifying of its formal elements. "A system of infinite subtlety, but answering, it must be acknowledged, most important purposes, has been framed upon this Statute, while most of the evils which it was meant to remedy remain."^ The frauds and other mischiefs which our ancestors, not without reason, apprehended have been in part provided against by the systematic development and refinement of the jurisdiction of the Court of Chancery, and in part by the ingenuity of counsellors. In the course of the seventeenth and eighteenth centuries conveyancers worked out a system of private investigation of titles which is still in use, and which, though exceedingly^ cumbrous and expensive, is fairly effectual. One can only say fairly effectual, for there remain possibilities of fraud which no ordinary precaution can exclude, and from time to time great hardship is thereby caused to persons who have laid out their money in good faith and have not failed in any point of due diligence. Other causes too were at work to break up the ' First Report of Real Property Commissioners, p. 8. ^ I should have said intolerably, but for the fact that land- owners have so long tolerated it. 108 THE LAKD LAWS chap. feudal scheme of land law. The military services were obsolete. Spain was teaching the world the power of disciplined standing armies, and the growing use of firearms brought in a need of copious and uniform munitions of war such as the old assize of arms ^ could not meet. Warlike equipment and organisation were henceforth more and more to be the immediate charge of the State. In this and other ways the feudal tenures had lost their original significance ; and the money pay- ments to the Crown and other lords appeared no longer as natural incidents of tenure, but as vexatious burdens on the full dominion of an owner. Even the fact that the decrease in the value of money had already greatly lessened such of them as were set at fixed amounts operated, we may well think, in the same direction. The same man who as tenant would cheerfully pay a substantial rent will, as soon as he regards himself as owner, resent the payment of much lighter but ap- parently casual and arbitrary demands. Then the besetting ambition of lordship, " an excess of will in men's minds, affecting to have assurances of their estates and possessions to be revocable in their own times, and too irrevocable after their own times, "^ was all the more stimulated ; nor did professional astuteness fail to devise the means of satisfying it. How the feudal tenants of England were definitely made real owners, and how the licence of disposition and posthumous control which the ' Sundry statutes and ordinances known by this name enjoined all freemen to keep in readiness a contribution of warlike weapons and equipment, whioli was graduated according to their rank. ^ Bacon, fading on tJie Statute of Uses (Works, vii. 409, ed. Spedding). IV LEGISLATION AND TRANSFORMATION 109 Statute of Uses seemed to make unlimited was, not by the baffled legislature but by a fresh exercise of judicial power, confined within certain though liberal bounds, shall be told in the following chapter. The Statute of Uses marks the close of the medieval epoch, and intro- duces a time of transition during which a new system works itself out, a system which after the Eestoration is continued without any notable break down to our time. A landmark so prominent and so memorable demands a pause before we pass on. CHAPTEE V DEVELOPMENT OF THE MODERN LAW The century that followed the passing of the Statute of Uses was a time of great legal activity in all directions. During the latter half of the sixteenth and the first half of the seventeenth centuries the books were pro- duced which, for most practical purposes, have long been regarded as the ultimate evidence of English Common Law. Coke's Commentary upon Littleton's Tenures, eminent in the charmed (and long since closed) ^ circle of "books of authority," was published in 1628. About the same time began the publication of printed reports of judicial arguments and decisions, a kind of publication which has gone on increasing till it has now swollen to a vast and unmanageable bulk. The advice of Coke to students, that " it is ever good to rely upon the book at large," has become a precept beyond human powers to fulfil, and the text-books of special subjects which digest and methodise the matter of the reports are themselves groaning under the burden. But in ^ Sir Michael Foster's Discourse of Crown Law (1762) was, I think, the latest addition. Blackstone, and some few later books, come very near, but only near, to being authoritative in the technical sense. CHAP. V DEVELOPMENT OF THE MODERN LAW 111 those days lawyers had more leisure. They found time to debate fully all the points of interest raised by a case, whether the solution of them was necessary for the actual decision or not ; and, with all their pedantry and occasional perversity of intellect, they held fast to a high and serious conception of the profession they followed, and their duty to it as a science and an art, which the modern practitioner must for the most part be content to envy. The law of real property got its full share of the discussion and development that went on in this period. Yet the results of the Statute of Uses were long in ripening. , Indeed the final touch was not put to them till the year 1833, on the very eve of the modern period of reform, when the limits of the " rule against perpetuities " were settled by the House of Lords. Practically, however, the structure of family settlements of land had assumed its modern form before the Eestoration, and the practice of con- veyancers became, from the seventeenth century on- wards, a fixed and well understood routine, improved in details from time to time by addition or variation, but preserving its main points unchanged. The general scheme of a " strict settlement " is of this kind. Let A be a living landowner who has a living son B, and wishes to keep his estate " in the family " as long as possible. For simpHcity's sake, we will suppose A to be tenant in fee simple to begin with. A convey- ance is executed by which A becomes tenant for life, and B tenant for life after A's death ; or else A by his will makes B the first tenant for life. Then an estate tail (generally in tail male) is given to the first son of B — a person not yet in existence, and who may never 112 THE LAND LAWS chap. come into existence ; and similar estates are given to B's second and other sons in succession. Each of these dis- positions can take effect only if every previous estate tail fails to begin, or comes to its natural end, by persons entitled to it not existing or ceasing to exist. Thus if the estates are in tail male, and B has three sons, the third son, or any of his descendants in the male line, can become tenant in tail only if the two elder sons and all their descendants in the male line, if any, are dead, and none of them has exercised his power of barring the entail — a power of which no tenant in tail can by any device whatever be deprived. There will follow, prob- ably, similar dispositions in favour of B's sons in tail general, so as to admit their descendants in the female line after those in the male line ; then other and some- what less elaborate ones in favour of B's daughters ; then, again, A's children other than B, if he has any, and their possible descendants,^ and also (in the case of a settlement made in A's lifetime) A's children who may yet be born, are provided for in a series of successive " limitations " of the same type as those already de- scribed. In the remote contingency of all these disposi- tions running their full course till every specified line of descent is exhausted, the ultimate "remainder" (see p. 81 above) is usually declared to be for A's " right heirs,'' that is, the person or persons who, at the time when the previous interests have been worked out, may answer the description of A's heir according to the common course of descent of a fee-simple estate. ' It is a question of detail, on wHoli the practice varies according to the desire of parties in the particular case, whether the descend- ants in the male line of the settlor's younger children are postponed or preferred to descendants in the female line of the elder ones. V DEVELOPMENT OF THE MODERN LAW 113 This may seem pretty complicated, but it is the least complicated part of a modern settlement. Practically it is only in rare cases that the remoter interests so carefully mapped out ever come into possession. If all things happen as is desired, the regular course is this : — B becomes tenant for life on A's death. He has a son C (born either during A's life or afterwards, it matters not which), who is " tenant in tail in remainder." English lawyers do not say that he mil he tenant in tail when the life tenancy comes to a end, but that he is tenant in tail subject to B's life estate. Suppose that C comes of age in B's lifetime. He can now without B's consent bar the entail as against his own issue, but not as against those "in remainder"; in other words, he can neither sell nor borrow to much purpose, for a title depending on the continuance of his descendants, which is all he can give to a purchaser or lender, is a precarious and specula- tive thing. ^ This check on C's power of alienation is a result of the technical necessities of the old " common recovery," which could not be carried through to its full effect without the aid of the person in actual posses- sion of the freehold. When fines and recoveries were abolished in 1833, a new but sustantially equivalent check was provided by the Legislature, the purpose of ' The curious kind of estate created by the conveyance in fee simple of a tenant in tail not in possession, without the concur- rence of the owners of estates preceding his own, is called a base fee. Though uncommon, it is not unknown in practice ; and it has been used by George Eliot in Felix Holt, with great effect and with perfect correctness, as part of the machinery of the plot ; insomuch that conveyancers reading the novel have been known to lament seriously, as if the thing had happened to one of their own clients, that the parties did not take better advice. I 114 THE LAND LAWS chap. that measure being only to simplify the form of proceed- ing and put an end to fictions and useless expense. Since that date B would be " protector of the settlement," and his consent to any disposition by C is an express and direct condition for its validity against the ulterior interests created by the settlement. B then stands in a commanding position towards C ; B and C between them are masters of the estate, while C alone can do very little, and B alone — except in the extraordinary contingency of all the subsequent limitations failing — can do nothing beyond the term of his own life. And when B proposes that C, in consideration of being adequately provided for during the rest of B's life, shall join in making a new settlement in which C shall be only a tenant for life, and other successive limited interests shall be laid out in the same fashion as in the former one, but with C and C's issue in the place of B and B's issue, there is not much probability that C can or will refuse. This process is repeated, as occasion serves, from generation to generation ; and so long as it can be kept up the estate is never in the hands of an absolute owner.^ Hence arose the necessity for the further complicated machinery which, as already hinted, must be added to these dispositions : I speak for the moment without regard to recent legislation, of which more in its due place afterwards. If things stood merely as now de- ' The process of settling and resettling famUy estates lias re- peatedly been described in forms intended to be more or less intel- ligible to the lay public. I have purposely -written my own account without referring to any other ; not that I can hope to be more successful than my predecessors. V DEVELOPMENT OF THE MODERN LAW 115 scribed, there would be no provision made out of the estate for the younger branches of the family (save in the remotely contingent event of some of them coming into possession of the whole), and no means of making any. Neither would there be any means of effecting a sale or exchange of any part of the settled estate, grant- ing leases, or exercising many other necessary and usual functions of an owner, during the minority of a tenant in tail. And here the wide and flexible applications of the Statute of Uses (undreamt of by its authors) come in. It would be possible, indeed, without the aid of the statute, to effect most of the purposes of a modern settlement, if not all, by conveying the whole estate to trustees in the first instance, who would, like feofi'ees before the statute, have the legal fee simple, or, in other words, would be the only persons whom a court of common law would recognise as owners. Such a course might nowadays be viewed with equanimity by lawyers and landowners, and in fact it is sometimes adopted in settlements made by will. But two centuries or even one century ago there was still a kind of magic in men's apprehension about the " legal estate.'' It is still true in strictness of theory that an " equitable estate," as we call the interest of a person beneficially entitled under a trust, is not perfect ownership — a right available against all the world — but is only a right to claim the profits and enjoyment of ownership from the determinate person who is the trustee. The beneficiary is in the same position as the old " cestui que use " before the Statute of Uses. Increased security has been given to this kind of interest by the full but gradual development of the jurisdiction and jurisprudence of the Court of Chancery; 116 THE LAND LAWS chap. the trustee's obligation being extended to all persons deriving or claiming title through him otherwise than for value, in good faith, and with due diligence of inquiry. This process was still in its youth at the time of the Eestoration, and it may well be thought that nothing less than the power given by the Statute of Uses to devise complex modifications of the legal ownership itself would have sufficed to strengthen the hands and embolden the invention of Orlando Bridgman and his contemporaries.^ Broadly speaking, there are two kinds of objects for which the special powers and provisions now in question are introduced : the benefit and support of wives and younger children of the successive holders of the estate, and the general management of the estate for the advantage of all parties interested in its profits. These objects are effected partly by the action 'of the person for the time being in possession, partly by the intervention of trustees named for that end in the settle- ment itself, and through an ingenious and intricate machinery which is far too artificial to be described here. It is enough to say that the machine has been brought to great technical perfection by the experience of about two centuries, and, when constructed by a skilful artist who takes advantage of all known improve- ments, works with much less friction than might be expected. On the other hand, if the workmanship is not first-rate to begin with (and for various reasons — ^ Orlando Bridgman, who gained great fame as a, conveyancer before and during the Commonwealth time, and was afterwards Lord Keeper, is credited with the invention of the ' ' strict settle- ment " in substantially its present form, avoiding certain risks to which it had before his time been subject. V DEVELOPMENT OF THE MODERN LAW 117 haste, ignorance of country practitioners, or what not — it often falls short of this standard), very serious trouble may ensue. With regard to the special trustees just mentioned, it must not be supposed that they exercise much discretion of their own. As a rule, they are little more than wheels in the machine, the management of which is in the hands of the family solicitor, fortified at need by the opinion and advice of counsel. It would be considered meddlesome if they offered to exercise their powers before they were asked. At the same time they are fully answerable for what is done by them on the usual request and advice ; and they are mostly persons of such weight and standing as to be above suspicion of conniving at any reckless or improper dealings with the settled estate. In short, the trustees of a family settlement are something like the constitutional safe- guards of a complex political system ; their presence is, in ordinary circumstances, hardly perceived, but they hold great powers in reserve, which may be used with effect on an emergency. The parallel just now suggested is not a casual one. There is nothing, perhaps, in the institutions of modern Europe which comes so near to an imperivm in imperio as the settlement of a great English estate. The settlor is a kind of absolute lawgiver for two generations ; his will suspends for that time the operation of the common law of the land, and substitutes for it an elaborate con- stitution of his own making. These constitutions are in fact all modelled on the same, or very nearly the same, type ; for the desire of great landowners has constantly been to make the strictest settlements which the law would allow, and the law, as we shall immediately see, 118 THE LAND LAWS chap. has set bounds, though liberal ones, to the power of fettering inheritances and suspending absolute owner- ship. And the ingenuity of conveyancers, devising how to satisify private ambition to the utmost within the field left clear to it by public ordinance, has produced that curious and exquisite structure which, a hundred years hence, will probably be as much abandoned to the care of a few legal antiquaries as the learning of remitter and collateral warranty. But a correction is needed when we speak of the family settlement as a constitu- tion ; for though it resembles a modern political system in its multiplication of checks and counter-checks, and in the tacit or half-expressed understandings by which its working is made practicable, there is another respect in which it is more fitly compared to the customs of an Oriental despotism. When the authors of a settlement have once made their disposition, it is like the law of the Medes and Persians rather than the acts of a modern legislature. They are as powerless as the great king of old to alter their decrees, unless they have made special provisions for that purpose which are not usually made in practice, and which indeed would be inconsistent with the main object of making the estate as a whole as nearly inalienable as possible. It might be a topic of curious meditation for the student of comparative jurisprudence to note how well the English landowning families have striven, though all unconsciously, to reproduce in our modern society something like the image of an archaic Aryan household. There is at the head of all the eldest living male ancestor, revered by his descendants and inferiors, and exercising great control and influence in divers ways. V DEVELOPMENT OF THE MODERN LAW 119 and yet with his power over the family property strictly limited — in the ancient household by custom, in its modern counterpart by accustomed convention. There are the women and younger generations of the family, subject to the house-father in one sense, and yet having rights in the common inheritance which he cannot inter- fere with. The free dependents, not members of the family, who swelled the state and substance of the ancient household, may well enough be represented by the modern tenant-farmers. Only the highest and lowest extremes of the community are wanting. Slavery and serfdom are happily long since extinct in our land ; and so are the rites of the domestic altar, which were for the ancient Aryan family the bond of fellowship, the assurance of the departed ancestors' welfare, and the pledge of their favour to the living. Some brilliant theorists have sought in these the very origin of pro- perty and inheritance ; and to this day the worship of household gods and ancestors is in full force in India, and lies at the foundation of the law which English judges administer to the Queen's Hindu subjects. The most fearful imprecation by which a Hindu can bind himself is that, if he speak falsely, he may have no son to deliver his soul from hell by due performance of the sacrifice. But hardly can a faint survival of such old- world usage (though much of the abiding human passions which only change their garment from age to age) be traced by the modern observer's fancy in an English estate. He may speculate on a remote connection of Pitris and Lares with the domestic chaplain, the an- cestral monuments in the parish church, or the family pew. Or, when he notes who is now the real possessor 120 THE LAND LAWS chap. of the secrets of the estate, the real familiar spirit at whose bidding the magical powers of the settlement are called forth, and without whose aid no matter of weight can be undertaken, he may peradventure dream that " the disestablished Lar," to use the term of a late ingenious writer,^ is not dead but transformed, and lives embodied in the family solicitor. We have said that the modern form of settlement dates, roundly speaking, from the Restoration. But already in the sixteenth century, when controversy was keen about Uses and the effect of the great statute, lawyers had an inkling of what was coming. And their opinions of the invidiousness and social inconvenience of strict settlements (or even of dispositions falling far short of what we now understand by the term) were such as would not many years ago have been called Radical. Coke has been quoted in the last chapter. Let us hear Bacon : — " First [it is said] that jit is a wisdom and foresight for every man to imagine of that which may happen to his posterity, and by all ways establish his name. To this I answer that it is a wisdom, but a greater than even Solomon aspired after, who had a large heart, as the Scripture saith. For I find that he uses other language, when he says that he must leave the fruit of his labour to one of whom he does not know if he shall be a fool or a mse man. And yet does he say that he shall be an usufructuary or tenant restrained in a perpetuity ? No ; but the absolute lord of all that he had by his travail. So little did he know of these establishments." . Whether Solomon had any land by his travail, in the ^ Hearn, The Aryan Hoiiseliokl. V DEVELOPMENT OF THE MODERN LAW 121 modern sense of having a thing in absolute ownership, we need not stop to inquire. As king he was no doubt, in a sense, lord of the soil of his kingdom ; but, if he was like other eastern kings, he could not lawfully dis- possess the meanest of his subjects who duly paid the customary revenue. Private ownership of land, at any rate, has been of slow growth in the East. Under the ancient rule of custom which is slowly decaying before our eyes in India the land belonged to the village or to the family, and was inalienable, though the possession and enjoyment might change hands within the circle of joint owners. The Hindu joint family is the true and ancient perpetuity. Later attempts at founding per- petuities by way of entail, mmjorat, strict settlement, and the like, are the rebellion of a privileged class against the tendency of the civilised world towards absolute and several ownership, even as the schemes of Socialists and other visionary reformers for "Land Nationalisa- tion," or whatever else it may be called, are the rebellion of those for whom the education of the world goes too slowly, and who vainly seek a charm that shall redress in one instant the balance of life between rich and poor. But Bacon has more to say to us. Let us hear him further : — " Some young heir when he first comes to the float of his living outcompasseth himself in expenses, yet perhaps in good time reclaims himself, and has a desire to recover his estate ; but has no readier way than to sell a parcel to free himself from the biting and con- suming interest. But now he cannot redeem himself with his proper means, and though he be reclaimed in mind, yet can he not remedy his estate. 122 THE LAND LAWS chap. " So, passing over the considerations of humanity, let us now consider the discipline of families. And touching this I will speak in modesty and under cor- rection. Though I reverence the laws of my country, yet I observe one defect in them ; and that is, there is no footstep of the reverend potestas patria which was so commended in ancient times. . . . This only yet remains : if the father has any patrimony and the son be disobedient, he may disinherit him ; if he will not deserve his blessing, he shall not have his living. But this device of perpetuities has taken this power from the father likemse and has tied and made subject (as the proverb is) the parents to their cradle, and so not- withstanding he has the curse of his father, yet he shall have the land of his grandfather." ^ The term perpetuity, as here used by Bacon, and constantly used by modern lawyers, signifies a dis- position which attempts to make property inalienable beyond certain limits fixed, or conceived as being fixed, by the general law. What, then, are those limits ? It was long before they were ascertained. In Bacon's time, and for a century later, the legal question was still open to be argued on wide grounds of policy. There was no obvious limit to the time within which trusts, or dispositions operating under the Statute of Uses, might be calculated to take effect. Yet it was felt that a limit there must be. It was agreed that " the rules of the law to prevent perpetuities are the polity of the king- dom," but nobody knew what the rules were. A long series of experiments, extending into the second quarter ' Arguments in Chudleigh'a case. Works, vol. vii. pp. 632-35, ed. Spedding. V DEVELOPMENT OF THE MODERN LAW 123 of the eighteenth century, determined the limits of safety for founders of families. Yet another century passed before the rule was finally settled in all points. In order to understand the form which it took, we must go back to the older law. Quite apart from the learn- ing of Uses, it was possible, at any time after the statute De Bonis} to grant by one and the same conveyance an estate for life, say in the manor of Dale, to Markham ; and, subject to his life interest, an estate tail to Martin ; after the determination of that estate another estate tail to Newton, and so on through any number of persons, till some one was named to take an ultimate fee simple, after which there was nothing more to dispose of. Martin, being of full age, could at any time, but not without Markham's concurrence while Markham's life estate was in existence, " suffer a recovery,'' in the manner explained in the last chapter, and thus acquire and dispose of the fee simple, when all the subsequent dispositions made by the original grant would go for nothing. So far, then, the manor is inalienable until Markham's life estate ceases, or Martin, if an infant at the date of the original conveyance, is twenty-one years old, whichever of these events happens latest. But it was not necessary that Martin, a person in existence, should be named as the successor after the first life estate. After some doubt ^ it was settled that an estate for life might be given to Markham, and an estate " in remainder " for life, in tail, or in fee, to Markham's first son, though at the time of the gift he had no son. The ' The legal reader will remember that a remainder over could not be limited upon a fee simple conditional at common law. 2 Digby, History of Laic of Real Property, c. 5, § 3, 124 THE LAND LAWS chap. series of dispositions which form the groundwork of a modern settlement, as above described, are only the full application of this principle. But it was not, and is not, allowable to give an estate for life to Markham's first son, after that another estate for life to the first son of that son, and so following — by which means, if allowed, the estate might have been made inalienable for an indefinite time. The title could not be secured for Markham's descendants in the male line further than by giving an estate tail to his first son ; nor indeed was the security perfect even so far, for if no son was in existence at the date of the settlement it was still in the father's power to destroy the " contingent remainder,'' which, when a son was born, would become "vested" in him and beyond the father's control. Suppose a son born and his interest unimpaired. That son might possibly be born in the last year of Markham's life, or even some months after his death. In that case twenty-one years might elapse from Markham's death before any one had power to alien the fee simple of the manor. But whenever a tenant in tail of full age is in possession, he can dispose of the fee simple notwith- standing any attempt made by the terms of the settle- ment to prohibit or restrain him. So it comes to this, that the furthest period until which alienation can be restrained by settlement, apart from provisions taking effect by way of use or trust, is the end of twenty-one years, and exceptionally something more, after the death of some person living at the date when the settle- ment is made. Now uses and trusts were not directly amenable to these rules, Being originally nothing else than direc- V DEVELOPMENT OF THE MODERN LAW 125 tions to the " feoflfees to uses " ^ to allow the profits of the land to be enjoyed by persons named by the feoffor, or to be ascertained in the manner pointed out by him, which directions were enforced by the Chancellor as binding on the feoffee's conscience, they had nothing to do with the rules of the common law as to the creation of estates. These newly -modelled interests might, under cover of the feoffee's legal possession, be made to shift about in the most capricious fashion. By the common law Fortescue could not give an estate to Brian and his heirs so long as they should use the name and arms of Fortescue, and if they did not, then to some one else. But when uses were established For- tescue might infeoff Moyle and his heirs to the use of Brian and his heirs until any of them should cease to use the name and arms of Fortescue, and then to the use of Prisot and his heirs. And as far as theory went, the contingency might have been something wholly unconnected with the estate and beyond the parties' control — for example, until Heme's oak in Windsor Park should fall, or until a new Pope should take the name of Leo. There were other contrivances, more complex in form but similar in principle, by which like results could be produced for a variety of purposes.^ i 1 P. 94, above. ^ "Uses of this class,'' said a learned and accurate modern writer (William Hayes, Elementary View, etc., 1840, s. 31), "may be described aa altogether eccentric ; as deviating from the course prescribed by the original laws of the system, not without con- siderable disturbance of its ordinary operations." It is said that by common law "a man may have an inheritance in fee simple in lands, as long as such a tree shall gi-ow, because a man may have 126 THE LAND LAWS chap. Neither the judges who had broken through the ex- press terms of the statute De Bonis, nor the successors who inherited their tradition, could be expected to tolerate this new-found way of fettering the possession and inheritance of lands. And, in fact, there was not (what one might look for) a crowd of daring attempts to push forward to all lengths in that course, but settlors of estates rather felt their way step by step till it was known how far they might venture ; so little doubt was there that the Courts would once more hold themselves authorised by the general "policy of the law " to make new law for a new mischief. The rule was at last settled by analogy (an analogy more felt than expressed) to the course of the common law. Not without sundry fluctuations of both opinion and judicial decision, it came to be understood, towards the middle of the eighteenth century, that a disposition of any kind of property ^ intended to take effect in the future must be so framed as necessarily to take effect, if at all, after a lapse of time not exceeding twenty-one years from the death of some certain person, or of the sur- vivor of certain persons, who are living at the date of the settlement if it is made by deed, or at the date of the testator's death if it is made by will. As late as 1833 it was decided by the House of Lords that these twenty-one years need not have reference to the infancy an inheritance in the tree itself" (Liford's case, 11 Co. Rep. 49a). This raises a very curious and difficult question, on which the learned reader is referred to Note G. In any case, a, remainder over could not he limited after such an estate. ^ The rule is not confined to land, though we now have to do with it only in that application. V DEVELOPMENT OF THE MODERN LAW 127 of any person. In particular cases the rule of law which treats an unborn child as a person in existence for all purposes of his benefit may have the effect of adding some months to the time allowed. A disposition which goes beyond these limits is said to be "bad for perpetuity." Thus a gift to the first child of A (having at the time no children) who shall attain the age of twenty-one years, or to all his children who shall attain that age, is valid. A gift to the first son of A who shall attain the age of twenty-five years is of no effect, and it makes no difference that A may in fact have a son who attains that age in his own lifetime. Again, a gift by will to the first person who shall climb up the cross of St. Paul's after the testator's death is " void for remote- ness,'' though somebody may peradventure do the thing within a week ; if the gift were to the first person who should so climb within twenty-one years after the testator's death, it would be good. The rule, it must be well marked, does not prevent an estate in remainder or reversion from being made expectant on an interest which may, or which naturally will, endure for a much longer time. For English lawyers regard an estate in remainder or reversion not as a future interest, but as a present interest subject to the " particular estate " which together with the re- mainder or reversion makes up the fee simple. To illustrate this by common cases : — If Holt grants an estate tail to Powell, with remainder to Finch in fee, Finch has the fee simple subject to Powell's estate tail, and can dispose of his interest at once, though it is un- certain when it may take effect by the failure of Powell's heirs under the entail, or whether (since Powell or any 128 THE LAND LAWS chap. of his successors can bar the entail and acquire the whole fee simple) it will take effect at all. Holt will be in the same position if, instead of granting the fee simple subject to the estate tail as a remainder, he keeps it as his reversion ; he can at once deal with it for what it is worth as he thinks fit. Again, Bayley grants a lease to Buller for ninety-nine years. If Bayley tries to provide that at the end of the ninety-nine years the person who is then his eldest living male descendant shall have the land, this is " void for remoteness," as the creation of an interest to take effect after the lapse of more than twenty-one years. But if he leaves it alone, he has a perfectly good reversion by the common law, which he may dispose of now or hereafter as he thinks fit, no less than if it were land in his actual possession. If he sells, it is commonly said that he sells the ground-rent ; but in legal conception and language he sells the land itself subject to the lease ; and if the lease were for a thousand years, the legal doctrine would be the same. Such is in outline the " rule against perpetuities," which, as regards these general principles, is now a settled part of our law. Many curious questions might still be raised on the application of it in various cir- cumstances ; but the skilled and cautious routine of conveyancers is for the most part, though not invariably, successful in keeping clear of them. Certain points in the legal theory are also still unsettled, or have been settled only in the last few years, but it would be un- profitable to explain or discuss them here.^ The rule prevents property from being made certainly ' See Note G, on Settlements and Perpetuity. V DEVELOPMENT OF THE MODERN LAW 129 inalienable beyond the limits of time fixed by it. But it would not be correct to say that the property must therefore be in fact alienable when the restrictive dis- positions have run their full course. There must then be some person absolutely entitled, but that person may be an infant or a lunatic, so that yet another generation may elapse before there is an owner with full and active power of disposal. The questions thus decided arose for the most part upon dispositions by will. The Statute of Uses, as we saw, was intended to make the devise i of land by will thenceforth impossible ; but the prohibition gave so much offence as to be soon removed, though for the time to a limited extent only, by the Statute of Wills of Henry VIII. (p. 102 above). And it was held that, dispositions by will being once allowed, either the Statute of Uses or its analogy (it matters not which) was applicable in their favour. Thus a man could, and he still can, by his last will, as well as by acts executed in his lifetime, create interests unknown and indeed repugnant to the doctrine of the old common law. Persons making wills are notoriously more adventurous and capricious than those who make settlements to take effect in their lifetime ; in part, it may be, because the obscurity or perplexity of their directions can vex only posterity ; moreover, they are less under the influence 1 A gift by will of freehold land, or of such rights arising out of or connected with land as are by English law classed with it as real property, is called a demse. A gift by will of personal property (a term including leasehold as well as movable property) is called a leqxiest. Dispositions of real property of a kind not conforming to the rules of the common law before the Statute of Uses are called, when made by will, executory c K 130 THE LAND LAWS chap. of discreet advisers. Hence dispositions by will pro- duced much the greater part of the litigation which defined the bounds of legal control over future genera- tions. Complete freedom of dealing with land by will was a result of the abolition of military tenures, an event which preceded in time the settlement of the rule against perpetuities, and may be taken as marking the full close of the medieval stage of the law. Though in form it stands as one of the first acts of the Restoration, in substance it was one of the many law reforms under- taken in the time of the Commonwealth. Most of those reforms were rejected or dropped at the Eestoration. This one was too popular and necessary to be so dealt with ; nor was unexceptionable authority wanting, for a similar measure had been strongly advocated by Coke.^ The Act of the first Parliament of Charles II. for abolishing the military tenures and their incidents was passed in 1660. No notice was taken by it of the previous Act of the Commonwealth which had been passed in 1656, all statutes of the Commonwealth being treated under the restored monarchy as made without lawful authority, and mere nullities ; but the same pro- visions were enacted in a rather more elaborate form.^ All freehold tenures were reduced to the one type of " free and common socage," ^ with an important twofold ^ See the details and references in Mr. Digby's History, eh. ix. It is a gi-eat error, but it has been a common one, to overlook these facts, and regard the legislation of 1660 as an arrangement between Charles IL and the great landowners. ^ According to Madox {Rist. Exch., i. 620, ed. 1769), the work- manship of the statnte leaves much to be desired, ^ See pp. 59 and 63, above, V DEVELOPMENT OF THE MODERN LAW 131 result. First, all the vexatious incidents of military- tenure disappeared with the tenure itself ; only ancient money rents might remain payable by the tenant, which already had become, by the changes in the value of money since they were fixed, almost or altogether nominal. Next, inasmuch as the statute of 1540 had enabled tenants in fee simple to dispose by will of the whole of their socage lands, and socage was now made the only freehold tenure, the whole of the fee-simple land in the kingdom became disposable by will. Feudal tenancy was converted for all practical purposes into full ownership. The "honorary services of grand serjeanty," that is, services due to the Sovereign in person by the terms of the tenure, were exempted from abolition ; and tenure in frank-almoigne, by which the greater part of ecclesiastical lands were and still are held, was left untouched.^ It had been proposed at an earlier time to compensate the Crown for the loss of feudal dues by assessment of a fixed money rent on the enfranchised lands.^ This method was not now adopted, but the excise duties which had been invented by the Long Parliament and renewed under the Commonwealth were granted to the King, " to the intent and purpose that his Majesty, his heirs and successors, may receive a full and ample recompense and satisfaction " for the feudal incidents, and for certain other profits and privileges abolished by the same Act. There is a not uncommon impression, I believe, that the land tax was imposed as part of, or in ^ This, however, was not a real exception, for every non-military free tenure is tenure in socage. 2 Coke, 4 Inst. 202. 132 THE LAND LAWS chap. connection with, the same transaction, and to make up to the public revenue for the income formerly received from the military tenants of the Crown. But this is not so ; we first hear of the land tax in 1692, and then as part of a scheme for a general property tax.^ The Land tax itself has now shrunk into relative insig- nificance beside Schedule A of the Income tax, which in our time is the real and effective tax on landed property for national as distinct from local purposes. It is a true property tax, being, unlike the tax on profits under Schedule D, assessed on the gross, not on the net value. And if any one thinks land does not bear its fair share of taxation, and wants to raise the question in a practical form, his proper course would be to propose a differential rate under Schedule A. But it is easier, and for some purposes more profitable, to put forth sounding generalities than to work out a definite plan.^ There were other branches of law relating to land which the development of the seventeenth and eigh- teenth centuries brought into substantially their present condition. Of these the law of mortgage is the most important. The power and practice of making a debtor's property, and especially immovable property, a security to the creditor for the payment of his debt, are well-nigh as old as the legal recognition and judicial enforcement of any rights of property whatever. The 1 Appendix L to Mr. G. Brodriok's English Lmid and JUnglish Landlords (by Mr. Humphreys-Owen). ^ A landowner who keeps land in hand pays not only as owner under Schedule A, but also as occupier under Schedule B, whether he occupies for proiit or not. I doubt whether this is generally understood. V DEVELOPMENT OF THE MODERN LAW 133 reflection is trite that both parties gain — the creditor by having security, the debtor because a secured creditor can let him have money on easier terms. Good security means low interest as certainly as high interest means bad security. Certain reformers who wish on economical grounds to abolish mortgage alto- gether must take, it seems to me, an extremely sanguine view of the facility of making a radical change in the convenience and the desires of mankind. The forms, however, in which English law has given efifect to this all but universal practice have been singularly ill chosen. Beginning with fictitious and impracticable stringency, our practice has ended in a wide and dangerous laxity, which breeds doubtful titles and litigation, and is no small encouragement to fraud. In the medieval period we meet with two ways of giving land in security. One of these consists in handing over possession to the creditor, who repays himself out of the rents and profits. This is called in Latin vivwm vadmm, in French mf gage, because the pledge in the creditor's possession is, as it were, alive for the debtor's benefit in working off the debt. In English it is called a Welsh mortgage ; in modern practice it seldom or never occurs. The other and prevailing method is a conditional sale of the land by the debtor to the creditor. If the debtor repays the money with interest to the creditor at a stated time, the creditor must give him back the land; if not, it remains the creditor's. And this is properly called mortgage, mortuum vadium, in opposition to the "vif gage," where the growing profits go to extinguish the debt, because the profits of the land are, as it were. 134 THE LAND LAWS chap. dead to the debtor. It must be difficult for any one but a lawyer to believe that so clumsy an operation is to this day the regular means of securing a debt on land in England. It is true that its harsh meagreness was amplified by the judgments of the Court of Chancery into a full and elaborate system of rules, abounding even to overmuch caution in safeguards for the debtor. The terms of the transaction were — as they still appear to be — that the debtor must pay his money to get back the land (redeem it, in the technical phrase) at a stated time, generally six months after the date of the agreement, or it would become the creditor's absolute property. But the Chancellors laid down that, notwithstanding these terms, the debtor should be ad- mitted to redeem after the set time had past. The right thus allowed him in the Court of Chancery was called an " equity of redemption." In other words, the jurisprudence of that Court, looking to the substance of the dealing between the parties and disregarding the form, treated the borrower as still the real owner of the land, and the land as only pledged for the lender's security. But a corresponding right had then to be given to the lender, his nominal right to deal with the land as owner being made useless. He might indeed take possession when the debtor failed to pay on the appointed day, but he did so at his peril; the Coxu-t held him strictly accountable to the debtor, so that the plight of a mortgagee in possession ^ is one of the most unenviable known to the law. Therefore a more just and convenient redress was given him. He was allowed, ^ The borrower who pledges his land by way of mortgage is called a mortgagor ; the lender who holds it in pledge is a mortgagee. V DEVELOPMENT OF THE MODERN LAW 135 if he could not get payment of his debt after due notice, to sue in Chancery, that the borrower might be ordered to pay the principal and all arrears of interest, or in default be "foreclosed of his equity of redemption," that is, no longer protected against the strict legal consequences of his agreement according to its form. This, however, was found expensive and tedious; and it became an almost universal practice to insert in the deed special provisions enabling the mortgagee to sell the land and repay himself out of the proceeds, if the debt were not paid off on notice, or if the interest fell into arrear for more than a certain time. Recently this " power of sale " has been conferred on mortgagees by Act of Parliament, so as to supersede in ordinary cases the necessity of expressly repeating the provisions which had become well settled by usage. Power has also been given to the Court to order a sale of the property instead of foreclosure. Shortly, the result is this : — A deed of mortgage means something very different from what it says, but this has so long been an understood thing that nobody is misled by it. The mortgagee never expects to be repaid on the day nominally appointed for redemption (not uncommonly, indeed, there is an express stipulation that the principal shall not be paid off till after a certain number of years), and the mortgagor's possession is safe as long as he pays the interest punctually. But the Court of Chancery embarked on other adven- tures, and discovered new methods of encumbering land. It was not enough to remodel the old mortgage into accordance with present convenience and the real expec- tation of the parties. Nor was it enough to hold that a 136 THE LAND LAWS chap. beneficial owner who had not the " legal estate " might raise money on the security of his beneficial interest in the same manner as a legal tenant. It was held, first, that an agreement to give a mortgage was not a mere personal agreement, but operated as a charge on the land itself in the hands of any one who knew of it ; and then that an " equitable mortgage " might be created without any written agreement at all. The latter step came about thus : — As the first thing required of an English landowner by any one dealing with him for the land is to produce his title-deeds, the handing over of the title-deeds to a creditor, pending the preparation of a formal security, is a fairly effectual precaution for the creditor's safety. When the Court of Chancery took notice of this practice, it became, and it still is, the established doctrine that such a deposit of deeds consti- tutes an " equitable mortgage " of itself. This doctrine being founded on the supposed equitableness of executing the agreement between the lender and borrower, and not on any virtue of the title-deeds in themselves, it does not matter whether the documents actually deposited are the whole of the debtor's evidence of title, and it even seems that it does not matter whether they are title-deeds at all, so long as they are stated so to be, and accepted as such by the creditor. A clever and un- scrupulous borrower may thus deposit part of his title- deeds with one creditor, part with another, make an express agreement for security with a third (if he can lull the suspicion excited by the deeds not being forth- coming, a feat which, though not easy, has been per- formed with success), and execute a legal conveyance by way of mortgage to a fourth, while the value of the V DEVELOPMENT OF THE MODERN LAW 137 property on wMcli they rely for payment is wliolly insuificient to satisfy them all. The judicial adjustment of the conflicting claims thus created, and the determina- tion of the order of preference to be given to the several creditors, demand the solution of most intricate legal problems. The Court of Chancery has manfully faced these difficulties, and, with the laudable intention of doing complete justice, has introduced various rules and dis- tinctions which in their turn, being worked out in great detail, have led to unforeseen and sometimes unreasonable consequences. It may be doubted whether these refine- ments have not, on the whole, done more harm than good. Endeavours have been made by Parliament at different times, from the end of the seventeenth century onwards, to deter borrowers on landed security from committing fraud on lenders by the concealment of earlier charges. These, however, have been but moderately successful, and the present state of things is anything but satis- factory. A lender may assure himself that there is at least one mortgage prior to his own, but he cannot be sure that there is only one, or any definite number.^ Neither the Court of Chancery nor Parliament is much to blame. The power of encumbering land by secret and informal charges is but the natural outcome of the events which followed the Statute of Uses. When once the Statute of Enrolments was successfully evaded, and no fresh measure taken to restore the original purpose of the Statute of Uses, it was settled for many generations to come that dealings with land, instead of being, as the old law required, open and notorious, ' Cp. Solicitm-s' Jouriial, xxx. 268. 138 THE LAND LAWS chap, t might be private and secret. The refinements and per- plexities of " equitable mortgages " have only carried out this principle. And if the Legislature or the public are minded to have any effectual amendment, they must in turn consider whether there is any way short of wholly reversing the principles of modern conveyancing, and going back, in principle though not in form, to the ancient ways of the fathers of the Common Law. CHAPTER VI LANDLORD AND TENANT A LANDHOLDER whose land is more than he can occupy and manage himself, and who does not choose to part with the ownership, can use it in the following ways : — He may, where slavery exists, cultivate the land by slave labour under overseers. He may cause it to be culti- vated, still at his own risk and for his own profit, by the hired labour of freemen under the management of an agent or bailiff. Or he may hand over the cultivation and the profits to some person who makes his own advantage of it on agreed terms and conditions, paying for the use of the land either in money or by a share of the produce. The first of these methods is not now possible in any considerable part of the civilised world, and has not existed in Europe at any time or place which need be regarded for our purpose. The second is every- where possible in law, and is common, I believe, on some parts of the Continent, but in England a large owner who farms his own land is now met with only as an occasional exception. The third plan, namely, letting the land to a tenant-farmer, has been, in one or another form, generally adopted in modern times. Yet another, which cannot be classed with any of these, was prevalent liO THE LAND LAWS chap. in the Middle Ages ; that is, the holders of small tene- ments were bound by custom to do agricultural work, of stated amount and at stated seasons, on other land possessed for his own use by the lord under whom they held. These customary services may be called labour- rents ; they probably represented what we now call the letting value of the land, but we must remember that they were not dealt with as a matter of contract. As often as not they were rendered by personally freemen ; in the case of a bondman's holding the services were more burdensome and the tenure less assured, though a comparison of the medieval text-writers with such detailed records as are accessible suggests that by those writers, and still more by the later ones who have built upon their statements, the precariousness of even the bondman's tenure is exaggerated. At least as early as the thirteenth century these labour-dues were largely commuted into fixed money payments. In the latter half of the fourteenth century the depopulation and scarcity of labour following on the Black Death went far to break up the system, and by the beginning of the sixteenth century money-rents appear to have fully taken the place of personal service. The same causes which made the old labour-rents worthless also made it unprofitable to the landlord to farm his own land by bailiffs, and thus the way was cleared for the modern usage of letting to tenants. ^ ^ Archdeacon Hale's Introduction to Domesday of St. Paul's, p. Ivi. Thorold Rogers's History of Prices. Vinogradoff, Villainage in Migland. Mucli interesting and useful information on the social aspects of the history of tenancy in England is also to be found in Part 1. of Mr. G. Brodrick's English Land and English Landlords. VI LANDLORD AND TENANT 141 Not that this usage is of modern origin. The begin- ning of it may be traced at least a century further back than the Norman Conquest. Religious foundations, the greatest owners of land in the earlier Middle Ages, and also the least able to give direct attention to its manage- ment, made temporary grants to tenants from an early time. Towards the end of the tenth century we find Oswald, Bishop of Worcester, explaining in a letter to King Edgar his practice of letting church lands to a tenant for three lives (the lives being the tenant's own, and those of two other persons named by him as suc- cessors) on the terms of rendering specified services ; and several examples of these grants are preserved.'- Two or three centuries later the practice of the Dean and Chapter of St. Paul's was to let their farms for life (sometimes to two tenants for the life of the survivor), the tenant rendering a fixed supply of provisions for the use of the house, with or without money payments in addition.^ The farm buildings and live stock were found by the landlord, and the stock had to be accounted for at the end of the tenancy. The mitayer system of farm-holding, still prevalent in many parts of the Con- tinent, is not unlike this. When tenancies for a fixed term of years came into use is not certain; they were well known, however, in the thirteenth century, and were common in the fifteenth. We may perhaps assume that letting for a certain term was suggested by the rotation of crops. A three years' 1 Cod. Bvpl. No. 1287, and Introd. vol. i. p. xxxiv. It does not appear whether the two succeeding lives must be those of persons in existence during the original tenant's life. 2 Domesday of St. Paul's, xxxviii. 122 sqq. 142 THE LAND LAWS chap. lease, according to the old-fashioned husbandry, would enable the tenant to complete the round and secure the advantage of the wheat crop ; and a term of three years is in fact the shortest that occurs in practice. Yearly tenancy, now the commonest form of holding in England, is not a letting for one year, but "from year to year," that is, for an indefinite time, determinable by either party giving notice (formerly a half-year's notice, but now a whole year's in the absence of agreement to the contrary) ^ at the proper season. Perpetual leases, after the fashion of the Eoman emphyteusis, are unknown to English law, but there is no legal limit to the certain number of years for which a lease may be granted, and therefore no technical difficulty in the way of making it as good as perpetual. In practice twenty-one years is the greatest length of agricultural and occupation leases.^ Building leases are commonly made for ninety-nine years (but of late often for not more than eighty), and mining leases (which, as being in truth a sale of the minerals to be worked by the lessee, are in sundry ways peculiar) for terms varying with local usage, not as a rule ex- ceeding sixty years. Longer terms, as of 200, 500, or even 1000 years, are conferred upon trustees as part of the machinery of family settlements, and were for some time commonly used in mortgages ; it is enough here, if it be not too much, to mention their existence. In these cases there is no rent and no real tenancy. Leases for 1 Agricultural Holdings (England) Act, 1883, s. 33. ' ^ In many American States the length of farming leases is re- stricted by statute to ten or twelve years. On the other hand, a lease for a term of years with a covenant for perpetual renewal is a common form of urban tenure. VI LANDLOED AND TENANT 143 lives, renewable by custom on payment of a fine as often as a life fell in, were until lately the common method of managing ecclesiastical and corporate lands. The object of adopting this plan was simply to benefit the existing bishop or corporators at the expense of the see, college, or other corporate body, the fines, as they were received for successive renewals, being treated not as capital, which they really were, but as revenue. Modern reforms have put an end to this, I believe, in every case. Leaving exceptional cases aside, we pass on to con- sider the position of the tenant who holds either for a term of years, or as tenant from year to year. In the feudal plan of society there is no place for him ; and accordingly the legal doctrine starts from the conception that the relation between the landlord and the tenant is simply a personal contract. This conception is at the bottom of all the differences between freehold and lease- hold tenure, and, though largely qualified in its effects, must be borne in mind in order to understand even the most modern form of the law. The lessee's interest is now beyond question property, not the mere right to the performance of a contract. Still, being in legal theory the creature of contract, it has neither the dig- nities nor the burdens peculiar to freehold tenures. It is not the subject of feudal modes of conveyance, nor of the feudal rules of inheritance. No particular form of words is necessary for its creation ; and the custom of creating it by deed has become a legal requirement (and that not in every case) only by modern statutes. It could always be disposed of by will if the tenant died before the expiration of the term ; and in case of such death the law deals with it in the same way as cattle or 144 THE LAND LAWS chap. money, and it goes to the executor, as part of the " per- sonal estate," to be administered by the same rules as movable property. If undisposed of by will, the lease- hold tenant's interest belongs on his death to the same persons, and in the same proportions, as cash or railway shares which he has not disposed of. There is no such thing as an heir of leaseholds. In one word, which for the lawyer includes all that has been said, a leasehold is not real but personal estate. From a strictly feudal point of view there is not an estate at all, only a per- sonal claim against the freeholder to be allowed to occupy the land in accordance with the agreement. But as early as the thirteenth century two points were settled, which together constituted a true right of pro- perty in the tenant. If he was ejected in breach of his landlord's agreement, he could recover not merely com- pensation for being turned out, but the possession itself ; and this not only against the original landlord, but against a purchaser from him. Already the pur- chaser could not say to the tenant whom he found on the land, "I have made no contract with you; look for your redress to the man with whom you did contract." The farmer's possession was as secure while his estate lasted as the freeholder's.^ On the foundation thus laid the modern law has been completed, partly by judicial usage and partly by express legislation. Broadly speaking, both the landlord's and the tenant's successors in title enjoy, while the term of the tenancy lasts, the 1 "Won magis poterit aliquis firmarium eiicere de firma sua quam tenentem aliquem de libero tenemento suo." — Bracton, fo. 220h. For details see the section on "The Term of Years," in Pollock and Maitland, Hist. Eng. Law, ii. 105-17. VI LANDLOED AND TENANT 145 rights conferred at its creation upon the landlord and tenant respectively, and are subject to the burdens imposed on them. Exceptions may still occur, too rare and technical to be now further specified, which are just enough to show that the old notion of a mere personal agreement, though decayed, is not dead. There is, however, one ancient and peculiar incident in the relation between landlord and tenant which the theory of contract is incapable of explaining. This is the landlord's right of distress. Early records, both of English customs and of those of kindred nations, point to a time when distress was almost the universal form of civil remedy. When cattle were the only movable property of any value, and courts of justice had no swift or certain means of enforcing their orders, the most natural thing for a man to do who complained of wrong at his neighbour's hands was to drive off some of the neighbour's cattle, and keep them till the owner would either satisfy his claim or refer the matter to the decision of an impartial authority. Still more obvious is the form of self-help, preserved in our law to this day, which is called " distraint damage feasant " ; that is, impounding cattle •*• which trespass upon one's land as a security that the damage shall be made good. When courts of justice began to compel the attendance of parties before them (for there are traces everywhere of a time when they professed only to do justice between such as willingly submitted themselves), they applied the same kind of compulsion that was in familiar use in private quarrels. 1 Goods of any kind may be distrained damage feasant, and a locomotive has been so dealt with by a railway company ; but animals only need to be regarded for the present purpose. L 146 THE LAND LAWS chap. As local jurisdiction passed into the hands of lords of manors, distress became the regular means of compelling persons subject to the lord's jurisdiction to appear in his court, and also, with or without judicial proceedings in the court, of enforcing payment or performance of the rents and services due from tenants. Some local customs even allowed the lord to seize the land itself. By steps of which nothing certain is known, it came to be under- stood that an agreement for the occupancy of land, though it created no feudal tenure, and therefore no service in the proper sense, entitled the owner, if the rent fell into arrear, to seize any goods he could find on the land as a security for its payment. Probably the right was claimed and exercised without dispute by analogy to the rights of the lord of a manor against his freehold tenants. It may have contributed to the readier allowance, as it certainly does to the apparent justice of the proceeding, that in the Middle Ages the live stock of the farm were mostly supplied by the landlord. In such a case, if the tenant became insolvent, a landlord who seized the stock was only resuming his own. Be the early history what it may, the right has existed ever since English common law took a definite shape. ^ In the course of the last two centuries legislation has made it a far more efficient instrument in favour of the landlord ; ' The student may consult, besides Sir H. Maine's chapter on the Primitive Forms of Legal Remedies, the chapter on Distraint in Mr. Bigelow's History of Procedure in Englaind. For the com- mon law in its settled form, see Blackstone, hook iii. ch. i. Our early medieval law was in general stricter against self-help than the modern law ; see Pollock and Maitland, Hist. Eng. Law, ii. 573. Nevertheless the landlord's right of distress without judgment grew up apparently without dispute. VI LANDLORD AND TENANT 147 whicli has indeed been the tendency of all legislation concerning landlords and tenants until a pretty recent time. By the common law things taken in distress were a mere pledge for payment of the rent. Therefore nothing might be distrained which could not be restored in the same condition, such as corn in sheaves ; nor anything which grew out of or was fixed to the soil, such as standing corn ; and the landlord had no power to sell distrained beasts or goods, nor to deal with them in any way as owner, even for the owner's benefit, except, perhaps, in case of evident necessity.^ And, moreover, as Blackstone tells us, "the many particulars which attend the taking of a distress used formerly to make it a hazardous kind of proceeding ; for, if any one irregularity was committed, it vitiated the whole," and the person distraining became a mere trespasser. This has already been noticed by Sir Henry Maine as evi- dence of the archaic nature of the institution. "The excessive technicality of ancient law" clings to all ancient customary remedies unless and until, as in this case, modern legislators remove it for the benefit of the parties, or one of them. In 1689 the power of selling things taken by distress, after notice to the tenant, was first given to the landlord by Act of Parliament ; it was further secured and defined by subsequent statutes of ^ It has been said that he had no right to milk a cow. The learned reader may compare the somewhat similar proposition that milking the testator's cow will make a man executor de son tort. But the statement is only in EoUe's Abridgement, i. 673, and is directly contradicted by the report of the same case in Cro. Jac. 147, where things done of necessity and for the owner's benefit are excepted from the general rule. 148 THE LAND LAWS chap. the eighteenth century, and the old restrictions on the kinds of things liable to distress were greatly relaxed. "The summary power of sale now exercised has been created by the recent statute law with more attention to the profit of the rich than to the rights which were secured to the poor by our ancient jurisprudence." ^ It will be understood that the right of the landlord is other and greater than the tenant could give him by contract : it is to take chattels found on the holding, whether the tenant's property or not. This, if it were simply a matter of agreement, the tenant could of course not enable the landlord to do : contracts cannot impose liabilities on persons who are not parties. In the case of under-tenancies, a thing unknown when the rule of law was fixed, great hardship may be the result, since the goods of a sub-tenant may be distrained for rent due from his immediate lessor to the superior landlord. This " great loss and injustice " (in the words adopted by Parliament) was remedied in the case of lodgers by an Act passed as lately as 1871. The expediency of the right of distress as regards farm holdings has in recent years been much discussed. It is rational to say that, since the landlord cannot help giving credit, he ought to be in some shape and to some extent a preferred creditor : but the existing law of distress gives effect to this principle in the most rough and irrational fashion. It would seem that the advocates of complete " freedom of contract " in the relations between landlord and tenant ought to have been the first to demand the abolition of an anomalous customary privilege which contract could never have been engendered ; but such was by no means ^ Palgrave, English Commonwealth, i. 182. VI LANDLORD AND TENANT 149 the case. By the Agricultural Holdings Act of 1883 (46 & 47 Vict. c. 61) the landlord's right to distrain, which at common law extended to six years' back rent, is limited to one year as regards holdings within the Act — that is, farm, pasture, and market-garden lands : and hired machinery, and live stock not being the tenant's own property, are exempt from distress — the former absolutely, the latter with certain qualifications. It may be regarded as a kind of set-off to the power of distress that other summary powers and remedies which lessors are in the habit of securing to themselves by contract have lately been mitigated in their opera- tion, so as to prevent them from being abused beyond their true function of securing the landlord's interest. * For a long time it has been the common practice to insert in leases, especially of town property, provisions enabling the landlord to re-enter and put an end to the lease if the tenant fails to perform his obligations. Non-payment of rent is the cause most commonly specified ; another is neglect to keep the premises in- sured, where the tenant has covenanted to insure ; frequently the clause of re-entry extends to breach or neglect of any one of the tenant's covenants. ^ With regard to non-payment of rent, the tenant became entitled, first by the practice of the Court of Chancery, and then by modern Acts of Parliament, to be relieved from forfeiting the lease on payment of the rent in arrear and costs. About twenty years ago neglect to insure was also made remissible on certain conditions. And now by the Conveyancing Act of 1881 (sec. 14) ^ Covenant, whicli etymologically is a mere synonym of agree- ment, signifies in English law any promise made by deed. 150 THE LAND LAWS chap. the Court has power to grant relief to the tenant in its discretion, and on such terms as it thinks fit, except in the case of non-payment of rent, as to which the former Acts remain in force. Another general rule of law which bore hardly on tenants has gradually been relaxed in their favour, namely, that which declares that things affixed to the soil become for legal purposes part of it, and cease to be the tenant's. The truth is, and it may as well be stated at this point, that the law of landlord and tenant has never, at least under any usual conditions, been a law of free contract. It is a law of contract partly express, partly supplied by judicial interpretation, and partly controlled by legislation and sometimes by local custom. So far as the terms and conditions are ex- press, they are in the vast majority of cases framed by landlords or their advisers. The tendency of judicial interpretation has also been, until lately, to incline the scale of presumption in favour of the landlord on doubtful points ; and the same may be said of the ruling tendency of legislation down to the middle of the present century. The allowance of local customs, which might have done much to redress the balance if taken up betimes, depends on the tendency of the judges. When special customs were looked on as a kind of natural enemies of the common law, and strict proof of them was required, they got little help in court. Probably many tenants have in past times failed to establish customary rights, or have been discouraged by the failure of others from asserting them, in cases where the decision would now be the other way.^ As 1 A good example of the present tendeney to give full effect to VI LANDLORD AND TENANT 151 to the point now in hand, it was settled early in the eighteenth century that a tenant might at any time before the end of his term remove fixtures set up by him for purposes of trade. Early in the present century the judges, with unfortunate timidity, declined to ex- tend the same principle to buildings and fixtures pro- vided by a tenant-farmer for purely agricultural use. What they did not see their way to declaring as common law has by successive steps been enacted by Parliament. By an Act of 1851 a tenant who, with the landlord's consent in writing, put up farm buildings or machinery at his own cost, was enabled to remove them as his own property, subject to an option on the landlord's part to take them at a valuation. The Agricultural Holdings Act of 1875 extended this right of the tenant (with some small variations of language in his favour) to engines, machinery, and fixtures affixed to the hold- ing even without the landlord's consent, saving, how- ever, to the landlord a right to object to the erection of a steam-engine : lastly, the Agricultural Holdings (England) Act, 1883, has re-enacted this provision without the saving. ^ There remains a serious economical question between landlords and tenants, for which the common law failed to provide a solution. No general rule of law gives an outgoing tenant any compensation for permanent or unexhausted improvements made by him. A tenant who has a pretty long lease may no doubt so order, matters as to get the full benefit of his improvements ; established local usage is afforded by the case of Tucker v. Linger, decided by the House of Lords in 1883, Law Rep. 8 App. Ca. 508. 1 S. 34. 162 THE LAND LAWS chap. though even so it is the worse for the land to make it the tenant's interest to leave nothing in it. But in England yearly tenancy is the rule, and leases for fixed terms the exception. In some parts of the country local customs exist vphich have been found fairly satis- factory, providing a scale of compensation for the tenant's outlay on lime, artiiicial manures, artificial feeding stuffs, and the like. But these are of limited extent, and applicable, as a rule, only to the soils and agricultural usages of the particular districts where they prevail. After much discussion, and an experiment by way of permissive legislation in 1875, on which it is no longer necessary to dwell, the matter was dealt with by the Agricultural Holdings Act of 1883. This Act is somewhat clumsy in form, and bears in every part, after the manner of all English legislation involving a conflict of interests, the marks of compromise and abundant caution. It has established, however, in opposition to the old common law, the leading principle that an outgoing tenant who has improved the holding is entitled to get by way of compensation "such sum as fairly represents the value of the improvement to an incoming tenant.'' The improvements for which com- pensation may be obtained are, it is true, defined and classified with excessive minuteness ; and the previous consent of the landlord to permanent improvements (such as building, irrigation, planting, and reclaiming) is a necessary condition of their being a subject of compensation. As to this class of improvements, therefore, the only effects of the Act are to make the landlord's consent, if given, binding on the land and on VI LANDLORD AND TENANT 153 his successors (which, however, is important, having regard to the great amount of land in settlement, or in the hands of trustees or public bodies or officers with limited powers), and to throw on the landlord the burden and odium of refusing consent where the im- provement is clearly a proper one. Perhaps it was . impossible to go further consistently with the funda- mental assumption of English leasehold tenure, that the lessor is entitled to have back his land at the end of the term in the same condition in which the lessee took it. As to drainage, a middle course is observed : the tenant need not obtain the landlord's consent for drain- ing the land, but he must give him notice of the in- tended work, and then the landlord may do it himself if he thinks fit, and charge the tenant with an addition to his rent by way of interest. Exhaustible improve- ments, such as liming and manuring, may be made, and will entitle the tenant to compensation, without the landlord's consent. Claims for compensation are to be settled by reference ; the local county court may ap- point a referee or umpire, as the case may require, in default of either of the parties or their referees acting, or either party may call for the nomination of an umpire by the Board of Agriculture. The Act contains, moreover, a number- of minor provisions and safeguards for which the text itself must be consulted. ^ It seems, on the whole, fitted to carry out its objects with as little friction and waste of power as, our system of ^ There are some later amendments and extensions of the Act : see the Tenants' Compensation Act, 1890, and the Market Gardeners' Compensation Act, 1895. Similar but distinct provisions were made for tenants of allotments by the Allotments and Cottage Gardens Compensation for Crops Act, 1887. 154 THE LAND LAWS chap. legislation being what it is, can fairly be expected. But it remains a misfortune that the principle of com- pensation declared by the Act was not in the first instance adopted by the common law. Local usage and a certain amount of judicial decision would then have settled the method and details of the tenant's compensation much better than the clauses and schedules of a modern Act of Parliament can do it. We are no longer able to declare a broad principle and leave it to work itself out ; we expect every question of detail to be met beforehand, and entangle ourselves in intolerable minuteness without, after aU, being secure against material oversights. And we shall hardly escape or mitigate this evil until either we greatly extend the modern device of delegating legislative powers, or in some way return to the older fashion of legislating by way of general instructions and conferring a large dis- cretion on the executive authority. This, however, belongs to the problems of the general theory of legis- lation. Except as regards the special legislation for agricul- tural tenancies which has just been noticed, and which dates only from the middle of the present century, there is no difference in law between urban and agricultural leaseholds. But in practice and custom the difference is very great. Farm holdings are always or almost always taken by the tenant direct from the freeholder, and there is generally something of a personal relation between them (even where the landlord is a college or other corporate body) beyond the mere re- ceipt and payment of rent. The farmer is legally bound to pay the full amount of his agreed rent, VI LANDLORD AND TENANT 155 without regard to the goodness or badness of the season ; but in bad years it is the constant practice for the landlord to remit such a percentage of the rent as to leave the tenant answerable only for so much as the farm seems fairly capable of paying under the circum- stances. A great landlord who refused to follow this practice would be entirely within his legal rights, but would certainly be thought the worse of in the country. In many counties, unfortunately, reductions of 15 or 20 per cent (sometimes even more) have now been rather the rule than the exception for some years past. The landlord in return expects a certain amount of deference and compliance in various matters from his tenant. Not only does the farmer meet him half-way on questions of shooting rights, and allow free passage to the hunt, but his political support of the landlord is not unfrequently reckoned on with as much confidence as the performance of the covenants and conditions of the tenancy itself. In the case of holdings from year to year it may be not unfairly said that being of the landlord's political party is often a tacit condition of the tenancy. In the letting of buildings for occupation, whether for business or as dwelling-houses, and especially in large towns, there is nothing of this kind. The trans- action and the relations of the parties are purely a matter of commerce. The owner of land on which there stands a shop or a factory sells the right to occupy it to the shopkeeper or manufacturer just as he in turn sells his goods, that is, for the highest price he can get. And the same rule holds as to leases of mines and collieries, where the sale is not merely of the right 156 THE LAND LAWS chap. to use the land, but of a portion of the land itself, namely, the coal, ores, or other minerals to be worked by the lessee. As regards occupation leases, again, underletting, instead of being the exception, is the rule in the south of England. The freeholder of building land in or near a town makes a lease of it to a builder for a term, which used to be of ninety-nine years, but is now (at least in and about London) commonly not more than eighty. The builder undertakes to cover the ground with a house or number of houses, according to requirements laid down by the landlord in more or less detail. At the end of the term the buildings become the landlord's property, the lessee finding his compensa- tion in such profit as he can make of them during the term by subletting to occupying tenants. In houses of a good class, which are occupied as a whole, there is commonly only one underletting. But there may be more ; and in the case of smaller tenements the actual dweller often has to pay several intermediate profits, getting in return bad quarters and a precarious tenure. The system of building leases is convenient for the free- holder, as it saves him all trouble beyond that of laying out the plan or the general character of the buildings to be put on the land, and ensures him the possession of his property greatly increased in value at the end of the term. He might build himself and let direct to occupiers, thus getting the full profit at once, but he would then incur also the trouble and risk which by the usual arrangement are thrown on the first lessee, commonly known as the "speculative builder.'' For all other parties, and it would seem for the public, the plan is a thoroughly bad one. As the builder has no VI LANDLORD AND TENANT 157 permanent interest in Ms work, lie has no motive for making it durable, and is tempted to make it, on the contrary, as bad as the ground landlord's surveyor can be induced to pass, and occupying tenants to accept for their habitation. He buys the use of the land from the freeholder for a limited time, and retails it to the public either directly or through a middleman. All the risks of the retail market fall on him ; some time must elapse at best before he can make a proiit, for it takes time, even with hxuried and flimsy work, to build houses and make them look fit to live in. There is no business in which profits are more uncertain, and none, I believe, in which there is more insolvency ; on the other hand, great fortunes have been made by well- known builders in London and elsewhere. When the builder's or middleman's turn comes, it is of course his interest to make the most he can out of the under- tenant for the time allowed him. Dwelling-houses, therefore, are necessarily rack-rented.^ By well-to-do sorts of people this is perhaps not felt as a grave incon- venience, but the pressure and hardship on the smaller under-tenants are very great. It is evidently absurd to speak of freedom of con- tract in relation to such a system. Desirable building ground near towns, and still more the ground of towns and cities already long occupied, and eminently those districts and sites which are favoured by business or fashion, are a monopoly in the hands of the landowner. ' Kack-rent, I need hardly explain, is the highest annual rent that can be obtained by the competition of those who desire to becsme tenants. It is not a strictly legal term, though sometimes used in Acts of Parliament ; in legal documents it is represented by " the best rent that can be obtained without a fine." 158 THE LAND LAWS ohap. The landowner dictates his terms to the building lessee, who in turn dictates them to the occupier, making the occupier's obligations, for his own protection, exactly follow those of the original lease. In this way the population of whole cities may be said to live at the will of a few great landlords. Over whole square miles of what is commonly called London the Duke of West- minster or the Duke of Bedford may without appeal or control forbid any given kind of building to exist, or any given kind of business to be carried on. In the eye of the law the Westminster, or Bedford, or Port- man estate is simply so much land in the administrative county of London, which the freeholder need not have built upon, or allowed others to build upon, unless he chose. If he does not choose to sell, and chooses to grant building leases, it is for people to take or leave, —as, where it is difficult to find a tenant for a farm in depressed times, it may be for the landlord to take or leave the tenant. Accidents of site and social con- ditions may give the freeholder a supreme monopoly, but that is his good fortune, an extraneous matter with which the law does not concern itself. That such powers are used, on the whole, with so little ground for complaint as they are, is much to the credit of human nature and of the training of English gentlemen. But the matter involves more than one serious problem for the next generation, if not for our own. The value of land for occupation in towns has increased out of all proportion to the increase in the value of other land ; it is comparatively little affected by the depression of agricultural industry, and no reason appears why 'the increase should not continue. This affords another VI LANDLORD AND TENANT 159 potent reason why the fortunate owners of city ground- rents should go on leasing rather than selling ; for to sell the freehold of such ground as the bounds of the city of London encompass is to part — however great be the price in hand — with potential riches beyond all present valuation. We are in sight of an accumulation of wealth and power in a few hands, and concentrated on vital parts of the commonwealth, such as is without example in history, and might conceivably be a danger to the State. But it is far more easy to perceive the danger than to devise a remedy. In Scotland the practice is to grant the land in perpetuity for a fixed rent (which under the feudal rules of Scottish law is easily done by a form of conveyance called ili modern practice a " feu-contract ") ; building leases of the Eng- lish type are, I believe, unknown. In the north of England it is not uncommon to do what comes nearly to the same thing in a less direct form by conveying the land in fee-simple and taking an annual payment out of it under the name of rent-charge. ^ But this is anything but satisfactory, for every part of the land is liable for the whole rent : therefore, if the subject- matter of the original grant is subdivided among other purchasers (which constantly happens), the occupying owner will find himself under liabilities which are un- certain and in possible events may be ruinous. Leases for a very long term, such as 999 years, have sometimes been used for the same object, and are open to the same ' Since the statute of Quia emptorcs (p. 70 above) a rent proper cannot be reserved upon a grant in fee-simple except by the Crown. There was never any corresponding enactment in practical force in Scotland, and the statute is not in force in some American States. 160 THE LAKD LAWS chap. inconvenience. In other parts of the north country the land is sold off in plots to builders who become absolute owners ; the simplest plan of all, and, if the parties will use it, the most rational. It has been proposed to confer on lessees of house and cottage property the power of acquiring the free- hold by way of compulsory purchase, or, to put it in a neater and perhaps more easily understood form, to attach a statutory option of purchase to all leases of a certain length. The terms would have to be in each case settled either by arbitration or by some kind of judicial process if the parties could not agree ; the refer- ence clauses of the Agricultural Holdings Act might serve for' the first lines of a working plan. The first effect of such a law would probably be to check the supply of building land and dwelling-houses by making landowners unwilling to grant new leases, and thus to aggravate for the time the evils of monopoly. What would be the ulterior effects it is not easy to foretell. It seems doubtful whether the acquisition of permanent interests by the people most in want of them would really be much facilitated. And it does not seem clear that building lessees are now, as a rule, anxious to acquire the freehold : for if they were so, they could and would find opportunities even under the present system. Urban freeholders are not often willing to sell, but forced sales are every now and then brought about by various causes. Still, it is something to recognise that a problem has to be faced. The idea of " leasehold enfranchisement " has at least such elements of a solution as to deserve consideration, and in London and one or two parts of the country it has attracted VI LAN'DLORD AND TENAKT 161 serious attention. In 1886 a Select Committee of the House of Commons was appointed to consider questions connected with town holdings ; it finally reported on leasehold enfranchisement in 1889.^ The report allowed that some advantages were likely to attend the conversion of leasehold into freehold tenure, but recom- mended only the partial application of the principle through powers to be given to local authorities over limited areas. Nothing has been done to give effect to this report. In later years there has been less heard of local enfranchisement and more of " betterment " and the taxation of ground-rent ; but the intricate con- troversy which has beset the last-mentioned topics is beyond the scope of this work. It seems fit to be weighed whether any systematic reform of urban tenure should not aim at making the municipalities rather than individual occupiers the ultimate owners. One or two of our northern cities, by using such occasions as presented themselves from time to time of acquiring property within their own borders, have already made some way in this direction. The recent Allotments Act, of which some account must now be given, is in the same direction so far as it goes. Good landlords have long been in the habit of en- couraging labourers dwelling on their estates to become tenants of small allotments, and cultivate them for their own profit and pleasure in spare time. But it has appeared that individual goodwill and enlighten- ment cannot always be trusted to make reasonable ' Pari. P. 1889, xv. 1. On the ground-rent question there is a further report of 1892. M 162 THE LAND LAWS chap. provision of this kind; and Parliament has now con- ferred on local authorities a limited power of acquiring land for allotments if satisfied that adequate voluntary arrangements cannot be made. The Allotments Act, 1887, 50 & 51 Vict. c. 48, enables the local sanitary authority, being so satisfied, to buy or hire suitable land, by agreement if possible ; failing agreement, the County Council may, on petition from the sanitary authority, making a provisional order conferring powers of compulsory purchase similar to those exercised by railway companies and other public bodies for other purposes of public interest. The provisional order must be confirmed by Act of Parliament in order to become effective. Parks, gardens, and the private grounds of dwelling-houses are absolutely excepted from compulsory taking. Allotment land acquired under the Act is to be so let and managed as to make the undertaking self-supporting. Allotments are not to exceed one acre, and sub-letting is forbidden. It will be seen that the Act rather holds out a distant prospect of compulsion than attempts any direct inter- ference with private discretion. It was intended to operate as a gentle stimulus to landowners to follow, where practicable and not yet practised, the example already set by the best of them ; and in this way it has done considerable good. Sometimes, I believe not often, the compulsory powers have been exercised. By the Small Holdings Act, 1892, local authorities are empowered to acquire land in their own names and sell or let it in small parcels for agricultural occupation; and under the Local Government Act, 1894, parish councils have power (which may be made com- VI LANDLORD AND TENANT 163 pulsory by an order of the County Council, on proper cause being shown) to hire land for allotments. It would be useless to give details here.^ ■^ Information as to the working of the Allotments and Small Holdings Acts can easily be obtained from the publications of the Rm-al Labourers' League (95 Colmore Row, Birmingham). CHAPTEE VII MODERN REFORMS AND PROSPECTS For about a century and a half there was no material change in the English law of real property. There was, indeed, in the early part of the eighteenth century a period of minor reforms, which are at this day perhaps not sufficiently remembered. The mechanism of judicial and other proceedings was in some respects simplified, and some opportunities of fraud still afforded by obsolete doctrines were removed. But the work of the preceding century was in the main left untouched. The statute which abolished the military tenures marks the end of one period of transformation. The Act for the Abolition of Fines and Recoveries passed by the first reformed Parliament marks the beginning of another which is still in progress. As a help to memory, the Restoration and the Reform Act of 1832 may well be thus taken as the landmarks, although the current form of settlement of landed property was not fully worked out until towards the middle of the eighteenth century, and on the other hand the movement leading to the series of legal reforms now to be spoken of was formally begun, by the appointment of a commission of inquiry, some years before it bore fruit, and was in substance due in great CHAP. VII MODERN REFORMS AND PROSPECTS 165 measure to the labours of Bentham. And Bentham's criticism was in turn provoked and made possible by Black- stone's artistic exposition of the settled form of the law which in his own time was still recent. So it is every- where, both in nature and in the affairs of man ; what seems to be repose is in truth the preparation of change. It will be convenient to take the modern alterations of the law not in order of time, which would lead to great intricacy and to the confusion of distinct topics, but according to the subject-matter. And first we will take the formal simplification of dealings with land. From the latter part of the sixteenth century onwards the common mode of conveying freehold land, as we have seen, was by the ingeniously artificial process of "lease and release" (p. 106 above), which involved the bulk and expense of two deeds. The length of these documents also steadily increased until about the begin- ning of the present century; partly by reason of the greater complication of the aifairs and interests to be dealt vidth, and the new precautions devised by lawyers to meet newly-discovered dangers to titles, partly because, by an evil usage which has only in the last few years been abated, professional skill was recompensed merely in proportion to the amount of writing produced. In 1845 (after one or two rather clumsy experiments, which it is needless to specify) the first of these troubles was removed in a manner so simple and elegant as to com- mand the approval of even the conservative school of conveyancers. The ceremony of feoffment was always in- applicable to an estate in reversion or remainder. Livery of seisin — the formal transfer of possession which has been described elsewhere (p. 75 above) — could be given 166 THE LAND LAWS chap. only by one who had the actual possession to one who had it not. A freeholder not having the right of actual possession (as where the land was held by a tenant for life or a lessee) disposed of his interest by deed without other ceremony ; it was an " incorporeal hereditament " which could not be delivered. The same rule applied to such rights over the lands of others, rent-charges, rights of way or common, and the like, as were capable of dis- position and transfer. All rights and interests that could be so dealt with were said to " lie in grant." The Act to amend the Law of Real Property passed in 1845 ex- tended this rule to estates in possession.^ Since it took effect everything which can be done by feoffment, bargain and sale enrolled, or lease and release, can be equally well done by a simple deed ; and although the older forms of conveyance are still open to any one who might choose to make use of them, the readier way provided by this Act has been almost universally followed. Certain further simplifications as to the use of particular words (not all of them really new) have been introduced by the Con- veyancing Act of 1881. They are too technical to be dwelt on here ; only it is worth notice that they do not touch the substance of the old feudal rule — namely, that a grant to a man without specifying what interest he is to have will give him no more than an estate for life.^ ' " After the said first day of October, one thousand eight hundred and forty-five, all corporeal tenements shall, as regards the conveyance of the immediate freehold thereof, be deemed to lie in grant as well as in livery." — 8 & 9 Vict. u. 106, s. 6, The capacity which a feoffment formerly had of practically giving a purchaser a better title than his vendor's (p. 82 above) was abolished by the same statute. ■' 44 & 45 Vict. 0. 41, s. 49, etc. These provisions have already VII MODERN REFORMS AND PROSPECTS 167 As to the other point of the prolixity of deeds, much more was left for the Conveyancing Act to do. In the same year (1845), and I suppose under the same auspices as the Act to amend the Law of Real Property, a statute was made with the laudable intention of substituting short forms for the accustomed verbose ones. It was a complete failure, partly from defects of workmanship which made it positively misleading, but chiefly because no legal practitioner who adopted it would, under the vicious scheme of remuneration then in use, have been tolerably paid for his work. The abbreviated forms provided by the statute were seldom or never used, and for another generation deeds went on being framed as before, though the practice of the best conveyancers was now to study conciseness. At last in 1881 a new Act ("The Conveyancing and Law of Property Act, 1881 ") dealt with the form of deeds relating to land in a far more thorough-going and elaborate manner. The use of a few prescribed words will now incorporate in a deed, according to the nature of the case; one or other of the several accustomed clauses which were formerly inserted at full length ; and in ordinary straightforward cases this may conveniently and safely be done. At the same time another Act (the Solicitors' Remuneration Act) abolished the necessity or supposed necessity of paying for the preparation of deeds according to length, and thus left the way clear for the Conveyancing Act. This last, however, is not compulsory ; people may go on using the old forms as much as they please, but the given rise, I believe, to a popular error that the use of the word grant is not only unnecessary but improper, whereas it remains as proper as ever. 168 THE LAND LAWS chap. new provisions have been generally adopted. Supple- mentary Acts were passed in 1882, and (for the further protection of lessees) in 1892, but the points dealt with (as well as many of those covered by the Act of 1881) are too technical to be specified in this work. The credit of these measiu'es belongs to both parties in the State ; they were substantially prepared under the direction of Lord Cairns in the later years of the Con- servative Ministry which held oflBce from 1874 to 1880, and were taken up and passed into law with little alteration by the Liberal Ministry which succeeded. The same may be said concerning the Settled Land Act, of which we have to speak later. The cumbrous machinery of fines and common re- coveries (which has been described in Chapter IV.) was swept away in 1833, and a tenant in tail is now em- powered, by means of a deed enrolled in Chancery, to make either himself, or any one to whom he wishes to dispose of his estate, a tenant in fee simple. If he is in possession of the freehold, he can generally do this without any other person's consent. If he is not in possession, he must, as a rule, have the consent of the "protector of the settlement," who is commonly the tenant for life, otherwise he can create no greater estate than a hase fee (p. 113 above). The name and function of the " protector " were introduced by the Act of 1833; but the purpose and the effect of the Act were, while simplifying the necessary proceedings, to preserve intact in substance the law and practice of family settlements. Considered as an improvement in form, which is all it professed to be, the Act for the Abolition of Fines and Eecoveries has been of great VTi MODERN EEFOEMS AND PROSPECTS 169 service, and has worked with singular freedom from difficulties of any kind. This success is due in some measure to the circumstance (not so common, unhappily, that notice of it should be superfluous) that the framing of the Act was entrusted to a man who thoroughly understood the matter he was to deal with. It seems worth while to quote a few sentences from the First Report of the Eeal Property Commissioners to show how the problems of law reform presented them- selves two generations ago to learned and enlightened persons. It will be seen that the limitation of their proposals was the result not of opposition or compromise but of free and deliberate conviction. "The owner of the soil is, we think, vested with exactly the dominion and power of disposition over it required for the public good, and landed property in England is admirably made to answer all the purposes to which it is applicable. " Settlements bestow on the present possessor of an estate the benefits of ownership, and secure the property to his posterity. The existing rule respecting perpe- tuities has happily hit the medium between the strict entails which prevail in the northern part of the Island, i and by which the property entailed is for ever abstracted from commerce, and the total prohibition of substitutions and the excessive restriction of the power of devising established in some countries on the Continent of Europe.^ In England families are preserved, and pur- ^ It was then possible to create inviolable entails in Scotland ; the law has since been changed, and alienation is as easy as in England. ^ Meaning probably those which adopted or imitated the Code NapoUon, 170 THE LAND LAWS chap. chasers always find a supply of land in the market. A testamentary power is given, which stimulates industry and encourages accumulation ; and while capricious limitations are restrained, property is allowed to be moulded according to the circumstances and wants of every family." Thus the Commissioners in 1829.^ Lord St. Leonards would have been in their eyes a rash innovator, Lord Cairns a revolutionist, and for Lord Halsbury parlia- mentary language would have failed them. But we are anticipating. Another point where simplification was urgently called for, and was carried out a few years later, was the disposal of land (as well as of other property) by will. The state of things before 1838 (brought about by steps which we need not consider) was that the formalities required for a will of lands were excessive, those for a will of leaseholds or movables absurdly and danger- ously defective. There were also peculiar rules and exceptions (partly by common law, partly by custom, and partly by statute) in particular cases, and, on the whole, it was found on careful inquiry that there were ten difi'erent laws for regulating the execution of wills under different circumstances.^ An Act of 1837, which applies to all wills made since the end of that year, abolished all these intricacies, together with many other incon- veniences and doubts, and established the uniform rule that a will must be signed by the maker of it, and attested by two witnesses. It also applied to wills the rule — still not applicable to deeds — that a simple and unquali- 1 First Keport, p. 6. ^ Fourth Report of Real Property Commissioners, p. 12. vn MODERN REFORMS AND PROSPECTS 171 fied gift is to be taken as disposing of the giver's whole interest. It is not the length of deeds, however, nor any for- mality required for either deeds or wills, that is the real source of expense and trouble in dealing with English land. The peculiar system of private conveyance which has grown up under the shadow of the Statute of Uses has made it needful for purchasers to secure themselves against the " constant danger from secret transfers and secret charges " ^ by an examination of the history of the property as evidenced by the title-deeds. Except in the rare cases where estates have been settled by Act of Par- liament, and in the case of land registered under the Land Registry Act of 1875, undisturbed possession under a continuous title for a certain length of time is the only proof of the rightful ownership of land in this country. The title-deeds are the written history of the possession, and of the right in which it has been exer- cised ; and from this point of view their contents are, or in strict prudence ought to be, examined by every purchaser. Almost always this operation requires some professional skill ; often it requires much. In all but the simplest cases the process is a long and costly one. Not only solicitors and counsel have to be paid for their intellectual work, but a good deal of clerical and mechanical labour is involved in making and verifying copies or abstracts,^ searching certain official records ^ First Report of Real Property Commissioners, p. 8. ^ The digest of documents and facts laid before counsel as the basis of Ms opinion is called an "abstract of title." It is the solicitor's business to verify the contents of this, the counsel's to point out, if necessary, what further verification is required, and to advise whether, assuming the statements to be correct, they 172 THE LAND LAWS chap. which might disclose matters affecting the title, and the like. All this falls on the buyer, and in the case of small properties acts as an exorbitant tax added to the purchase-money ; for the cost and difficulties are no less for a small property than for a large one. Indeed large properties are better off; for there are some estates of which the possession and title are historical, and so well known that, if a portion is to be sold or let, the owners can afford to make what terms they please as to dispens- ing with inquiry. Less fortunate sellers have to examine their own titles before going into the market, and to guard themselves against requisitions which it might be impossible or ruinous to comply with by selling only on carefully framed special terms, which are known, if the sale is by auction, as " conditions of sale." Very few landowners are in a position to make out their title if strict proof is required at every step, which it may be ^ in the absence of special conditions. One need not dwell on the inconvenience of such a state of things in a great commercial country. It was amended in some particulars by an Act of 1874, which made some of the most usual special conditions part of the general law. The root of the evil, however, is un- touched. Many times it has been proposed to establish an official registry either of deeds or of titles. The whole subject was elaborately considered by the Eeal Property Commissioners in 1830; they collected much informa- sufficiently show the vendor to have power to dispose of what he oifers to sell. ^ The practice of conveyancers is to be content in sundry matters with less than would be accepted as sufficient evidence in a court of justice ; but even so the burden on the vendor under an "open contract " is a grievous one. VII MODERN REFORMS AND PROSPECTS 173 tion both at home and from abroad, and recommended the establishment of a general registry ; but nothing was then done. Since that time the experiment of permissive registration of titles has twice been made, with next to no result. In 1862, in Lord Westbury's chancellorship, an office of land registry was established, and the offer of a State guaranty and great simplicity in future deal- ings was held out to owners who could satisfactorily prove their title. But the requirements of the initial proof were for most owners more formidable than the inconveniences of the existing system. Those incon- veniences, moreover, fall chiefly and visibly not so much on existing owners as on those who buy from them ; while every man can see the mischief of exposing his own title to a rigorous official scrutiny which may dis- close an unsuspected flaw, and must invite his neigh- bours to raise questions of boundaries. It seemed the more prudent part to let sleeping lions alone. Solicitors, again, disliked the scheme, and gave it anything but encoittagement ; as with the attempt to shorten deeds, the matter was complicated by the problem of remunera- tion. In 1875 Lord Westbury's plan, being supposed to have failed through over-ambition, was supplanted by a more tentative and elastic one. But the office of land registry as reconstituted by Lord Cairns's Land Transfer Act, though not now, as it was for several years, an office of sinecures, is still avoided by most landowners.^ The inevitable weakness of every scheme of merely optional ' "Registration under this Act is optional, and its success has not been sufficient to justify any lengthened account of it in an elementary work like the present": Joshua Williams on Real Property, 17th ed., by Mr. T. Cyprian Williams, 1892, p. 566. 174 THE LAND LAWS chap. registration is that it is not the apparent interest of the landowner, save in exceptional cases,^ to register. It is not found in practice that the selling price of land is sensibly diminished by its being sold under special condi- tions of the ordinary kind. In other words, what is called a " marketable title," that is, a title such as the Court would compel a purchaser to accept upon an "open con- tract," is worth nomore to the owner than a "good holding title," such as the vast majority of titles are. The tempta- tion of acquiring a statutory " marketable title " is there- fore spread in vain in the sight of English freeholders. Practically the problem is unsolved. At least three distinct schools of opinion exist among competent per- sons, not counting those who are satisfied with things as they are. Some are for registration of title ; some (now but few) are for only keeping an official record of dealings vnth the land ; others hold that nothing effectual can, or at any rate will be done until the sub- stance of our land laws is greatly simplified ; and per- haps these last are in the right.^ In recent years Land Transfer Bills have been promoted by Chancellors of both parties and approved by the House of Lords. They have included what may be described as a mild and modified compulsion, and a provision, adopted from the Australian colonies, for insuring registered owners against fraud or mistake by means of a special fund. This last provision is thought to be of very great utility and importance by those who have studied the " Torrens system " in detail. * As where contiguous lands acquired under different titles are to be thrown together, and sold or let in small parcels. ^ For references to recent literature on the subject see Note H, vii MODERN REFORMS AND PROSPECTS 175 Local registries of dealings with the land, not of title, were established in the last century in Middlesex and Yorkshire, and are still in use. Unfortunate de- cisions of the Court of Chancery made these county registers far less efficient than they were intended to be, and the Middlesex Eegistry is for this and other reasons worse than useless.^ The Yorkshire Registries (one for each riding) have worked better ; they have been remodelled by Acts of 1884 and 1885. Registration of assurances is well known in other English -speaking countries ; it has long been in force in Ireland, in Scot- land (where the system acts to a great extent as a registry of title also), and in several of the United States. Registration of title exists in most Continental countries, if not in all, and in our own Australian colonies. In England itself every manor has a register of the title to the copyhold property held of it, namely, the court rolls. But the law and the circumstances of English real property are so peculiar that the state of other lands in this respect is rather an object of envy than a presently practicable example for imitation. Neither can the archaic practice of copyhold tenure help us much. One thing seems tolerably clear : English landowners cannot both eat their cake in the form of secret conveyances, informal modes of raising money, and complicated settlements, and have their cake in the form of cheap and easy transfer of land. Either way has its advantages and its drawbacks, but we cannot follow both at once. The country must choose between them. Sir Henry Maine has pointed out that the estab- lishment of a complete registry would tend to have con- 1 See Land Transfer, 1886, p. 21. 176 THE LAND LAWS chap. siderable ulterior effects on legal doctrines and concep- tions. The passage ^ should be read and considered by every one interested in the subject. Meanwhile, we remain in the condition very fairly described in one of the answers sent in to the Real Property Commissioners : "It is possible to attain to such a degree of certainty as commonly satisfies a prudent man in the ordinary business of life, but with considerable delay, difficulty, and expense ; the great evil is not that titles are un- certain or unsafe, but that the investigation of them is difficult, tedious, and costly.'' Not many titles are really bad, but under the existing law there must always be an unknown element of risk. The acquisition and proof of title to land itself, and to rights over the land of others, have been put on a more rational footing by the Prescription Act of 1832, and the Real Property Limitation Acts of 1833 and 1874. Broadly stated, the results are as follows: Twelve years' possession of land (whether it begins vnth an appearance of right, or by mere intrusion or " squatting ") gives a good title as against claimants not disabled by infancy or the like from asserting their rights for themselves ; and for the benefit of such per- sons or others claiming in their right, a further time of six years is allowed, counted from the removal of the disability or the death of the person subject to it ; but in no case may the period open for the claims of persons out of possession be extended beyond thirty years in all from the time when possession begins to count against a freeholder entitled to an estate of inheritance. But where there is a tenant for life in possession his acts or ^ Early Law and Custom, pp. 351-61. vn MODERN REFORMS AND PROSPECTS 177 neglects cannot affect the title of those in remainder, and therefore when land is in settlement a much longer period may elapse before a title resting merely on adverse possession can become safe. Rights over the land of others are established by twenty years' unin- terrupted enjoyment if they are easements (that is, such rights as that of passage or of having free access of light and air, which do not involve taking anything from the land), thirty years if they are profits, that is, rights of taking something off the land, of which the most familiar case is a right of common. Possession is, of course, not available as a ground of title against a person whose superior title is acknowledged, by the payment of rent for instance, by the person in actual possession. This branch of the law, even with the modern improvements, is not an elementary one ; and it will be understood that there are many special points and distinctions of which I do not profess to say anything here. It is an arguable question whether the establishment of titles by long possession is consistent with a complete and efficient system of registration. In Scotland, where there is such a system, there is nothing answering to our Statute of Limitation as regards land. And in England the Land Transfer Act of 1875 has actually excluded the operation of the Statutes of Limitation as regards land which has been put on the register. No amount of adverse possession will prevail against a registered title. This provision, which I believe is not generally known even to lawyers, makes, in the opinion of one of our most learned and experienced real pro- perty lawyers, " a vast and very mischievous change in 178 THE LAND LAWS chap. the law." 1 At present " in English law, all title to land is founded on possession," ^ and has been so for at least seven centuries. The reversal of this principle in the case of registered land is likely to give rise to many difficulties apart from the danger of actual fraud. It stereotypes all unsettled questions of boundaries, whereas under the existing and well -understood practice they are constantly disposed of by friendly adjustment, often without even a formulated agreement, and in a genera- tion or so lapse of time cures all. Those who think that registration of title is not compatible with e2q)ecting purchasers to satisfy themselves that the land is held in accordance with the written title, as a prudent purchaser always does now, seem to forget that title is not the same thing as ownership. They have attempted to create a system of registered ownership before the legal profession or the public is ripe even for registered title. More will be heard on this point before the operation of the Land Registry is made general or compulsory. We come now to alterations more nearly touching the substance of the law. In 1833 the law of intestate succession was amended in various details. It does not seem worth while to give a particular account of either the old or the new law. The death of a fee-simple holder of land without making a will is in modern times an exceptional case. Popular knowledge of the law of descent is confined to the fundamental rule that the eldest son (or his descendants as representing him) takes the whole of the land to the exclusion of younger ' Sir Howard Elpliinstone in Law Quart. Bev., xi. 360 (Oct. 1895). ^ Williams on Real Pi-operty, 17tli ed., p. 631. VII MODERN REFORMS AND PROSPECTS 179 children ; ^ while daughters, on the contrary, share equally in the absence of sons. Nor is it often necessary for the practising lawyer to remember the further rules of collateral succession. The prevalence of settlements, and the complete freedom of disposition by will that has existed since the Restoration, have caused this whole head of the law to shrink into comparative insignificance. Nevertheless, proposals have from time to time been made to change our unique law of inherit- ance as regards land by assimilating it to the law of succession of personal property, which is substantially the same as the Eoman law in its final form and the present law of the rest of Europe. Throughout the States and colonies which took their law from England this change has already been generally made. At home these proposals have been strenuously resisted, and hitherto they have always failed. Why has a point of comparatively slight intrinsic importance been thus made a centre of active controversy ? Simply because the point actually at issue has been understood on both sides to be the symbol of much more. The attack was in form directed upon a rather small anomaly ; but in substance it was aimed at the privileges of landowners and the custom of settling lands from generation to genera- tion on the eldest son. If the legal rule of primogeniture in the strict sense were abolished, the artificial primo- geniture of our family settlements could not long survive ^ His title may be subject to the right of his father's widow to dower, i.e. to the enjoyment of one-third of the land for her life. But this right has become in various ways, which it would be too long to specify, practically obsolete in England. In the United States it is still active, and is the subject of quite modern text- books. 180 THE LAND LAWS chap. it. Immovable property would become assimilated to movable both in law and in sentiment ; or to speak more correctly, the change in the law would be an all but conclusive mark that public feeling was changed already. As it is, by far the greater number of the well-to-do people who make wills (I say well-to-do, because so much may fairly be inferred from a man sending his will to be settled by a conveyancing counsel) throw their land into a common stock with the rest of their property for equal division among their children. What is called " making an eldest son " is the exception. Sometimes the proposal to abolish the rule of primogeniture on intestacy, leaving intact the power of disposal by will in favour of the eldest son or any one else, is confounded with a proposal to introduce compulsory division of the whole or some considerable part of the heritage, accord- ing to the system which has always prevailed on the Continent. An error of such magnitude barely needs, one would think, to be pointed out. The two plans belong to different orders of legal institutions. Since Tory Ministers have proposed the abolition of primo- geniture (this first happened in 1887), its most ardent supporters can hardly expect to see it last much longer. Already the Intestates Estates Act, 1890, gives land as well as personal property to a widow, if there are no children, where the husband has died intestate and worth less than £500 in all. Copyhold lands have been incidentally mentioned in their historical aspect in former chapters. Many of their incidents are relics of an antiquity to which the Norman Conquest is a thing of yesterday. But how- ever venerable its antiquity, and however interesting to VII MODERN REFORMS AND PROSPECTS 181 the student of historical jurisprudence . the evidences which it preserves, this tenure is found in many vfays inconvenient in modern practice ; nor is the advantage of copyhold titles being registered on the court rolls of the manor sufficient to outweigh the inconveniences. From imperfect identification of boundaries, and for other reasons, dealings with property of which part is freehold and part copyhold (a very common case) are apt to be exceedingly troublesome. The fines and other payments due from the tenant to the lord of the manor are mostly not a serious burden, but they may be so in exceptional cases. On the whole, the conversion of copyhold into freehold tenure, for which the proper word is enfranchisement, is generally felt to be desirable. It could always be done by agreement between the lord and the tenant ; but the lord had to be willing, and the tenant as well as the lord under no disability, to say no- thing of the possibility of disagreement as to the terms on which the lord's rights were to be extinguished. Under certain modern Acts of Parliament, of which the principal one was passed in 1852, enfranchisement is now compul- sory on the requisition of either the lord or the tenant. Provision is made for valuing the manorial rights if the parties cannot agree. By the operation of these Acts copyhold tenure is gradually but surely disappearing ; and its end is being hastened by yet further legislation. A series of Acts, now consolidated by the Copy- hold Act, 1894,^ have increased the facilities for enfranchisement in various ways. Every new copy- holder must have express notice that he is entitled to ' 57 & 58 Vict. c. 46. The latest and most important of the consolidated Acts was passed in 1887. 182 THE LAND LAWS chap. enfranchise Ms land on paying the lord's compensation and the steward's fees ; those fees are defined by a new scale ; additional provisions are made for the assessment of compensation to the lord, and paying it in the form of a reni^charge ^ with an option to the tenant to redeem at twenty-five years' purchase ; and in certain cases the Board of Agriculture (formerly the Land Commissioners) has power to enfranchise the copyholds of a manor bodily if such is the desire of two-thirds of the copy- holders. Any rights of common which may be attached to copyhold tenures are expressly saved from being prejudiced by the change of tenure. If one form of ancient customary law is becoming obsolete in copyholds, there is another branch of the same stock which, within the memory of the present generation, has been called into fresh life, and recovered much of its old importance : I speak of the law con- cerning rights of common. Such rights may be regarded as the remnant of the ancient communal system of holding and cultivating land. Being useless to modern agriculture, and inconvenient for the larger landowners, they were long looked upon with disfavour by both the judges and the legislature, and were in a way to disappear even faster than copyhold tenure, but for the discovery that they were a valuable means of securing open spaces for public exercise and recrea- tion. Two or three centuries ago, it is almost needless to say, there was no lack of open spaces even in London, nor any visible prospect of it. Wholesale inclosure, 1 Such rent-charges, though doubtless discoverable by applica- tion to the Board, may in course of time be lions in the path to subsequent purchasers, not knowing (as they sometimes well may not) that the land was ever copyhold. VII MODERN REFORMS AND PROSPECTS 183 begun in more or less irregular ways in the sixteenth century, was the deliberate policy of the social reformers and philanthropists of the eighteenth. To bring as much land as possible under cultivation seemed to them the just ambition of the landowner who would serve the commonwealth. As long ago as the thirteenth century the statute of Merton had authorised the lords of manors to " approve," ^ that is, inclose for their own profit, as much of the waste land as would leave enough uninclosed for the use of the commoners.^ Inclosures might also be made without limit if the consent of all persons entitled to rights of common was obtained ; but this was seldom practicable. Many commons were inclosed under local Acts of Parliament, and by various old statutes inclosure for particular purposes, such as planting timber, was encouraged. In 1801 a general Inclosure Act was passed (superseded by a completer ^ "Approve" in this sense is not the common \vord = Lat. approhare, neither does it stand for appropriare, as has been con- jectured with some plausibility. It represents an old French verb, aproer or approuer, to profit or enrich, from^'ew ov pivti, itself au obscure word, which in modern French survives in "ni prou ni pen " (of. preux, prouesse). In Latin of the late thirteenth century (Statute of Westminster and Fleta) it appears in reflective construc- tion as se appruare. " Se appruare de" . . . is therefore "to make one's profit of " . . . exactly what in the Statute of Merton, where the word itself does not occur, is expressed by the phrase "commodum suum facere." The corresponding substantive is "approment" or " approwement " in the English of the fifteenth and sixteenth centuries. In the seventeenth century the spelling approve, approvement, and with it the mistaken derivation from ap- probare, came in. See the words in the Oxford English Dictionary. ^ It has been doubted whether the law declared by the statute was really new : Joshua Williams on "Rights of Common, " pp. 107, 108; Elton on "Commons and Waste Lands," 177-86; but see Pollock and Maitland, Hist. Eng. Law, i. 612. 184 THE LAND LAWS chap. measure in 1845), and a standing commission was establislied to conduct and regulate the inclosure of common lands. The object of this policy was not merely to bring fresh land into cultivation, but also, and as regards the Act of 1845 chieily,^ to get rid of the old customs of cultivation by scattered parcels or shifting allotments in common fields. In 1845, how- ever, there was already a kind of presentiment of coming change ; for village greens were specially saved, and the inclosure of land near large towns was made subject to the consent of Parliament — a consent which some years later was made necessary in all cases. Provision was also made for the devotion of part of the land to public purposes. The Act, however, was still entitled "An Act to facilitate the inclosure and improvement of commons and lands held in common." The reversal of the inclosing policy may be dated from 1865. Towards that time the rapid increase of building in the neighbourhood of London conferred a new value on lands which had been worthless to the lords, and created a new danger to the people, who saw themselves being deprived of their playgrounds. The cause of the commoners became the cause of the public. Litigation ensued about various common lands, Hamp- stead Heath among others. In 1865 a Committee of the House of Commons considered the question of preserving open spaces near London, and the Commons Preservation Society was founded by private exertions. In the following year the Metropolitan Commons Act, ' This appears from the proceedings of the Select Committee in 1884. Probably most of the earlier special Inclosure Acts had the same character. Cp. Seebohm, English Village Community, p. 14. vii MODERN REFORMS AND PROSPECTS 185 1866, put a stop to further inclosures within the metropolitan police district, and empowered the In- closure Commissioners^ to prepare schemes for the management and regulation of the commons thus preserved. Some commons, Wimbledon, for example, were separately dealt with by special Acts. Meanwhile the movement for reviving and maintain- ing rights of common was extended into the country parts. At Berkhamstead an attempted encroachment on the lord's part was vigorously and successfully resisted. Mr. Augustus Smith, a tenant of the manor, who fortunately had the means of doing right to himself and his fellows, sent down men by train early in the morning, who in a couple of hours demolished about a mile of wire fence erected by the then Earl Brownlow, taking care (as persons abating an unlawful obstruction or inclosure always should) to do no unnecessary damage to the materials. In the litigation that ensued Mr. Smith satisfied the Court of Chancery that his action was lawful, and obtained a decree against any repetition of the encroachment. A series of contests followed, in which the commoners were for the most part, though not uniformly, victorious. In 1871 the rule was declared and acted on by Lord Hatherley, then Chan- cellor, that, where rights of common were shown to have been claimed and exercised for many years, it was the duty of the Court to find, if possible, an account of their origin which would justify them in law. Not that Lord Hatherley professed to lay down any new principle, but there can be no doubt that in fact he ^ This body has been transformed first into the Land Commis- sioners and then into the Board of Agriculture. 186 THE LAND LAWS chap. dealt with the case in a very different spirit from that of the judges of the Elizabethan age.^ In 1874 the case of Epping Forest was taken up by the City of London. Here the peculiar and once oppressive rights of the Crown under the old forest law were made to play an unexpected part in restraining encroachments on the common. Owners and occupiers of land within a royal forest were .subject by that law to various burdens and restrictions ; thus they might not have fences above a height fitted to allow free passage to deer ; and by way of compensation they were allowed rights of common, subject to the forest rules, over the whole waste of the forest without regard to the boundaries of particular manors. This was an insuper- able bar to the lord of a small or decayed manor law- fully making himself absolute owner of the waste, as has not unfrequently been done elsewhere, by buying up the remaining rights of common. By a mere accident the Corporation of London were owners of a small property within the bounds of the Forest, and were thus entitled to undertake, as no private commoner could effectually have done, the research and expense necessary to establish their rights to the full extent. At the instance of the lords themselves, who thought ^ Warrick v. Queen's College, Oxford, L. R. 6 Ch. 716. In the later case of Goodman v. Mayor of Saltash, 7 App. Ca. 633 (1882), a still further application of this principle was made by the House of Lords. The decision established a right (nominally for inhabit- ants of Saltash, practically for the public) to take oysters in the Tamar. I am disposed to think the reasoning by which it was arrived at may be found to support claims to rights over land also, which have hitherto been supposed incapable of legal existence ; this, however, is too speculative a question to dwell upon here, nor have any recent decisions thrown light upon it. vii MODERN REFORMS AND PROSPECTS 187 they were choosing the lesser risk, measures were taken by Parliament for preserving the waste land not yet inclosed; but the lawsuit of 1874 (the course of which Parliament refused to stop) showed that many of the inclosures made within the previous twenty years, and untouched by the Act of Parliament, were unlawful. In the result a considerable amount of space (estimated at more than 2000 acres) was not only preserved but restored to the public.^ Meanwhile the Parliamentary Commission proceeded with its work, and in 1878 a new Act of Parliament constituted the Corporation of London conservators of the Forest, with full powers of regulation and management, and with the charge of preserving it "uninclosed and unbuilt on, as an open space for the recreation and enjoyment of the public." ^ In 1876 a general Commons Act was passed, which took the great step of establishing a presumption for rural as well as suburban commons in favour of regula- tion rather than inclosure. Improvement is expressly defined so to include adding to the beauty of a common, a thing never dreamt of in the former Inclosure Acts. Special provision is made for giving facilities for recrea- tion and preserving objects of historical interest. In short, the Act, even where it leaves most to administra- tive discretion, is an emphatic instruction to the Inclosure '* See for further details the Report of the Commons Preserva- tion Society, 1877, which is to be had for 6d. 2 41 & 42 Vict. u. ccxiii. Outstanding questions were by the same Act referred to the absolute arbitration of Sir A. (now Lord) Hobhouse. " All questions of importance in relation to the Forest are now settled, and the arbitrator has made his final award, speci- fying by a map the lands confided to the care of the Corporation of London, and dedicated to the public enjoyment for ever." — (Report of Commons Preservation Society, 1882.) 188 THE LAND LAWS chap. Commissioners to reverse the policy of the earlier part of the century. Among other points which mark a new- spirit, it makes encroachment on village greens a public nuisance. The general interest in these matters has con- tinued to increase. Local associations for the preservation of commons and footpaths have been formed in several parts of the country. In some cases they have effectually withstood encroachment ; in others the mere knowledge of their existence has probably prevented it ; sometimes they have been able to promote the adjustment of dis- puted rights in the interest of the public and all parties. Finally the Statute of Merton, though not repealed, has been rendered practically inoperative, "approve- ment " being no longer valid without the consent of the Board of Agriculture. '^ A new form of encroachment has lately been found to require new vigilance. Eailway companies and such like bodies are prone to put forth their hands on common rather than on inclosed land when there appears to be any choice, that they may have the less compensation to pay ; and their managers naturally take little thought for preserving the beauty or the quietness of the country affected by their schemes. But several projects of this kind have been signally discomfited in Parliament ; and when, at the twenty-first meeting of the Commons Pre- servation Society in 1886, the work done by it since its foundation was summed up by Mr. Shaw-Lefevre and Mr. Bryce, Mr. Bryce was able to say " the railways have begun to fear us." ^ ^ Law of Commons Amendment Act, 1893. The whole history may be seen more at large in Mr. Shaw-Lefevre's English. Commons and Forests. ^ Report of Proceedings, 1885-86, p. 34. VII MODERN REFORMS AND PROSPECTS 189 Important as the reaction against inclosure of open land may justly be thought, it has, on the whole, been effected through change in the spirit and working of the law rather than by changes in the law itself. One remarkable case of recent change remains to be noticed, — a case where, after the law had repeatedly been de- clared by persons speaking with authority to be perfect and in no need of reform, a large measure of reform, designed by the greatest lawyer of the Conservative party, was accepted almost without a whisper of opposi- tion. We have seen what grievous dislike was incurred by strict settlements of land when they were a novelty. Coke and Bacon, opponents in most things, agreed in denouncing them. Pamphleteers wished " that there might be no estate but absolute, for life or inheritance, without conditions or entails . . . and this would shorten all suits about estates." ^ Until the close of the seventeenth century the judges followed the Chan- cellor's allowance of the new-fashioned dispositions with undisguised reluctance. But when once the rule against perpetuities was fixed, the landowners and conveyancers had things all their own way for some generations. The scheme of settlement devised by certain ingenious coun- sellors of the Restoration passed, in the esteem of land- owners and even of lawyers, into the category of things immemorial and sacred, bulwarks of Church and State. The Eeal Property Commissioners could find no fault in it. Lord St. Leonards, a score of years later, was of the same mind. Objections could, in his view, proceed only from ignorance or perversity. In such words as these he expounded to the English landowner (the ^ 3 Jurid. Soc. Papers, 598. 190 THE LAND LAWS chap. farmer had then hardly entered into his reckoning) the beatitudes of an English settled estate : — " A desire has often been shown, not merely to im- prove the law of real property, but unnecessarily to alter it, and to admit only simple settlements after the fashion of the Code Napol6on. But the present plan of a strict settlement in this country is free from all objection. It does not place land extra commercium, but within reason- able limits enables the owner to transmit it to all his posterity ; and from its very nature leads to successive settlements, which alone have kept many estates in the same families. . . . " Thus are estates quickly resettled, and the State does not, that I am aware of, suffer any inconvenience from such repeated settlements. No man in this country can justly complain that there is not sufficient land in the market on sale. . . . " It is objected that these purposes are effected by a complicated and an expensive machinery ; but who ever complained of the complex movements in a well-finished watch ? "We admire the connection of its parts depend- ing on each other, and all necessary to form the combina- tion which produces the desired results." Metaphors are dangerous things. A well-finished watch is desirable not for the beauty of its machinery, but that its wearer may know the time of day. He will prefer a cheap watch that keeps fair time to a sidereal chronometer which is out of relation to men's common count of hours. Moreover, a watch is far from being a perfect machine ; the watchmaking trade is peculiar in its constitution and division of labour, and improvements have been introduced only slowly and with great diffi- \n MODERN REFORMS AND PROSPECTS 191 culty. But this elaborate combination, it is said, " pro- duces the desired results." Desired by whom and for whom ? There is the point. Behind the question of legal machinery there is the question of social policy, which has gradually been forcing itself on men's attention, Lord St. Leonards and other persons of authority notwithstanding. The results of a strict settlement are doubtless desired by its makers, and the purpose is effected with exceeding skill. But are they desirable for the common weal ? Dealing with the land is hampered; not made impossible or impracticable, as sometimes appears to be thought, at least not if the settlement is framed by competent hands ; but hampered it is, and it is idle to deny it. Powers of sale and management are useful, but they are not ownership. I have power to consult any book in a great law library by walking a few score paces from my chambers. But if there is a book I should rather like to consult, but can do without, the chances are that I shall not go. Every scholar knows the difference between having books at one's elbow and having to go out to see them. So it is with attempts to restore the uses of absolute ownership after dismembering its substance. And there is a still graver economical objection to strict settlements than any real or supposed want of power to deal with the land. It is that they destroy the ordinary means and motives of an owner for dealing with it in the way of improvement. The life-tenant of a settled estate, " limited owner," as he is expressively called in modern Acts of Parliament, seldom has any capital to put into the land. Incumbrances created by previous settlements commonly leave no margin of 192 THE LAND LAWS chap. revenue beyond what is needed for barely keeping up the estate. And if the limited owner has other means, he is still without much inducement to apply them to the improvement of the estate ; if he does so, he spends for the exclusive benefit of an eldest son who is already preferred by the settlement. He is much more likely to use his money for the benefit of younger children ; and justly, so far as they and he are concerned ; but this divorce of capital from land is a loss to the land itself, to the cultivator, and in the long run to the commonwealth. Much has been said and written of the evils of strict settlement since the renewal of law reform which began about sixty years ago, and sundry partial measures have been taken to palliate them. It is needless to describe these, for they have been mostly superseded by the Settled Land Act of 1882, due to Lord Cairns as its chief author, the object of which is to confer on " limited owners '' as large and effectual powers of using the land to the best advantage as are compatible with settlements existing at all. By this Act a tenant for life ^ has powers of selling and leasing, and other powers of administration, without any special provision in the settlement, and, subject to certain safeguards intended to prevent abuse, and not burdensome, he may exercise them at his own discretion. He cannot, however, sell a principal mansion-house or heirlooms without leave of the Court. His powers cannot be released or bargained away, and no disposition purporting to abrogate or 1 Certain other limited owners, whose position differs only technically from that of a tenant for life, are included in the Act. There are later amending Acts of too detailed and technical a nature to call for notice here. VII MODERN REFORMvS AND PROSPECTS 193 restrain them is valid. The purchase -money of land sold under the Act may be spent (among other purposes) on the discharge of incumbrances, or on improve- ments of several specified kinds. Expenditure on im- provements is subject to the control of either the Land Commissioners (a body constituted by the amalgamation of the former Inclosure, Copyhold, and Tithe Com- missions) or the Court. No power is given to raise money from the land by mortgage as distinct from sale in order to execute improvements on it; this can be done, by a somewhat tedious process, under an Act of 1864. The list of authorised improvements, however, is extended by the Settled Land Act. It must be ob- served that the Act does not alter the beneficial title to the settled land, or the investments representing its value, or the increased value given to part of the land with money obtained by selling another part. The tenant for life cannot enlarge his own interest ; he can only decide in what form the property shall be handed on to those who come after him. He may leave them fewer acres with equal or greater value, in the shape either of incumbrances taken off or of improvements added. He may even leave them money instead of land. So far he can alter the subject-matter of their rights, but the rights themselves he cannot touch. It was reserved for another Tory Lord Chancellor in 1887 to carry through the House of Lords a proposal for the automatic expansion of estates tail into fee simple. Even this, however, would in the main operate as a simplification of form. The authors of the Land Trans- fer Bill professed not to interfere with the general principle of family settlements. 194 THE LAND LAWS chap. It is still too early to say what will be the economi- cal effects of the Settled Land Act. No such carefully devised measure of administrative reform in our land laws has been passed since the Fines and Recoveries Act. Probably its framers have succeeded in their intention of removing every obstacle to dealing with settled land that is not involved in the very existence of settlements. Experience must show how far the coimtry will be satisfied. Reforms of this kind are most useful for their time, and their promoters deserve all honour. But they all add to the complications of a system abeady too complex ; and the day seems to be at hand when the system will no longer bear this process, and recon- struction must be faced. The traditions of English character, the circumstances of an age of commerce, and the example of other Eng- lish-speaking communities, all seem to point to the sim- plification of landholding and the encouragement of absolute ownership. On the other hand, there are cer- tain speculative reformers, of late very loud and busy, who will have it that private ownership of land is altogether wrong, and instead of desiring that, so far as may be, every man should be his own landlord, are for making the State every man's landlord. To discuss such schemes is not within my province. As Lord Hobhouse has well said,^ " they are hardly proposals for the alteration of the laws relating to land, or of any other branch of municipal law commonly so called ; but rather for shifting the very basis of English society, and remodelling it on different theories of property, and of the relation between the State and individuals." 1 Tlis Dead Hand, 1880, p. 164. VII MODERN REFORMS AND PROSPECTS 195 It will be time for lawyers to take a serious view of what is barbarously called " Land Nationalisation " (but it deserves a barbarous name) when its advocates have shown themselves capable, as thus far they have not, of appreciating the enormous difficulties that would beset an attempt to give tangible form to their vast and vague idea. No law could be framed for destroying private property in land, or (what some of these reformers take to be the root of mischief) the relation of landlord and tenant, without hundreds of keen-witted men forthwith setting themselves to evade it. The history we have surveyed has given us, perhaps, examples enough of what happens when the letter of laws is matched against the nature of man and of things. As to the question of public economy, I have nothing to say of it. Not only is it not my business here, but I cannot imagine that those who were not convinced by Mr. Fawcett ^ are likely to be convinced by anything I could add. It is said that working men in London are found apt to be taken with these projects of national socialism. It may be so ; too few of them are ballasted in their course by much, or any, of the magic of owner- ship. But the artisans of the north country are already in great part, through the operation of building societies, full owners of the homes they dwell in. Let such a working man understand that our new social reformers will confiscate his well-earned cottage neither more nor less than the squire's mansion and the manufacturer's mill — that he too is to become a rent-paying tenant of the State, or a so-called occupying owner taxed up to ^ ' ' State Socialism and the Nationalisation of Land, " Macmillan's Magazine, July 1883 (since republished in pamphlet form). 196 THE LAND LAWS chap, vii the rental value of his holding — and I think he will prefer his risks of poverty to such progress as the social enthusiasts of the East or the West have to offer him. The extreme Socialists, after threatening in brave words to extinguish both the great historical parties of our Commonwealth, have now (1895) been signally rebuffed by the electors of Great Britain,^ and have lost what little representation in Parliament they had snatched in the last few years. We may be permitted to trust that the progress of English laws and society will be, as it has been, in a steady course of rational reform ; that men of power and wisdom may be with us in time of need in the future as they have been in the past ; and that the terrible fascination of revolution may remain, as it yet is, alien and impotent among our people. Dea magna dea Cybebe, dea domina Dindymi, procul a mea tuus sit furor omnis, era, domo : alios age incitatos, alios age rabidos. ^ In Ireland there has never yet been room for them. APPENDIX Note A. — The Germanic Land System Few passages have been more discussed than the short account of early German agriculture in the Germania of Tacitus (c. 26). The best exposition is still, perhaps, that of Waitz {Deutsche Verfassungs-Geschichte, vol. i. p. 104 sqq., ed. 1865). There are some useful notes in Mr. H. Furneaux' recent edition of the book (Oxford, 1894). The critical words are these : — " Agri pro numero cultorum ab universis viois [al. in vices] occupantur, quos mox inter se secundum dignationem parti- untur. Facilitatem partiendi camporum spatia praestant. Arva per annos mutant, et superest ager." I would translate to this effect : "The German township as a whole takes up a tract of land according to the number of its husbandmen. The land so occupied is then allotted among the members in proportion to their rank, the extent of open ground making this process a simple one. The parcels under the plough are shifted yearly, and there is land to spare." First, what does universi mean 1 It might mean a whole tribe or nation, a State, as we^ may fairly call it, if we remember that it was not organised like the modern State. But the partition of lands, newly occupied or otherwise, between individual cultivators is the work not of the State, but of the smaller community which appears on various scales and under various names, and for English purposes may best be called a township. It is possible, and indeed 198 THE LAND LAWS probable, that from tbe earliest times not only the kings but other great men had large portions of the land allotted to them in separate lordship. Such men were not members, but lords of the communities which might be settled on their lands. But it is also probable from what follows that it is a communal system that Tacitus here has in view. The alternative reading has much more MS. authority for it, vids being the reading of only one MS. But, though it is defended with great ingenuity by Orelli (1848), and is tempting in some ways, it makes a difficult and ambiguous sentence. It would imply that (1) Tacitus was at a loss to give a more definite name than universi to the German agricultural unit ; (2) the same ground was successively taken up by different communities. We should thus get, at first sight, a closer agreement with Caesar's account (B. G. vi. 22, quoted in Stubbs's Sel. Ch. p. 53). But it is by no means clear that Tacitus meant to agree with Caesar, who appears rather to be describing clans still in a nomadic stage, while Tacitus seems to be speaking of the permanent settlement of fresh tracts, " new takes," as they say on Dartmoor. As to the process of partition, it is clear that already there were distinct degrees of rank, that the more worshipful man got more land allotted to him, and the less worshipful less. " Arva per annos mutant " is taken by the best modern authorities to refer to the course of husbandry, and signify the alternation of crop and fallow. But it may also import (what seems otherwise probable) that every man had, or might have, a different parcel allotted to him each year. In like manner " superest ager " may refer to the unallotted common land of the township, or to vacant land not yet appropriated by any township, or to both. Fustel de Coulanges considered this passage of Tacitus in his Recherches sur quelques probl^mes d'histoire, Paris, 1885. He thinks that Tacitus wrote it for the information of Eoman readers as farmers rather than as publicists, and that this accounts for the want of technical definition which has given so much trouble to modern commentators. " Tous les termes se rapportent k la culture." It is quite true that the leading word occupantur expresses a bare fact, and does not involve APPENDIX 199 any particular inference as to the kind of right in which the acts are done ; on the otlier hand it does not exclude any which is otherwise plausible. References to recent criticism of Fustel's work may be found in the Introduction to the French translation of Maine's Village Oommunities, sub tit. Etudes sur I'histoire du droit, Paris, 1889. An article in the Quarterly Review for October 1895 opens an interesting field of comparison, and a new one to most English readers at any rate, on the history of land communities in Spain. Note B. — Symbolic Transfer in Early English Customs Professor Sohm, following K. Maurer and Schmid, assumes that the A.S. " book " was the common mode of assurance, and goes so far as to deny that symbolical transfers were used at all before the Conquest. (Frdrikisches Recht und romisches Recht, Weimar, 1880, p. 30.) But the absence of records appears quite inconclusive. It was not a native Anglo-Saxon or Germanic custom, but a foreign practice introduced under clerical influence " iuxta exempla Romanorum," to keep any written records at all. Besides, it is not the case that we find before the Conquest "keine Spur der Investitur." There are half a dozen examples in the God. Dipl. See Nos. 12, *37, 104, 114, *177, 1019. In No. 114 a sod from the place is expressed to be delivered along with the book. In the other oases it is mentioned to have been placed on the altar or on the book of the Gospels. Two of the charters, here also noted by an asterisk, are marked by Kemble as forgeries. But the like incident must have been found in genuine originals which the forger was imitating. Two other cases may be added from the Black Book of Peter- borough, in a record printed by Dr. Stubbs in 1861, and accepted by him as genuine. Early in the eighth century " jEthelred, the glorious king of Mercia, on the occasion of a visit to Medesham-stede, gave to the brethren he found there thirty manentes at Lengtricdun, and confirmed the gift by placing on the Gospels' Book a sod taken from the place." Again, a purchase of lands at Cedenanac from the 200 THE LAND LAWS king " was ratified at Tonitun (Northampton ?) in the king's chamber by joining of hands, and by placing a sod from Cedenandc on the Gospels' Book, in the presence of Bishop Saxnlf." Here we seem to catch the very moment of the final struggle of the older symbolical ceremony with the " book," which for great occasions was to supersede it — if, indeed, it did so ; for, as Palgrave has observed {Proofs and Illustrations, ocxxviii.), the silence of most charters as to any symbolical delivery really proves nothing. My own belief is that for the common occasions of private persons the symbolical transfer never went out of use. It seems to me no extravagant supposition that many of the symbolical customs still found in copyholds, such as surrender by a straw in the manor of Winteringham, in Lincolnshire {Academy, Nov. 19, 1881, p. 386), are really of immemorial antiquity. On Sohm's and Schmid's theory they would be late medieval imitations of the Franco-Norman custom. It seems far-fetched, however, to find in the laws of Alfred an attempt to convert book-land into family or customary land by way of reaction towards pure Teutonic principles (Mr. Lodge in Essays in Anglo-Saxon Law, pp. 70, 71). The law says that a man who has inherited book-land from his kindred must not give it from the family " if there be writing or witness that it was forbidden by those men who at first acquired it, and by those who gave it to him, that he should do so." Probably this was only a confirmation of existing law (compare the preamble), and it rather goes to show that restrictive clauses were often disregarded than anything else. But it does seem to allow validity to restrictions of this kind not expressed in the book itself, but only declared by the donor in the presence of witnesses — a point not noticed by Mr. Lodge. So far we may say that the peculiar quality and privileges of book-land were kept in check by the old family principles. Note C. — Thb Origins op the Manor It may be useful to state in a summary way the different APPENDIX 201 theories as to the origin of manorial tenures and customs which have prevailed among English authors at different times. For shortness' sake they will have to be put rather in their extreme forms than with the reserves and q^ualifica- tions which their more cautious maintainers have introduced. But this will not involve any injustice, for the object is not to give an exact account of this or that author's opinions, but to indicate the successive tendencies of historical specula- tion, the changes of fashion, as one might say, if the dignity of history siiffered it. Those who desire a more careful and detailed study of the different schools may find it in the Introduction to Mr. Paul Vinogradoff's Villainage m England, and Dr. Andrews' work on The Old English Manor.^ Down to the middle of this century, or later, the common opinion, founded probably on Blackstone's account, was that at the date of the Norman Conquest, and for some centuries afterwards, the greater part of English land was held at the absolute will and pleasure of the lords of manors. The medieval customs evidenced in surveys and court rolls were supposed to have arisen (so far as any rights of the tenant were concerned) out of mere sufferance, " by a long series of immemorial encroachments on the lord." ^ It was not uncommonly stated or assumed that before the Conquest the land was to a great extent in the hands of free yeo- men or peasant proprietors. These were supposed to have been either evicted or reduced to villenage by the Norman lords, leaving little or no trace in records of the Anglo- Norman or any later period. Kemble's great work, The Saxons in England, was pub- lished in 1848, but the opinions of his school were not fully established in the acceptance of scholars and the general use of teachers till about twenty years later. However, the acceptance, when it fairly set in, was so complete and un- 1 Mr. Scrutton's Commons and Common Fields (Cambridge, 1887) may also be usefully consulted. ^ Blackstone's Comm., i. 95. It is not clear whether Blackstone thought this process dated from before, the Conquest or not, but I think he would not have put it much before in any case. 202 THE LAND LAWS questioning as almost to invite a reaction. According to this school the original type of the Germanic community was, among the Saxon conquerors of Britain as elsewhere, an association of free men equal in rights, if not in social rank or wealth. The village or township was a group of joint owners, acknowledging no superior except for military service and other purposes of public order and justice, and regulating its own internal affairs. Personal lordship, how- ever, was among the early Germanic institutions, and territorial lordship was already growing up some con- siderable time before the Conquest ; i in fact feudalism was impending, and the Conquest only accelerated the reception and definition of feudal rules. Probably no serious writer of this school would have denied that many details remained obscure as to dates and otherwise. Nevertheless it was agreed in the main that, whether the process began earlier or later, servile tenants did not encroach on the lords as supposed by Blackstone and the modern law-books, but on the contrary the lords encroached on rights and liberties of the occupiers, which were more ancient than their own. It was not disputed that the Saxon freemen, like the citizens of a Greek or Roman city, had below them a considerable class of unfree men, slaves in fact. This indeed has never been disputed by any one, though in some popular descrip- tions of the free village community the existence of slavery may have been allowed to fall a good deal into the back- ground. The German word " mark " and the phrase " mark system" were freely adopted by writers of this school, and their doctrine is often called the " mark theory," though there is really no English authority for the use of " mark " as a synonym of " village community " or " township." Maine's Village Communities is justly esteemed the best general exposition of the theory for English readers, and this for the very reason that Sir Henry Maine professed, ^ In Kemble's view (Saxons in England, i. 307), "the ruin of the free cultivators and the overgrowth of the lords " had already gone far before the time of .fflthelred. In other words, the ancient free institutions had broken down at least a century before the Normans came. (Cp. Freeman, JV. C v. 462.) APPENDIX 203 as regards European institutions, only to take the results of the prevailing Germanic school as he found them. It must be added that this school has never regarded the village com- munity, township, or " mark," as a mere association of other- wise unconnected households or individuals, like a modern club or commercial company. Some bond of kinship, traditional if not absolutely historical, is always tinderstood to have made the association possible in the first instance. Kemble and Maine are both express, nay emphatic, on this point.i It is needless to remind educated readers how in- timately it is connected with the importance of Adoption in archaic societies. A fictitious brotherhood was the founda- tion of the Germanic comitatus, and played a considerable part in the early stages of feudalism, as M. Flach has lately shown. The Blackstonian theory (as we may call it for con- venience) had taken its facts from the legal formulas of the later Middle Ages, as interpreted by the still later classical text-writers of English law. The Germanic theory, dis- carding this unreal treatment of a large and complex historical process, corrected the former view, as we have just seen, by taking account of political and economic develop- ments, and bringing the purely English history into line with the general mass of Teutonic learning and tradition. But the later extensions of this method in such hands as Nasse's ^ prepared the way for a critical reaction. It was found that close attention to the details of medi- eval agriculture and administration raised new problems which the Germanists, if not as unconscious of them as the Blackstonians, had left in the background. Anglo-Saxon documents, when they do go into detail, show us the tillers of the soil in a condition not materially different from the villenage of the thirteenth century. The labour-dues, well- known in post-Norman inquests and court rolls, are there ; a lord to whom they are rendered is there ; the highly ^ See especially Kemble's Appendix A on patronymic place- names. 2 The Agricultural Community of the Middle Ages. (Eng. trans, by Col. Ouvry, Cobden Club, 1871.) 204 THE LAND LAWS artificial system of common-field cultivation, though, not described in terms, is plainly there too when one has learnt to trace it.i Free communities are not visibly there, only supposed survivals of ancient freedom, which after all may be regarded as ambiguous. Hence, when Teutonic enthusi- asm and the brilliance of Kemble's exposition had spent their early charms, the question became inevitable : Was the free village community ever there at all? Why should not the usual medieval relations between the lord of a manor and his tenants in villenage be as old as anything else in the manorial system ? Is not the whole system really of one piece ? These and such like points were raised by Mr. Seebohm's work in this country. Soon after his English Village Community had begun to engage serious attention, M. Fustel de Coulanges independently attacked the German school on the Continent. In addition to destructive criticism of the German "mark theory," some writers, and notably Fustel, have undertaken to maintain that the supposed Teutonic village community is nothing btit a continuation of the Eoman villa, which beyond any doubt was a great man's estate cultivated by servile work. Neither Fustel's methods nor his results have been generally accepted in his own country. M. Paul VioUet, M. Flach, M. Dareste, and M. Glasson, have all criticised his work (partly by way of self- defence) with more or less severity, and have pointed out that, although he constantly and justly insisted on strict accuracy in distinguishing fact from conjecture, his own use of authori- ties was not always adequate or even exact. M. Flach is of opinion that even in Gaul there was nothing like a general survival of Eoman municipal institutions. It is to be observed, in any case, that Fustel de Coulanges never examined English facts or documents himself, nor, so far as I know, expressed any opinion whether his conclusions, which were strictly limited to the materials before him, would be applicable to the English settlement in Britain. Whoever wishes to rely on Fustel's work for English purposes ought ^ This is less material ; for if there were no Anglo-Saxon evidence at all it would be impossible to believe that such a system was introduced after the Conquest. APPENDIX 205 to be prepared to show that the determining conditions, as to the continuity of Roman institutions and otherwise, were not materially diiferent in Britain and in Gaul ; and Mr. Ashley, now of Harvard University, the only English writer who has followed Pustel without reserve, has not to my knowledge attempted anything of the kind. Thus the new " villa " school of early English history, if one may call it so by way of contrast to the " mark " school, is a Roman- ising one, and stands more or less committed to alliance with the late Mr. Coote and other students who have from time to time endeavoured to show that the persistence of Roman elements in our civilisation has been underrated. Much learning and ingenuity have been spent on this endeavour. Without entering here on the subjeot,i I am bound to state it as my considered opinion that on the whole this expendi- ture has been in vain. It is possible, however, to deny or not to be satisfied that free village communities existed in England within historic times, and yet not to accept the pro- posed derivation of the dependent community from the Roman villa. Maine has suggested with much force ^ that it may have been the Roman villa that was influenced by the barbarian village, Teutonic or other. It would be quite natural for the internal economy of a great Roman estate in the provinces, or even in Italy, to be modelled to some extent on the customs with which the majority of the slaves doing out-door work had been familiar in their old homes. The '' villa " theory may be said to be, to a limited extent, a restoration of the Blaokstonian or older manorial theory. So far as the facts from the Conquest to the thirteenth or fourteenth century are concerned, its upholders agree with Blackstone in regarding the law of the king's judges as a correct expression of those facts. But, instead of holding that both facts and law dated from the Norman Conquest, . this theory carries them back as far as the English conquest of Britain. There seems to be in this school a tendency, and almost an avowed wish, to encourage modern optimist views of social progress by representing the tillers of the soil as ^ Cp. Andrews, The Old English Manor, pp. 34-40. ^ Early Law and Custom, p. 332. 206 THE LAND LAWS moving slowly but steadily throughout the early Middle Ages from a worse to a better position. Its adherents are therefore really very far from having returned to the Black- stonian point of view ; they take in a much wider field, and aim at much more complete explanation. They have certainly done good by demanding and enforcing a fresh examination of the evidence. And Mr. Seebohm has been the first modern writer to exhibit and prove what a yardland in the common fields really was.^ Meanwhile the conclusions of this school cannot be said to have established themselves either here or on the Continent. Mr. Elton,^ Mr. Kenelm Digby,3 and, I think, most of our legal scholars, adhere to the Germanic doctrine in the main, though not to the in- cautious generalities of secondhand expounders which are still repeated now and again in semi-popular writings. From Russia the Germanic champions are reinforced by Mr. Vinogradoff and Mr. Kovalevsky, — strong allies both, Vinogradoff with a knowledge of the English documents, published and unpublished, in which perhaps only Mr. Maitland can rival him, Kovalevsky with wide command of analogous facts and developments in Slavonic societies at every stage of civilisation. Mr. Kovalevsky derives the Russian village community from the expansion of a patri- archal family,* and calls attention to the fact that indepen- dent and dependent communities have actually existed in Russia side by side in historical times. This is of great importance when equally universal and unqualified claims are made on behalf of either type as the proximate original of the English manor. ^ It would be rather interesting to know when people began to forget what it was, but this could be ascertained (save by good luck) only with more labour than it would be worth. In some manors, e.g. Berkhamsted, name and thing had disappeared as early as the reign of James I. ^ Article on Mr. Vinogradoffs Villaiiiage in England, Law Quart. Rev., viii. 117 ; and see "Early Forms of Landholding, " Eng. Hist. Rev., i. 427. ^' Introduction to the History of the Law of Real Property, 4th ed. Oxford, 1892. * Modem Customs and Ancient Laws of Russia, Lend. 1891 ch, iii. APPENDIX 207 I may add that the number and position of English land- holders who were certainly free at the time of the Domesday Survey, or at the date of King Edward's death, to which the Survey constantly refers, are among the elements to be most carefully considered in dealing with this problem of origins as a whole. That number was much underrated by Mr. Seebohm in consequence of his not allowing enough for the variety of Domesday terminology. The " thanes " of Wessex, e.g., are mostly not lords but rather small freeholders ; and many tenants of the same class are entered merely by name. Thus the supposed "fact that Domesday does not mention free tenants in the greater part of England " (Mr. Ashley in Econ. Rev., iii. 165) is an illusion. With regard to the manorial courts, Mr. Maitland's researches (Select Pleas in Manorial Gowis, published by the Selden Soc, 1888; and see Mr. G. H. Blakesley, "Manorial Jurisdiction," Law Quart. Rev., v. 113) have proved, in my opinion, that the .jurisdiction is feudal and derived from tenure, and independent .alike of any earlier popular institutions and of economic organisation. The distinction between the " court baron " of the freeholders and the " cus- tomary court " of the base tenants is a relatively modern formalism ; early medieval records know only of a curia legalis when there is any adjective at all. I now have no doubt that private jurisdiction is very ancient, and grew up not by transformation of any more ancient popular court, but in direct competition with the hundred court. There is no sufficient reason to think that there was ever any popular court less than the hundred court, and the " mark- moot," which students were for some time taught to believe in as if it were as well known as the King's Bench, is a creature of ingenious but unwarranted conjecture. From all this it is a long and dangerous leap to the conclusion that the origin of our medieval land system is to any considerable extent not Germanic but Roman. I am no more disposed to accept any such conclusion myself than I was when I first wrote this book. In 1893 Prof. Earle suggested in the Economic Journal that the syllable well or wil in English place-names may 208 THE LAND LAWS sometimes be a survival of a Roman villa. I know no reason a priori why villa should not have survived sometimes, as castrum certainly often did, but even if the evidence came up to the mark of conviction, as I can hardly think it does, I do not see that our general view of the history of civilisa- tion in this island would be much affected. Note D. — Villenage, Villein Tenure, and Coptholds The first condition for a clear understanding of the " base tenures" of our books is to keep in view the distinction between the tenure of the land and the personal condition of the tenant. For this, as recognised in the thirteenth century, the classical authority is Bracton. He mentions tenants on the king's demesne, whose ancestors held by free services, but were put out by the strong hand, and were allowed to take back their holdings to be holden in villenage by base but certain and assigned services (cp. as to this Dial, de Scaccario, I. c. x.). These tenants are personally free, " since they perform their services not in regard of their persons, but in regard of their holdings." Compare as to this the old English document called Bectitudines Singularum Personarum, § 2, p. 372, in Schmid's collection: the "cot- setla," who renders labour-rent such as after the Conq^uest was called servile, is personally free, for " he is to give his hearth-penny on Holy Thursday as every free man ought " ; so Kemble has well noted, Saxons m England, i. 323. To return to the demesne tenants in Bracton, they cannot bring the ordinary real actions, " but only the little writ of right according to the custom of the manor " ; that is, the jurisdic- tion was in the king, not as king, but as lord of the parti- cular manor. Yet a tenant in ancient demesne might have the assize of novel disseisin against a stranger, and under exceptional conditions he or his feoffee might have it even against the lord, see the opinion set out in Mr. Horwood's Preface to Y. B., 20 & 21 Ed. I. p. xviii. They were called glebae ascriptitii (cp. again the Dial, de Scacc., loc. cit.), but Bracton regarded this as expressing not a bondage, but a right ; not that they might not go elsewhere, for " they shall APPENDIX 209 not be enforced to hold such a tenement unless they will," hut that the lord cannot remove them as long as they per- form the services (fo. 7a). Though not properly tenants in socage, they are regularly called socmen ; cp. Old Nat. Brev. tit. "Briefe de recto olauso," fo. 11, ed. 1584. Again, " the tenement changes not the condition of a free man any more than of a slave. For a free man may hold in mere villenage, doing whatever service thereto belongs, and shall none the less be free, since he does this in regard of the villenage and not in regard of his person. . . . Mere villen- age is a tenure rendering uncertain and unlimited services, where it cannot be known at eventide what service hath to be done in the morning, that is, where the tenant is bound to do whatever is commanded him" (Bracton, fo. 26a). Again, "another kind of tenement is villenage, whereof some is mere and other privileged. Mere villenage is that which is so held that the tenant in villenage, whether free or bond, shall do of villein service whatever is commanded him, and may not know at nightfall what he must do on the morrow, and shall ever be held to uncertain dues ; and he may be taxed at the will of the lord for more or for less . . . yet so that if he be a free man he doth this in the name of villenage and not in the name of personal ser- vice . . . but if he be a villein [by blood], he shall do all these things in regard as well of the villenage as of his person" (fo. 2086). The only difference in the services was in the merchetum on marrying a daughter, which was an incident of personal servitude (being a fine paid to the lord for depriving him of a slave. As to the origin of the custom and word see Mr. Pike's Introduction to the Year-BooJc of 15 Edw. III. in the EoUs Series. I venture to guess that not only, as he suggests, the word has to do with mearc, and means a fine for marrying beyond the manor boundary, but the original form was mearc -sceat, cp. the Domesday form circet = dric-sceat, which removes an otherwise serious objec- tion : it occurs in Sussex, D. B. i. 296, and in Oxfordshire (twice) 1545). Merchet was not of right demandable from the free man holding in viUenage : "nee enim tenebitur ad merchetum de iure, quia hoc non pertinet ad personam liberi 210 THE LAND LAWS sed villani " ; and even on this point tlie evidence is not uniform as in later books. Bracton's de iure suggests doubt whether the practice was fixed in his own time. Privileged villenage is then described as the tenure of the personally free men already mentioned in the passage fi.rst cited. Bracton also mentions a kind of qualified villenage in which the services are defined by express agreement, " ex conventione " ; this is not properly called privileged, but the terms agreed upon are enforceable against the lord (fo. 209a). But I have some doubt whether these sentences belong to the original text. I suppose the personally free man who held in villenage, privileged or not, might always perform the services by deputy if he could find one. Cp. Bracton, fo. 199, 200, on the "exceptio villenagii," and cases 70, 80, 281 in Bracton's iVote-BooA:, ed. Maitland, 1887. Thus in Bracton it is quite clear that the conditions of the tenure and the personal status of the tenant have no necessary connection. Not only a free man may hold servile land, but a bondman may hold and deal with free land as against everybody but his own lord (fo. 266), and he may in general sue any one but his own lord, and in some cases even him (1556) ; cp. Pollock and Maitland, Hist. Eng. Law, i. 399. It is not uncommon for a group of unfree villeins to find one man for particular works : this would usually be one of themselves, but I doubt whether the lord cared who did the work so long as it was done. And in Littleton, and Coke's commentary on his text, the distinction is equally clear (compare Co. Litt. 576, 58a, with Litt. s. 172, and Co. Litt. 116, 117). There is already a source of confusion, however, in the double meaning of villanus. At the time of Domesday the serf by blood was called servus (later nativus), never villanus. But gradually, by a kind of euphemism, the word servus was dropped, and villanus came to mean sometimes, as aforetime, a man holding by a certain kind of service, but often a man personally unfree. In Glanvill we see the terminology in a state of transition. He always calls the bondman nativus, but his condition (not merely the tenure, but his personal state) is called villen- agium ; a reclaimed bondman is said to be in villenagio APPEISTDIX 211 positus, or in vilknagium clamatus. Bracton freely uses mllanus as synonymous with nativus. By Littleton's time the word villein seems, as a rule, to have imported personal servitude, though a careful man might speak of a " villein in blood" if he meant to secure himself against mistake. The distinction in substance, as above said, is still perfectly clear. Coke, perhaps, misinterprets Littleton in one place, s. 209. Littleton says that the lord of a manor cannot prescribe for a fine payable by every tenant within the manor who marries a daughter without the lord's licence ; " for none ought to make such fine but only villeins." Coke's gloss on this is as follows : " that is, either villeins of blood, or freemen holding in villenage or base tenure." Comparing Bracton's more explicit statement, it seems Littleton should have meant to exclude, not to include, freemen holding in villenage (s. 174, which suggests the contrary, and is cited by Coke, is not part of Littleton's own text, but an interpolation. See the note in Tomlins' edition, 1841). But Coke's view may be supported by Y. B. 10 Ed. IIL 22, pi. 41, and 43 Ed. Ill 5, pi. 13. And if he is wrong, it is only on a detail. When we come to Black- stone '{Comm. II. 92-96) the confusion is complete. He writes as if wholly unaware that villanus ever meant any- thing but a personal serf : the liber homo tenens in villenagio somehowi escapes his notice altogether, and the early copy- holder is represented by him as an enfranchised bondman and nothing else. Blackstone's confused and misleading account has been adopted, so far as I know, by all modern text-writers before Mr. Kenelm Digby, who, going back to Littleton and Bracton, restored the distinction which Black- stone had obscured. The blundering of Blackstone's Gom- mentaries on this point is the more remarkable, inasmuch as his Considerations on Copyholds show that he at one time read and to some extent appreciated the earlier authorities. Putting the tract and the Gommentaries together, his theory, so far as he had one, appears to have been that personally free men holding in villenage existed, but were a very small class, and were ultimately represented by the customary tenants not said to hold at the will of the lord ; that the 212 THE LAND LAWS ordinary copyholder always represented a nativus ; and that the tenure of the nativus was not only burdensome, but wholly precarious until long after the Conquest. This is a long way from the facts, though perhaps not much further than the opposite one of regarding the normal township as a self-governing corporation, and villenage as a post- Norman development or abuse of Anglo-Saxon serfdom. As to the personal condition of the nativi, modern scholars seem pretty well agreed that legally it might have been one of great hardship, but in practice they were for the most part not so very badly off. The least favourable view of their position I have met with is Nasse's, the most favour- able, Mr. Larking's (Domesday of Kent, note 57) and Mr. Thorold Rogers's {History of Prices, vol. i.). These last writers appear, however, at least as much concerned to make out the state of the modern labourer worse as to make out the state of the old nativus better. Mr. Thorold Rogers certainly underrates the positive evidence that personal servitude, as distinct from attachment to the soil, was known long after the Conquest. For example, it was elaborately discussed in 1302 whether a neif marrying a free man became free absolutely or only during the coverture (Y. B., 30 '& 31 Ed. I. p. 164). Again, in 1305, it is pointedly stated by the Court, the question being whether free or villein service is due from certain land, that a plea of the tenant's personal freedom is irrelevant as regards the land (Y. B., 33 Ed. I., p. 10). Probably it was not common for bondmen to be sold apart from the land, but there is no doubt that it was some- times done (Madox, Form. Anglic. Nos. 756-62). In the absence of evidence it would be a plausible theory that the ignorance or carelessness of Norman surveyors and judges confused the lowest classes of free men with the bondmen, in which process the dependent free man lost something, but the bondman who formerly had no rights at all ultimately gained much. Such a view has been taken by the Bishop of Oxford (Const. Hist., I. 429) and also by Mr. Freeman. But the evidence shows a shifting (if anything) rather than a confusion of the names, and possibly to some extent the actual state, of different .classes, a Standesverschiebung as one APPENDIX 213 might say in German. In the Exchequer Domesday we find on the manors of St. Paul's four distinct kinds of tenants, Villani, Bordarii, Cotarii, Servi. In the Domesday of St. Paul's, A.D. 1222, or in round numbers a century and a quarter later, we find Tenentes of several descriptions (lihere tenentes, tenentes per vilenagium, tenentes terras operarias, tenentes de terra assisa, and others), Cotarii, and Nativi. It may he difficult to assign the true relation of this classi- fication to that of the Conqueror's surveyors (see Scrutton, Gommans and Common Fields, pp. 25 sqq.); but anyhow there is the fact that in the thirteenth century the minuter distinc- tions which the Conquest might he supposed to efface were as far from having disappeared as ever, and went on under changed or shifted names. It by no means follows, of course, that all these terms imply differences of personal status. It is hardly possible (especially after Mr. Seebohm's work) to maintain that bordarius and cotarius signify any- thing but the general nature and size of the holding. But the leading types are well enough marked. In the eleventh century there was the free man in the fullest sense compatible with having a lord at all, who might "go where he would with his land," or, if he chose, alienate it, " vendere vel dare potuit," " potuit vendere cui voluit," in the phrases of Domes- day Book. Then there was the dependent free man, who might seek a new lord if he chose, but at the cost of leaving the land : " potuit ire quo volebat, sed non cum terra." Such are the coliberti of Domesday in Dorset, elsewhere called "sokemen" (Eyton, p. 46),^ and the thanes in the western counties who held church lands and " could not be separated from the church." Then we have the still more dependent holder {ascriptitius in the language of the twelfth- thirteenth century authorities, such as Dial. Scacc, II. 14), who cannot quit the land without the lord's consent. Last of all comes the thrall or servus. In the thirteenth century, 1 Eyton treats the villani, in Dorset at any rate, as an unfree class ; but I fail to see why. Doubtless they were inferior to the censores, who already paid a fixed money-rent for all service ; but payment of labour-rents is quite compatible with personal freedom, otherwise no free man could ever have held in villenage. 214 THE LAND LAWS I apprehend, these distinctions were by no means out of use, except that the power of seeking a new lord " cum terra sua" had disappeared, as being repugnant to feudal principles. The liber homo, ascriptitius, and servus, were still separate and defined conditions of men. The "ascriptitii qui villani dicuntur " of Dial. Scacc., I. x., seem to be not serfs, but degraded free men. Again, we know that villenage by blood, as distinct from villein or customary tenure, existed and was recognised not only in books but in practice down to the seventeenth century. "Queen Elizabeth in 1574 issued a commission for the enfranchisement of most of the bond- servants in the manors belonging to the Crown. And there is a case reported ' as late as the fifteenth year of James I., in which an issue as to villenage was tried " (Elton, Custom and Tenant-right, p. 29). In the face of this it is impossible to hold that the sections of Littleton on villenage, for example, are the mere survival of legal pedantry describing obsolete institutions as if they were still alive.^ Thus much of the persons. Now for the tenure. It should really be needless at this day to contradict Black- stone's story that all our customary tenures were invented after the Conquest by the more or less capricious indulgence of the lords of manors. Still it may be worth while to re- capitulate the facts. 1. There is no reason whatever to suppose that the actual tillers of the land were materially disturbed by the Conquest, except where, as in Yorkshire, the land was harried as a punishment for stubborn resistance.^ 2. Customary tenants are, in point of fact, found existing from Domesday till now without any discontinuous change in their tenure or its incidents : unless we are to count for ^ As to the frequent and early commutation of services for money payment, see the Glastonbury Inquest, pp. 37, 69,100 ; in the two latter passages it appears as already of long standing. Even the dis- tinction between fixed and arbitrary service as a test of the character of the holding is not always and absolutely to be relied on : see the Peterborough Book (Appendix to Ohron. Petrob. Camd. Soc, 1849, p. 160), where we find sokemen bound for certain days in summer time to do "quidquid iusserit dominus." ^ Cp. Domesday of St Paul's, Introd. p. xxx. APPENDIX 215 such the final allowance, temp. Ed. IV., of their right to maintain their title in the King's courts even as against their own lord. 3. The customs themselves hear every mark of archaic origin. Transfer by symholic ceremonies (which Blackstone mentions in a sufficiently appropriate context, Gomm. II. 313, but without seeing its real significance) ; a variety of rules of descent, showing affinities to early customs preserved in other parts of Europe ; and the record and attestation of title by the manorial coTirts, in forms apparently modelled on ancient popular procedure ; all these things point not to feudalism or to usages springing up under feudalism, but to a state of society far older. 4. The country where common fields abound, as we learn from the facts collected and discussed by Nasse, is also the country of small copyholds. We know that common fields represent a primitive system of agriculture. If copyholds equally represent a primitive system of tenure, it is natural that where the one persists the other should persist also. If, as the account formerly received would have it, copyholds are of post-Norman origin, the coincidence is inexplicable. When Sir Edward Coke said that copyholders come of an ancient house, he spoke more truth than he knew. The ancient customary holders of land under a lord who were not great men enough to hold their estates by charter before the Conquest, and have them confiscated after it, are repre- sented by the modern copyhold tenants, and in most cases we may well think that the customary tenant practically had a, fixed tenure from the first. There is a significant entry in the St. Paul's inquisition of 1222 (Domesday of St. Paul's, ed. Hale, p. 52) : " Terre akermannorum qiias dominus potest capere in manu sua cum vult sine iniuriis hereditarie successionis." The words as they stand might be clearer ; but, "iniuriis"' seems to stand for the more usual "calumpniis," i.e. claims, and the meaning to be that in these particular holdings a deceased tenant's heir had not any customary right to succeed. This implies that in other cases there was such a customary right, especially when we bear in mind the canon of inquiry that in medieval surveys what is un- 216 THE LAND LAWS usual, not what is usual, is specially noted outside the common forms of description. After all, there is no reason to suppose that the lawyers of Westminster in the thirteenth century were less dogmatic or more disposed to pay attention to the real facts and working of local usage than their followers in the sixteenth or nineteenth. Their confusion of villanus and servus may have been to some extent deliberate (Pollock and Maitland, Hist. Eng. Law, i. 413). I do not say that in every case copyhold is an ancient customary holding. It is possible that servile tenements of the most precarious kind may, in the course of the Middle Ages, have grown to some extent into customary estates. Nay, I think there is one class of copyholds which may with considerable probability be referred to such an origin, the copyholds of so-called imperfect tenure which are still fre- quent in the western counties (Elton, Custom and Tenant- right, p. 63). These are the copyholds for lives or years where the fines are still uncertain, and there is no strict right of renewal, — a kind of estate which, in Mr. Elton's words, "by the severity of the lord's exactions, and the re- currence of a servile phraseology, is shown to have descended from the precarious holdings of the natives who could call nothing their own." Long as it has taken to establish even an inchoate tenant-right, I should be disposed to put back the origin of these holdings far beyond the Norman Conquest. They prevail only in the west of England, if indeed they are found at all elsewhere. There is no reason why this should be so if they began as the holdings of English serfs or degraded free men. But there is every reason why it should be so if they began as the holdings of a conquered British population who remained as the serfs and tenants at will of the English invaders on the land which had formerly been their own. In the east and south-east the British tillers of the soil were slain or dispersed. The remnant who may have been reduced to servitude left no distinct traces on the new form of society. In the west the English conquest was milder and more gradual. Still it was a conquest by the strong hand, and the relation of mere APPENDIX 217 ■sufferance which Blackstone imagined between the Korman lord and the English customary tenant is like enough to have really existed between the victorious English settler and the Welshmen who were the captives of his spear. It seems a fair conjecture that the servile holdings of the West- Welsh, little or not at all affected by the Norman Conquest, have all these centuries been growing into what we now know as copyholds of imperfect tenure. In the case of the peculiar " convention ary " holdings of the Cornish mining country, where the tenant has an in- heritable interest, but must be readmitted every seven years, something like proof of a Celtic origin is attainable. This custom, fully described in Concanen's special report of Rowe V. Brenton, presents distinct analogies to the customs of Brittany.i Considering, again, the extreme antiquity of tin mining in Britain, it seems at least probable that the "tin -bounding " custom of Cornwall (and formerly of the contiguous parts of Devon also), and the similar customs which exist in the mining district of Derbyshire, come down to us from a time earlier than the English occupation ; and, indeed, when we remember tliat in Cornwall the English occupation is allowed to have been of the most superficial kind, even by those who are least favourable to the preserva- tion of Celtic elements elsewhere, there is hardly room for two opinions on this point. What is above suggested as to the origin of the imperfect copyholds of the west of England may perhaps be strengthened by consideration of the manner in which personal villenage was distributed over the country at the date of the Domesday survey. The Abstract of Population in the second volume of Ellis's Introduction gives the facts ready to hand ; and in Mr. Seebohm's English Village Gommunity they are exhibited with the help of graphical charts (p. 86). We find that the proportion of servi to the whol? population steadily diminishes as we go from the southern and western parts northward and eastward. In Cornwall and Gloucestershire, the former still really a West- Welsh county, the latter still half- Welsh, the 1 Coutumier Qiniral, iv. 409 (usances locales du domaine conge- able de Cornoaille). 218 THE LAND LAWS servi are almost one in four. In Devonshire tliey are little short of one -fifth, in Shropshire over one -sixth, in Dorset and Somerset not much under, and in Hampshire they are over one -seventh; in Worcestershire and Wiltshire it is about the same.^ As we cross from west to east there is a notable change ; the average in the southern midlands, such as Berks, Oxfordshire, Warwickshire, is roughly one-eighth. As we go on to Leicestershire, Nottinghamshire, Stafford- shire, Norfolk, and Suffolk, we find the serfs ten per cent of the population at most. In Derbyshire there are hardly any serfs, and in Lincolnshire, Huntingdonshire, Kiitland, and Yorkshire there are none at all. In the north-west, again, Cheshire shows a small proportion, less than one- twelfth. In the south-east the distribution seems irregular ; in Essex the serfs are about one-ninth, in Kent something less than one-tenth, in Surrey something more, while in Sussex they are less than one -twentieth. Local accidents of forest and fenland may in part account for these diversities ; the less profitable soil would be more in servile occupation, and serfs would be employed as herdsmen in the uncleared or half- cleared forests. In any case the average is much below that of the western counties. In one word, the serfs are fewer in proportion as the English settlement of the country is more complete. This decreasing ratio from west to east cannot be an accident, and we are led to infer that both the descent and the tenures of the nativi represented by the present " imperfect copyholders " of the west were not English but Welsh. Not that there is anything I know of to exclude the possibility that the Welsh themselves before the English invasion had serfs, who may not have been Celts or even Aryans, or that these people left a permanent mark. Mr. Elton {Origins of English History, ch. viii.) has collected a mass of facts tending to show that the preference of younger to elder sons, which in this country survives in the form of borough-English, is derived from non-Aryan sources. And ethnologists appear to be satisfied that the physical characters of a prehistoric race of short and dark men who were ^ In Herefordshire the proportion is evidently high, but the numbers seem to be uncertain. APPENDIX 219 certainly neither Teutons nor Celts are amply visible in modem England. But for the present purpose I keep within historic limits. How the king's courts came to assume, as they did at least as early as the beginning of the fourteenth century (Y. B., 30 & 31 Ed. I. 168), that there was no villenage in Kent is an unsolved riddle. The doctrine went so far that a man's freedom was established by showing that any of his ancestors was born in Kent, " quia esset impossible servum procreare liberum" (Y. B., 33 Ed. I. 15). But in Kent itself such was not the opinion or practice, for villenage certainly existed there much later, as Somner has shown (Treatise on Gavelkind, pp. V3-75); to say nothing of the difficvilty of accounting for the total disappearance, within two centuries or thereabouts, of the servile class registered in Domesday. The legal doctrine, which no less certainly was acted upon by the courts out of Kent, is a curious example of the rapid growth and acceptance of fictions even when the facts con- tradicting them are not remote or difficult of access. Note E. — Primogeniture in Socage Lands There is still much obscurity about the manner in which primogeniture was extended from lands of military tenure to socage lands. It is certain that when Glanvill wrote — that is, about a century after the Conquest — gavelkind was still the prevailing rule. In Glanvill's chapter on inheritance (L vii. c. 1) the rule of primogeniture, as a rule of common law, " secundum ius regni Angliae," is expressly confined to lands held by military service. As to the land of a " free soc-man," it has to be ascertained whether the land was partible by ancient custom. If so, the sons take equally, saving that the first-born has the chief dwelling-house, on the terms of making recompense in value to the others. If the land is not partible, then, " according to the custom of some, the first-born shall have the whole inheritance ; according to the custom of others, however, the last-born son is heir." Thus primogeniture and borough-English appear as local customs on an equal footing. The best specific evidence of 220 THE LAND LAWS customary primogeniture I know of is the case of certain inferior tenants of the Prior of Canterbury, who were called Kberi sohmanni, but whose J:enure had servile incidents (Elton, Tenures of Kent, 106). It has been suggested that in this case the rule was imposed after the Conquest on socagers who were driven by need or the strong hand to accept grants in base tenure (Kenny, Essay on Primogeniture, Cambridge, 1878, p. 31); but this, considering Glanvill's language, appears by no means a safe inference. There are also local rules of inheritance in various customary tenures, showing a preference of the eldest (or sometimes the youngest) daughter. (Elton, Custom and Tenant-right, Appendix D.) These evi- dently have nothing to do with feudalism. See further as to customary primogeniture or indivisible holding Mr. (after- wards Sir R.) Morier's "Report on Tenure of Land in the Grand Duchy of Hesse," Pari. Papers, 1870, Ixvii. 193, 197. In Bracton, writing after the lapse of nearly another century, we find a significant difference, though not at first sight a great one (fo. 76a). It is still a question in each case whether the heritage is by ancient usage partible or not. But if not, then primogeniture becomes the rule without further inquiry. Local custom is to be considered only in the case of villein tenure. Fleta (1. 5, c. 9, § 15) copies Bracton almost word for word. Bracton's language shows the tendency of primogeniture to increase, and the desire of lawyers to hasten the process. We cannot accept it, however, as show- ing how far matters had really gone ; and for this reason, that it proves too much. What Bracton says does not admit of the existence of borough-English in any land of freehold tenure. But we know that there must have been borough- English freeholds in Bracton's day ; for there were such when Littleton wrote, and there are still such in our own day. Therefore, what we really learn from Bracton is that the lawyers of the thirteenth century, in their zeal to improve the position of primogeniture, were ready to outrun the actual law and practice as they existed then and have existed ever since. Again, Bracton does not say or suggest that there was any presumption for or against the partibility of socage lands. APPENDIX 221 Mr. Kenny {Essay on Primogeniture, p. 20) has called atten- tion to a case, decided in the year 1200, as showing that by that time the burden of proof was on those who claimed under the custom of division. The record is short enough to quote in full (Placita de term. S. Mich., 2 Joh., Abbrev. Placit. p. 286 ; since published from the original roll by the Selden Society, Select Oivil Pleas, vol. i. ed. W. P. Baildon, 1890, pi. 61):— " Rutland. Gilebertus de Beivill [petit] versus Willelmum de Beivill duas virgatas terre cum pertinentiis in Gunetorp que eum [sic] contingunt de socagio quod fuit patris eorum in villa, et Willelmus venit et defendit ius eius et quod socagium illud nunquarn partitum fuit nee debet partiri et hoc offert defendere per quendam liberum hominem suuni. [Gilbert offers two marks to the king to have an inquest instead of trial by battle on the issue] utrum terra ilia partiri solet et sit partibilis. Et ipse voluit ponere se in iuratam inde et quia Gilebertus nullam probam produxit consideratum est quod Willelmus eat sine die et quietus." This, however, does not show that there was a presump- tion in favour of primogeniture ; it only shows that there was no presumption to the contrary. The holding might or might not be divisible by ancient custom ; whether it was so was a question of fact. It was the plaintiff's business, as in any other case, to prove the facts on which he relied, and among them that the land had from ancient times been divis- ible. Failing to prove this, he lost his cause. It does not appear that if the elder son had been plaintiff he would not in like manner have been called on to prove that primogeni- ture was the local rule. It is certain, however, that during the thirteenth century primogeniture made great advances. Before the century was out it had come to be spoken of as the common law of England. In 1292 we find it said that "in some places as well the tenements holden in socage as other tenements are governed by the common law" ; and the custom of Kent is already spoken of as peculiar (Y. B., 20 Ed. I. in Record Office Series, pp. 327, 329): but then, and some years later, it was still an open question of fact whether given socage lands 222 THE LAND LAWS were partible or not (30 Ed. I. p. 50, a.d. 1302). There came a time, no doubt, and probably not long after this, when by making the rules of evidence more stringent the partible tenure, except in Kent, was extinguished. But specific evi- dence of the steps by which the process was completed appears to me to be still wanting. Note F. — " Cestui que use " at Common Law The action of assumpsit was about contemporaneous with the doctrine of uses in its introduction and development : it became at last the common method of enforcing simple con- tracts, but originally was not an action on the promise itself, but on the breach of a duty incident to the promise, a " tort founded in contract," as the modern phrase runs. The object, in fact, was to get a remedy where the contract itself could not be sued on in any of the recognised forms of action. I do not know what there was to prevent the judges, if they had chosen, from holding that a conveyance to uses created a duty in the feoffees which would support an action of assump- sit in case of breach or neglect. Again, they might well enough have given cestui que we an action of account for the profits, or an action on the case analogous to it. Neither of these remedies, however, would have been anything like so effectual as the Chancellor's decrees. Bacon's summary of the mischiefs aimed at by the statute is worth citing. {Works, ed. Spedding, vii. 418.) "The par- ticular inconveniences by the law rehearsed may be reduced to four heads : first, that these conveyances in use are weak for consideration " {i.e. made without fitting deliberation) ; " secondly, that they are obscure and doubtful for trial ; thirdly, that they are dangerous for want of notice and pub- lication ; fourthly, that they are exempted from aU such titles as the law subjecteth possession unto." Notwithstanding the dislike of the common lawyers to uses, the sages of the law did not scruple to employ them for their own convenience. Littleton had lands in use, and in 1481 disposed of the use by his will (see it set out in the APPENDIX 223 notes to his life in Tomlins's edition of his Tenures, 1841, p. xxxiv.). Note G. — Settlements and Perpetuity Estates for life, followed by one or more estates tail in remainder, were a known form of disposition as early as the fourteenth century. In the Calendar of Patent EoLs for 7 Ed. II. (p. 76, ed. 1802) this grant occurs :— " Rex contirmavit Ricardo de Pynelesdon pro vita, rem' diversis filiis suis in generali tallio omnia sua hereditamenta in Worthynbury in partibus de Maylorseysenells per servic' quartae partis unius feodi milit'." On the next page is a grant of successive estates tail with- out a preceding life estate : — " Rex confirmavit Johanni de EUerker in generali tallio rem' aliis in generali tallio cert' haeredit' in Ellerker ei dat' per Episcopum Dunolm' \sic] pro annuo reddit' 3s. 9d." In the later time of Edward I. there appear several grants in tail, often to a man and his wife, and sometimes there is a remainder in fee-simple (ib. p. 51, coL 2 at foot) ; but for the most part the grants abstracted are in fee-simple. The history of the further development of settlements was traced by the late Mr. Joshua Williams (Transactions of the Juridical Society, i. 45). He states, as the result of a care- ful examination of documents preserved in the MS. collections of the British Museum, that limitations in substantially the modern form to the use of a man for life, remainder to his first and other sons in tail, were not uncommon in the latter part of the sixteenth century ; the parties being content apparently to take the risk of the contingent remainders being destroyed by the feoffment (or other act amounting to a discontinuance), or forfeiture by waste, etc., of the tenant for life. The introduction of trustees to preserve contingent remainders in order to meet this risk is ascribed by a pro- fessional tradition, which was established by the middle of the last century (Blackstone, Comm. II. 172 ; cp. Lord Hard- wick in Garth v. Cotton, 1 Dick. 188, 191, and the argu- 224 THE LAND LAWS ments of the same case, 1 Ves. Sr. 524, 548), to Orlando Bridgman and Geffrey Palmer, whose invention, it is sug- gested, was sharpened by the troubles of the Civil Wars to the better securing of their clients' estates. In Orlando Bridgman's Precedents of Conveyances the limitation to a trustee, after the determination of the tenant for life's estate, and during his natural life, "upon trust only for preserving the contingent uses and estates herein- after limited, and to make entries for the same if it shall be needful," appears in regular use (though sometimes it is omitted), and in that form it does not appear to occur earlier. Limitations curiously similar in form, though differing from this in substance, as careful examination will show, were, however, in use nearly a century before. In Holcroft's case, Moore, 486, we find an estate settled by fine in the second year of Mary (1554-55) to the use of Sir John Holcroft the younger for life " so long and until he attempt to alien, and then to the use of Hamlet [Holcroft], and the heirs males of his body during the life of Sir John the younger ; and im- mediately after his death " remainders to the first and other sons of Sir John the younger in tail male, down to the fourth son inclusive, remainder to Hamlet Holcroft in tail male, re- mainders over. The estate limited to Hamlet during Sir John's life comes exactly in the place where, in Bridgman's and later settlements, the estate of the trustees to preserve contingent remainders would come. But, for whatever pur- pose it was inserted, it cannot well have been to preserve the remainders to Sir John's unborn sons ; for supposing that no son was in existence when Hamlet entered upon Sir John's attempt to alien (and assuming, which is not clear, that the " attempt to alien " refers to a tortious conveyance of the fee, and not merely to alienation of the life estate), then Hamlet's estate tail pur aider vie and his remoter estate tail in re- mainder would coalesce and shut out the intermediate con- tingent remainders. It seems to have been intended merely as a check on the tenant for life. Only one step, however, was needed to make an estate of this kind a protection for the remainders as well, and that step was taken by Bridgman and Palmer when they limited it to a trustee specially appointed APPENDIX 225 for the purpose, and with an express declaration of the trust. ' Modern conveyancing as a wliole, and the regular use of settlements in their modern form, certainly date from about Bridgman's time. A brief but clear and admirably written account of the elements of the subject, both doctrinal and historical, is given in the part of Blackstone's Commentaries, already referred to (Book II. ch. ii.) ; it ought to be needless to mention this, but the text of Blackstone is nowadays un- duly neglected, partly by reason of exaggerated criticisms which have obtained currency, partly from the ravages it has suffered at the hands of so-called editors. Advanced students will of course con,sult the late Mr. Davidson's exhaustive in- troduction to his volume on Settlements. As to the history of the rule against perpetuities, the following dates may be not without curiosity for readers of this note : — 1681-85. Duke of Norfolk's case (3 Ca. Ch.) Limita- tions to take effect within lives in being held good, and an attempted distinction between terms of years and freehold interests rejected. 1697. Lloyd v. Carew (Pre. Ch. 72, 106, and Shower, Ca. Pari. 137). Twelve months after lives in being, good. 1699. Scattergood v. Edge, 12 Mod. 287. Still doubt- ful whether a life or lives in being were not the extreme limit : " It was a great policy of the common law that alienation should be encouraged; for it is the greatest preserver and promoter of industry, trade, arms, and study ; and this was visible from the making of the statute Ve Bonis, until common recoveries were found out ; and these executory devises had not been long countenanced when the judges repented them, and if it were to be done again, it would never prevail ; and therefore there are bounds set to them, viz. a life or lives in being ; and further they shall never go by my consent, let Chancery do as they please " : per Treby, C. J. 1736. Stephens v. Stephens. (Ca. temp. Talbot, 228.) Executory devise to an unborn child of a living person when he shaU attain the age of twenty-one years held good. This was the conclusive decision. Q 226 THE LAND LAWS 1765. Blackstone. (Comm. IL 174.) "The utmost length that has been hitherto allowed for the contingency of an executory devise of either kind to happen in, is that of a. life or lives in being, and one-and-twenty years afterwards.'' 1787. Jee v. Audley, 1 Cox, 324, 1 Rev. Rep. 46. (Sir Lloyd Kenyon, M. R.) The rule considered to be settled : " it is grown reverend by age, and is not now to be broken in upon." 1833. Cadell v. Palmer, in House of Lords, 1 CI. & F. 372. The term not exceeding twenty-one years which may be added to lives in being is a term in gross, i.e. it need not have reference to any minority. A very full history of the doctrine is contained in Hargrave's argument in Thellusson v. Woodford, 4 Ves. 247 sqq. (also separately printed in his Juridical Arguments, vol. ii.), and Sir E. Sugden's in Cadell v. Palmer. The allowance of the period of gestation when gestation exists (but not as a term in gross of nine or ten months) is properly treated by the best writers on the subject as being not a special clause or pait of this rule, but an application of the general principle that an unborn child is _ treated, when- ever it will be for his benefit, as being actually in existence ; and I have so stated it in the text. It is now judicially settled, after much difference of opinion, that remainders are also subject to another rule which is not merely a particular example of ' the rule against perpetuities, namely, that an estate cannot be limited to any child or issue (as purchasers) of a person not in existence : Whitby v. Mitchell (1890) 44 Ch. Div. 85. Much might be said about the history of the doctrine, but this is hardly the place for it. The learned reader may consult Williams on Seal Property, 17th ed. p. 380, and the preface to voL viii. of the Revised Reports. A question in some measure connected with the doctrine of perpetuity is how far there can exist at common law a fee-simple determinable otherwise than by failure of heirs, or by the re-entry of the grantor or his heirs for a condition broken. To put the question in a concrete instance : apart from the Statute of Uses, and apart from any question of perpetuity, is a grant good in this form ; To J. S. and his APPENDIX 227 ieirs as long as Lincoln's Inn Hall, or as a certain tree, shall stand ? Plowden thought such a grant was good, and operated hy way of limitation, not condition (which distinction must be carefully observed) : Plowden, 557, cp. Scholastioa's case, ih. 414. And Coke in Seymor's case (10 Rep. 976) says there is a fee -simple determinable by limitation, as distinguished from one determinable by condition ; " as if A enfeolfs B of the manor of D to have and to hold to him and his heirs, so long as C has heirs of his body, and that is called a fee-simple limited and qualified." No remainder or reversion can be expectant on it. There is also a fee-simple determinable "implicit, and derived out of an estate tail," which is known as a base fee ; its nature is comparatively common learning. In Liford's case (11 Eep. 49a) we read that " a man may have an inheritance in fee-simple in lands, as long as such a tree shall grow, because a man may have an inheritance in the tree itself." But this appears to be said only in the course of argument. The Year-Book referred to (27 Hen. VIII. 296) gives the fanciful reason that an estate may as well depend on the life of a tree as of a man. This also appears to be extra-judicial. Neither the reason in Coke's report nor that in the Year-Book can be said to strengthen the proposition. On the other hand, Coke's contemporary Anderson, Chief Justice of the Common Pleas, appears to have been of the contrary opinion ; according to what is given in his reports (2. 138, 139) as the resolution of the Court in Corbet's case, the only determinable fee-simple (not upon condition) known to the law is a base fee, as, when a villein is tenant in tail, and his lord enters on him, or tenant in tail is attainted of high treason, the lord or the Crown has a fee-simple deter- minable on the failure of issue inheritable under the gift in tail : he seems to think parties cannot make a gift in fee- simple determinable in any other way than by entry for condition broken, which is a different matter. The cases and opinions in the Year-Books referred to in Anderson are partly unverifiable by reason of wrong or misprinted refer- ences, partly not to the point, being concerned with condi- tions against alienation by a tenant in tail, and in one place 228 THE LAND LAWS (13 H. VIL 24) there is aa unresolved difference of opinion. If there be a feoffment in fee for " so long as J. S. has issue," it is asserted on the one part and denied on the other that the "condition" (but a condition in the proper sense can hardly be meant) is void ; the reporter adds " ideo quaere." Modern writers are not agreed either. Some, and notably Mr. Challis, following Preston, are confident that such estates can exist, some that they cannot. On the whole, it appears that the point has never been settled. I cannot help thinking that the silence of Coke's commentary on Littleton (where surely he would have ex- plained the nature of such estates, and distinguished them from estates upon condition, had he deliberately allowed them : see especially 2 1 46) weighs much against the dicta in his reports. Also it is strange that Littleton himself (ss. 380, 381, etc.) had nothing to say of such estates, if in his time they were known to the law. It may be that after the statute De Bonis an estate to A and his heirs as long as B shall have heirs of his body (as put in Seymor's case) was exceptionally allowed, or thought allowable, by analogy to a base fee. If a fee-simple determinable on an arbitrary event is a possible estate at common law, it seems to be not within the rule against perpetuities. It also seems probable that since the statute of Quia emptores the possibility of reverter would belong not to the grantor or his heirs, but to the chief lord of the fee. Whether such possibility (if any such there can be) is a " possibility coupled with an interest " within 8 & 9 Vict t. 106, s. 6, so that it may now be disposed of by deed, is a point that may be considered at leisure by the curious. The learned reader is further referred to the remarks of Prof. J. C. Gray, of Harvard University, and Mr. H. W. Challis on Determinable Fees in the Law Quarterly Review, iii. 399, 403. Note H The law relating to the settlement and transfer of land, and various proposals for its amendment, are considered in the following recent publications : — APPENDIX 229 Land Transfer. Published by order of the Bar Committee. London : Butterworths, 1886. Cp. Law Quarterly Review, ii. 397. Statement on the Land Laws. By the Council of the Incorporated Law Society of the United Kingdom. London : Butterworths, 1886. Cp. the articles cited below. On the Transfer of Land. By H. W. Elphinstone. In Law Quarterly Review, 1886, ii. 12. At p. 23V of the same volume the same writer gives a concise critical account of a number of tracts and articles on the subject, of which the titles are there collected. The Council of the Incorporated Law Society on the Land Laws. Articles in the Solicitors' Jowrnal, 1886, January 30 to March 6 inclusive. Registration of title to land, and how to establish it without cost or compulsion. By Charles Fortescue-Brickdale. London : E. Stanford, 1886. Cp. Law Quarterly Review, iii. 86. The Land Transfer Bill. Solicitor's Journal, April 9, 16, 1887. The second article gives a list of publications, in- cluding documents on the Torrens system in the Australian Colonies. Transfer of Land by Registration of Title. By T. R. Colquhoun Dill. London : Cassell and Co., 1893. Notes on Land Transfer in various Countries. By Charles Fortescue-Brickdale. London: Law Times O&ce, 1894. INDEX The figures to which an asterisk is added refer to places whose techni- cal terms are explained : thus the entry " Entail, 68*," shows that the legal definition of entail will be found at page 68. Abstract of title, 171* Agricultural Holdings Act, 149, 151, 152 Aids, 62* Alienation of land, 66 Allotments Act, 162 Alodium, 27* Approvement, 183*, 188 Aryan household, 118 Assize of arms, 108 Assizes, 77 Bacon on perpetuities, 120 on uses, 96 Bargain and sale, 105 Base fee, 113*, 227, 228 Betterment, 161 Bondmen, 45-48, 61 Book-land, 20*, 23, 24, 26, 28, 35, 199, 200 Borough-English, 49*, 219 Building leases, 156 Cestui que use, 94*, 99 Church, lands held by, 89 Common lands, 5, 18, 22, 38, 39, 40, 115, 183, 184, 222 Common, rights of, 18, 41-43 their revived importance, 182 Communal property, 21 Compensation for improvements, 152 sqq. " Conveutionary " tenants, 51 Conveyances, secret, 104 Conveyancing Act, 166, 167 Copyhold Act (1894), 181 Copyholds, 19, 39, 43, 44, 48-51, 60 enfranchisement of, 181 origin of, 214 Crown lands, 37 Crown, sovereignty of the, 65 Customary rights, 41 Customs, local, as between land- lord and tenant, 51, 150 Dartmoor, 14, 42 Be Bonis (statute), 22, 68, 80, 88, 89, 91, 123 Deeds, 75 shortening of, 167 Devise, 129* Distress for rent, 145 sqq. Domesday, 35, 37 Dower, 179* Basements, 177* Enfranchisement of copyholds, 181 of leaseholds, 160 Enrolments, statute of, 105, 106 232 INDEX EntaU, 68*, 80, 87, 112, 123, 124 Epping Forest case, 186 Equitable estate, 115 Equity of redemption, 134* Esclieat, 65* Pee simple, 66* tail, 68 Peoifment, 76* Peudalism, 8, 30, 54 sqq., 80 Pine, 87 Pines and recoveries, abolition of, 168 Fixtures, law of, 150 Folk-land, 20*, 24, 37 Foreshore, rights of public on, 13 Forest (royal), 41* Prankalmoigne, 36*, 90, 131 Gavelkind, 58 Grant, conveyance of freehold in possession by, 166 Guardian, 63 Heielahd, 22 Heriots, 44 Highway, rights of public on, 13 Impeovemefts by tenant, com- pensation for, 151 sqq. Inclosure, 38 Inclosure Acts, 183, 184 Income tax, 132 Inheritance, modern law of, 178 Jews, civil rights of, in Middle Ages, 17 n. Jurisdiction, private, 27 L^N-LAND, 28* Lammas lands, 6, 38* Land, ownership of, 12 Land-tax, 62, 132 Land Transfer Act, 177 Lease and release, conveyance by, 106, 107 Leaseholding, 141 sqq. Leasehold tenure, peculiarities of, 142 Leaseholds, 7 proposal for compulsory en- franchisement of, 60 urban, 154 Leases for lives, 141 for years, 72 Limitation Acts, 176 Limitations, in settlements, 112 Limited owner, 9 Livery of seisin, 75*, 104 Lord and man, 29 Magna Charta, 71, 77, 91 Manor, 10, 34, 41-44, 200 sqq. Mark, 202* Merchet, 209* Merton, Statute of, 183, 188 Military tenures, abolition of, 130 Mines, leases of, 155 Mining customs, 51 Mortgage, 132 sqq. equitable, 136 Mortmain, 91, 92 " Nationalisation " of land, 195 Oath of allegiance, 65 Open spaces, preservation of, 184 Paeish lands, 39 Perpetuities, 111, 225 Bacon on, 120 the rule against, 122, 128 Personal laws, 17 Possession, 73 ■ symbolic delivery of, 74 Power of sale, statutory, in mort- gages, 135 Powers in settlements, 116 Prescription Act, 176 " Primer seisin, "63* Primogeniture, 57 relation of legal to customary, 179, 219 Public, rights of, on open lands, 14 Quia emptores (Statute of), 70 Raok-eent, 157 n. INDEX 233 " Real actions," 76 Real Property Commissioners, Report of, 169, 172, 176 Real Property Law Amendment Act, 166 Recovery, common, 83 Redemption, equity of, 134* Registry of titles or assurances, 172, 175 Relief, 63 Remainder, 80*, 112, 123 Remoteness, dispositions void for, 126, 128 Rents, conversion from labour into money, 140 in kind, 8 Reversion, 128 St. Leonards, Lord, on settle- ments, 190 Scotland, leasing customs in, 159 old law of entail in, 169 registration of assurances in, 177 Seals, use of, 75 Serfs, 46, 61 Serjeanty, grand, 131 Settled Land Act, 192-194 Settlements, 9 Lord St. Leonards on, 190 strict, nature of. 111, 117 . strict, disadvantages of, 191 opinion of Real Property Commissioners on, 169 Small Holdings Act, 162 Socage land, 28*, 59 military tenures converted into, 130 Subinfeudation, 70 [ Succession, intestate, to land, 178 [ Symbols in land transfer, 74 Tenube, 12 conventionary, 51 customary, 48 military, 33, 53 military, abolition of, 130 servile, 47, 61 Tin-bounding, 61, 217 Titles, investigation of, 171 registration of, 172 Town holdings, report of Com- mittee on, 161 Transfer of land, publications on, 229 Trespass, actions of, 15 Trusts, 104 (and see Uses) Uses, 93 Bacon on, 96 shifting, 125 Statute of, 2, 93, 99-109, 115, 122, 129, 137, 171 Village greens, 40 Villein, 45* Villenage, 45, 48 Vouchee, common, 86 Wardship, 63 Warranty, 82* Wills Act, 170 Wills of land, 95, 100, 129 Wimbledon Common, 14, 185 Writ of right, 77, 88 Yearly tenancy, 142 THE END Printed by R. & R. Clark, Limited, Ediniurgh. THE ENGLISH CITIZEN: HIS RIGHTS AND RESPONSIBILITIES. Edited by Henry Craik, C.B. Crown &V0. 2s. 6d. each. THE PUNISHMENT AND PREVENTION OF CRIME. By Colonel Sir Edmund du Cane. LOCAL GOVERNMENT. By M. D. ^Chalmers. COLONIES AND DEPENDENCIES. Part I. India. By J. S. Cotton, M.A. II. The Colonies. By E. J. Payne. THE STATE IN ITS RELATION TO EDUCATION. By Henry Craik, C.B. THE STATE AND THE CHURCH. By Hon. Arthur Elliot, M.P. THE STATE IN ITS RELATION TO TRADE. By Lord Farrer. THE POOR LAW. By the Eev. T. W. Fowle. THE STATE IN RELATION TO LABOUR. By W. Stanley Jevons. JUSTICE AND POLICE. By F. W. Maitland. THE NATIONAL DEFENCES. By Colonel Maurice, B.A. [in the Press. THE LAND LAWS. By Sir F. Pollock, Bart. Second Edition. CENTRAL GOVERNMENT. By H. D. Traill. THE ELECTORATE AND THE LEGISLATURE. By Spencer Walpole. FOREIGN RELATTIONS. By Spencer Walpole. MACMILLAN AND CO., LONDON. THE IRISH LAND LAWS. By Albxandee G. Richet, Q.C, LL.D., Deputy Eegius Professor of Feudal and English Law in the Uni- versity of Dublin. Crown 8vo. 3s. 6d. A HISTORY OF THE CRIMINAL LAW OF ENGLAND. By Sir James Fitzjames Stephen, Bart., K.C.S.I., D.C.L., formerly a Judge of the High Court of Justice, Queen's Bench Division. In three Vols, 8vo. 48s. A GENERAL VIEW OF THE CRIMINAL LAW OF ENGLAND. By Sir J. F. Stephen. Second Edition. 8vo. 14s. A DIGEST OF THE LAW OF EVIDENCE. By Sir J. F. Stephen. Sixth Edition. Grown 8vo. 6s. A DIGEST OF THE CRIMINAL LAW. (Crimes and Punish- ments.) By Sir J. F. Stephen. Fifth Edition by Sir Herbert Stephen, Bart., and Harry L. Stephen. 8vo. 16s. A DIGEST OF THE LAW OF CRIMINAL PROCEDURE IN INDICTABLE OFFENCES. By Sir J. F. Stephen and Sir Herbert Stephen, Bart. 8vo. 12s. 6d. INTERNATIONAL LAW AND INTERNATIONAL RELA- TIONS. An attempt to ascertain the best method of discussing tlie topics of international law. By J. K. Stephen, B.A. of the Inner Temple, Barrister-at-Law. Crown 8vo. 6s. ESSAYS IN JURISPRUDENCE AND ETHICS. By Sir F. Pollock, Bart. 8vo. 10s. 6d. AN INTRODUCTION TO THE HISTORY OF THE SCIENCE OF POLITICS. By Sir P. Pollock, Bart. Crown 8vo. 2s. 6d. _ LEADING CASES DONE INTO ENGLISH, and other Diversions. By Sir F. Pollock, Bart. Crown Svo. 8s. 6d. INTRODUCTION TO THE STUDY OF THE LAW OF THE CONSTITUTION. By A. V. DiCEV, B.C.L. of the Inner Temple, Barrister- at-Law ; Vinerian Professor of English Law ; Fellow of All Souls, O.'cford ; Hon. LL.D. Glasgow. Fourth Edition. Svo. 12s. 6d. THE ENGLISH CONSTITUTION. By Emile Boutmt. Trans- lated from the French by Isabel M. Eaden. With Introduction by Sir Frederick Pollock, Bart. Crown Svo. 6s. STUDIES IN CONSTITUTIONAL LAW— FRANCE, ENG- LAND, UNITED STATES. By Emile Boutmt. Translated from the Second French Edition by E. M. Dicey. With an Introduction by A. V. Dicey, B.C.L., Vinerian Professor of English Law, Oxford. Crown Svo. 6s. COMMERCIAL LAW. An Elementary Text-book for Commercial Classes. By J. E. C. Munro, LL.Jt. of the Middle Temple, Barrister-at- Law. Formerly Professor of Law in the Owens College, Manchester. Globe Svo. 3s. fid. [ComTiurcud G'iass Books. MACMILLAN AND CO., LONDON,