THE COMMUNE OF LONDON

THE
COMMUNE OF LONDON

AND OTHER STUDIES

BY J. H. ROUND M.A.

AUTHOR OF ‘GEOFFREY DE MANDEVILLE’,
‘FEUDAL ENGLAND,’ ETC.

With a Prefatory Letter by Sir Walter Besant

WESTMINSTER

ARCHIBALD CONSTABLE AND CO.

2 WHITEHALL GARDENS

1899

Butler & Tanner,
The Selwood Printing Works,
Frome, and London.

Prefatory Letter

DEAR MR. ROUND,

I have to thank you for kindly letting me see the advance proofs of your new book. It is difficult for me to explain the very great advantage which the study of your books has been to me in my endeavour to get at the facts, especially those of the 12th century, connected with the history of London. For instance, I have found in your pages for the first time a working theory of the very difficult questions connected with the creation of the municipality. I have adopted your conclusions to the best of my ability with, I hope, an adequate expression of thanks to the source from which they are derived.

I would also point out the great service which you have rendered to the history of the City by giving, for the first time, the exact truth regarding the conveyance of the Portsoken to the Priory of the Holy Trinity, an event which has been hitherto totally misunderstood.

Thirdly, I must acknowledge that it is only from your pages, especially a certain appendix to ‘Geoffrey de Mandeville,’ that one can understand the ordinary position of the clergy of the City of London in the 12th century.

It is unnecessary for me to enumerate many other obligations which I owe to your pages.

I remain, dear Mr. Round,

Very faithfully yours,

WALTER BESANT.

Office of the Survey of London,

July 6th, 1899.

Contents

Preface

The paper which gives its title to this volume of unpublished studies deals with a subject of great interest, the origin of the City Corporation. In my previous work, ‘Geoffrey de Mandeville’ (1892), and especially in the Appendix it contains on ‘The early administration of London,’ I endeavoured to advance our knowledge of the government and the liberties of the City in the 12th century. In the present volume the paper entitled “London under Stephen” pursues the enquiry further. I have there argued that the “English Cnihtengild” was not the governing body, and have shown that it did not, as is alleged, embrace a religious life by entering Holy Trinity Priory en masse. The great office of “Justiciar of London,” created, as I previously held, by the charter of Henry I., is now proved, in this paper, to have been held by successive citizens in the days of Stephen.

The communal movement, which, even under Stephen, seems to have influenced the City, attained its triumph under Richard I.; and the most important discovery, perhaps, in these pages is that of the oath sworn to the Commune of London. From it we learn that the governing body consisted at the time of a Mayor and “Échevins,” as in a continental city, and that the older officers, the Aldermen of the Wards, had not been amalgamated, as has been supposed, with the new and foreign system. The latter, I have urged, is now represented by the Mayor and Common Council. That this communal organization was almost certainly derived from Normandy, and probably from Rouen, will, I think, be generally admitted in the light of the evidence here adduced. This conclusion has led me to discuss the date of the “Établissements de Rouen,” a problem that has received much attention from that eminent scholar, M. Giry. I have also dwelt on the financial side of London’s communal revolution, and shown that it involved the sharp reduction of the ‘firma’ paid by the City to the Crown, the amount of which was a grievance with the citizens and a standing subject of dispute.

The strand connecting the other studies contained in this volume is the critical treatment of historical evidence, especially of records and kindred documents. It is possible that some of the discoveries resulting from this treatment may not only illustrate the importance of absolute exactitude in statement, but may also encourage that searching and independent study of ‘sources’ which affords so valuable an historical training, and is at times the means of obtaining light on hitherto perplexing problems.

The opening paper (originally read before the Society of Antiquaries) is a plea for the more scientific study of the great field for exploration presented by our English place-names. Certain current beliefs on the settlement of the English invaders are based, it is here urged, on nothing but the rash conclusions of Kemble, writing, as he did, under the influence of a now abandoned theory. In the paper which follows, the value of charters, for the Norman period, is illustrated, some points of ‘diplomatic’ investigated, and the danger of inexactitude revealed.

Finance, the key to much of our early institutional history, is dealt with in a paper on “The origin of the Exchequer,” a problem of long standing. On the one hand, allowance is here made for the personal equation of the author of the famous ‘Dialogus de Scaccario,’ and some of his statements critically examined, with the result of showing that he exaggerates the changes introduced under Henry I., by the founder of his own house, and that certain alleged innovations were, in truth, older than the Conquest. On the other, it is shown that his treatise does, when carefully studied, reveal the existence of a Treasury audit, which has hitherto escaped notice. Further, the office of Chamberlain of the Exchequer is traced back as a feudal serjeanty to the days of the Conqueror himself, and its connection with the tenure of Porchester Castle established, probably, for the first time. The geographical position of Porchester should, in this connection, be observed.

In two papers I deal with Ireland and its Anglo-Norman conquest. The principal object in the first of these is to show the true character of that alleged golden age which the coming of the invaders destroyed. It is possible, however, of course, that a “vast human shambles” may be, in the eyes of some, an ideal condition for a country. Mr. Dillon, at least, has consistently described the Soudan, before our conquest, as “a comparatively peaceful country.”[1] In the second of these papers I advance a new solution of the problem raised by the alleged grant of Ireland, by the Pope, to Henry II. As to this fiercely contested point, I suggest that, on the English side, there was a conspiracy to base the title of our kings to Ireland on a Papal donation of the sovereignty of the island, itself avowedly based on the (forged) “donation of Constantine.” No such act of the Popes can, in my opinion, be proved. Even the “Bull Laudabiliter,” which, in the form we have it, is of no authority, does not go so far as this, while its confirmation by Alexander III. is nothing but a clumsy forgery. The only document sent to Ireland, to support his rights, by Henry II. was, I here contend, the letter of Alexander III. (20th September, 1172), approving of what had been done. That he sent there the alleged bull of Adrian, and that he did so in 1175, are both, I urge, although accepted, facts without foundation.[2]

The method adopted in this paper of testing the date hitherto adopted, and disproving it by the sequence of events, is one which I have also employed in “The Struggle of John and Longchamp (1191).” The interest of this latter paper consists in its bearing on the whole question of historic probability, and on the problem of harmonising narratives by four different witnesses, as discussed by Dr. Abbott in his work on St. Thomas of Canterbury. This is, perhaps, the only instance in which I have found the historic judgment and the marvellous insight of the Bishop of Oxford, if I may venture to say so, at fault; and it illustrates the importance of minute attention to the actual dates of events.

Another point that I have tried to illustrate is the tendency to erect a theory on a single initial error. In “The Marshalship of England” I have shown that the belief in the existence of two distinct Marshalseas converging on a single house rests only on a careless slip in a coronation claim (1377). A marginal note scribbled by Carew, under a misapprehension, in the days of Elizabeth, is shown (p. 149) to be the source of Professor Brewer’s theory on certain Irish MSS. Again, the accepted story of the Cnihtengild rests only on a misunderstanding of a mediæval phrase (p. 104). Stranger still, the careless reading of a marginal note found in the works of Matthew Paris has led astray the learned editors of several volumes in the Rolls Series, and has even been made, as I have shown in “the Coronation of Richard I.,” the basis of a theory that a record of that event formerly existed, though now wanting, in the Red Book of the Exchequer.

The increasing interest in our public records—due in part to the greater use of record evidence in historical research, and in part, also, to the energy with which, under the present Deputy-Keeper of the Records, their contents are being made available—leads me to speak of the contributions, in these pages, to their study.

A suggestion will be found (p. 88) as to the origin of the valuable “Cartæ Antiquæ,” of which the text too often is corrupt, but which, it may be hoped, will soon be published, as they are at present difficult to consult. In the paper on “The Inquest of Sheriffs” I have proved beyond question the identity of the lost returns discovered at the Public Record Office, and so lamentably misunderstood by their official editor. But the most important, and indeed revolutionary, theory I have here ventured to advance deals with what are known as the Red Book Inquisitions of 12 and 13 John. It is my contention that this Inquest, the existence of which has not been doubted,[3] though it rests only on the heading in the Red Book of the Exchequer, never took place at all, and that these ‘Inquisitions’ are merely abstracts, made for a special purpose, from the original returns to that great Inquest of service (as I here term it) which took place in June, 1212 (14 John). It is singular that this conclusion is precisely parallel with that which experts have now adopted on another great Inquest. “Kirkby’s Quest,” it is now admitted, having been similarly misdated in an official transcript, and again, in our own time, by an officer of the Public Record Office, was similarly shown by a private individual to consist, as a rule, “of abridgments only of original inquisitions” ... “extracts from the original inquisitions made for a special purpose.”[4] Thus, under John, as under Edward I., “the enquiry itself was a much wider one” than would be inferred from the Red Book Inquisitions and “Kirkby’s Quest” respectively. And, in both cases, its date was different from that which has been hitherto assigned.

I cannot doubt that the theory I advance will be accepted, in course of time, by the authorities of the Public Record Office. In the meanwhile, I have endeavoured to identify all the material in the ‘Testa de Nevill’ derived from the returns to this Inquest, and thus to make it available for students of local and family history.

It is needful that I should say something on the Red Book of the Exchequer. One of the most famous volumes among our public records, it has lately been edited for the Rolls Series by Mr. Hubert Hall, F.S.A., of the Public Record Office.[5] The inclusion of a work in the Rolls Series thrusts it, of necessity, upon every student of English mediæval history. It also involves an official cachet, which gives it an authority, as a work of reference, that the public, naturally, does not assign to the book of a private individual. That a certain percentage of mistakes must occur in works of this kind is, perhaps, to be expected; but when they are made the vehicle of confused and wild guesswork, and become the means of imparting wanton heresy and error, it is the duty of a scholar who can prove the fact to warn the student against their contents.[6] It is only, the reader must remember, a stern sense of duty that is likely to compel one to turn aside from one’s own historical researches and devote one’s time and toil to exposing the misleading theories set forth in an official publication issued at the national expense. A weary and a thankless task it is; but in Mr. Eyton’s admirable words: “the dispersion of error is the first step in the discovery of truth.”

In my ‘Studies on the Red Book of the Exchequer,’ issued last year for private circulation only, and in two special articles,[7] I have partially criticised Mr. Hall’s work and the misleading theories it contains. Of these criticisms it need only be said that the ‘English Historical Review,’ in a weighty editorial notice, observes that “The charges are very sweeping, but in my opinion they are made out.... I am bound to say that, in my opinion, Mr. Round has proved his case.”[8] The further exposures of this official work, contained in these pages[9]—especially in the paper on “the Inquest of Sheriffs,” which illustrates its wanton heresies—justify my demand that the authorities should withdraw it, till revised, from circulation.

The paper on “Castle-ward and Cornage” not only proves that the two were distinct, and gives the real explanation of their juxtaposition in the ‘Red Book,’ but contains novel information, to which I would invite attention, on the constableship of Dover Castle. The early history of this important office has been altogether erroneous.

Lastly, I must speak, very briefly, of the criticism to which my work has been exposed, although I do so with much reluctance. Honest criticism one welcomes: difference of opinion one respects. But for that uncandid criticism which endeavours to escape from facts, and which is animated only by the wish to obscure the light, no excuse is possible. The paper on “Anglo-Norman Warfare” will illustrate the tactics to which I refer; and the weight to be attached to Mr. Oman’s views may be gathered from that on “Bannockburn.” But, apart from the necessity of these exposures in the cause of historical truth, the papers which contain them will, I trust, be found of some service in their bearing on the tactics and poliorcetics of mediæval England, and on the introduction, in this country, of tenure by knight service. It is the object also of the “Bannockburn” paper to illustrate the grossly-exaggerated figures of mediæval chroniclers, a point which, even now, is insufficiently realized. Here, and elsewhere, it has been my aim to insist upon the value of records as testing and checking our chronicles, placing, as they do, the facts of history on a relatively sure foundation.

I
The Settlement of the South-and East-Saxons

I would venture, at the outset, to describe this as a “pioneer” paper. It neither professes to determine questions nor attempts to exhaust a subject of singular complexity and obscurity. It is only an attempt to approach the problem on independent lines, and to indicate the path by which it may be possible to extend our knowledge in a department of research of which the importance and the interest are universally recognised.

It is the fine saying of a brilliant scholar, I mean Professor Maitland, that “the most wonderful of all palimpsests is the map of England, could we but decipher it.”[10] But the study of place-names has this in common with the study of Domesday Book. The local worker, the man who writes the history of his own parish, is as ready to explain the name it bears as he is to interpret the Domesday formulæ relating to it in the Great Survey, without possessing in either case that knowledge of the subject as a whole which is required for its treatment in detail. On the other hand, the general student, from the very wideness of his field, is deprived of the advantage conferred by the knowledge of a district in its details. In the hope of steering a middle course between these two dangers, I have specially selected two counties, both of them settled by the Saxon folk—Sussex, with which I am connected by birth; and Essex, with which are my chief associations. And further, within these two counties I restrict myself to certain classes of names, in order to confine the field of enquiry to well-defined limits.

The names with which I propose to deal are those which imply human habitation. And here at once I part company with those, like Kemble and other writers, who appear to think it a matter of indifference, so long as a name is formed from what they term a patronymic, whether it ends in-ham or-ton, or in such suffixes as-hurst,-field,-den, or-ford. To them all such names connote village communities; to me they certainly do not. If we glance at the map of Domesday Sussex,[11] we see the northern half of the county practically still “backwoods” eight centuries ago.[12] If we then turn to the Domesday map prefixed to Manning and Bray’s Surrey, we find the southern half of that county similarly devoid of place-names. In short, the famous Andredswald was still, at the time of the Conquest, a belt, some twenty miles in width, of forest, not yet opened up, except in a few scattered spots, for human settlement. The place-names of this district have, even at the present day, a quite distinctive character. The hams and tons of the districts lying to the north and the south of it are here replaced by such suffixes as -hurst, -wood, -ley, and -field, and on the Kentish border by -den. We may then, judging from this example, treat such suffixes as evidence that the districts where they occur were settled at a much later time than those of the hams and tons, and under very different conditions. The suffix -sted, so common in Essex, is comparatively rare in Sussex, and we cannot, therefore, classify it with the same degree of certainty.

Taking, therefore, for our special sphere, the hams, the tons, and the famous ings, let us see if they occur in such a way as to suggest some definite conclusions. The three principles I would keep in view are: (1) the study, within the limits of a county, of that distribution of names which, hitherto, has been studied for the country as a whole; (2) a point to which I attach the very greatest importance, namely, the collection, so far as possible, of all the names belonging to this class, instead of considering only those which happen to be now represented by villages or parishes; (3) the critical treatment of the evidence, by sifting and correcting it in its present form. The adoption of these two latter principles will gravely modify the conclusions at which some have arrived.

There is, as Mr. Seebohm’s work has shown, nothing so effective as a special map for impressing on the mind the distribution of names. Such a map is an argument in itself. But although I have constructed for my own use special maps of Sussex and Essex, they cannot here be reproduced.

I now proceed to apply the first principle of which I spoke, that of examining a single county in the same way as others have examined the maps of England as a whole. I doubt if any county would prove more instructive for the purpose than that of Sussex, of which the settlement was developed in isolation and determined by well-defined geographical conditions. Whatever may be said of other suffixes, Mr. Seebohm has shown us that, even allowing for a large margin of unavoidable error, the terminations -ing and -ham are not distributed at random, but are specially distinctive of that portion of England which was settled by the earliest immigrants and settled the most completely. As a broad, general conclusion, this is virtually established. Now, if we turn to the map of Sussex and ask if this general principle can also be traced in detail, the first point to strike us, I think, is the close connection existing between the hams and the rivers. The people, one might say, who settled the hams were a people who came in boats. Although at first sight the hams may seem to penetrate far inland, we shall find that where they are not actually on the coast, they almost invariably follow the rivers, and follow them as far up as possible; and this is specially the case with the Arun and its tributary the Western Rother. Careful examination reveals the fact that, while to the south, round Chichester Harbour and Selsea Bill, we find several hams, and find them again to the north in the valley of the western Rother, there are none to be found in the space between, which shows that the men who settled them found their way round by the Arun and not overland. I need hardly observe that the rivers of those days were far larger than the modern streams, and their water level higher.

It is anticipating somewhat to point out that the same examination shows us a large group of tons covering this district away from the river, where we find no hams. Evidently these suffixes do not occur at random.

And now let us pass from the extreme west to the extreme east of the county. Here, instead of a group of tons with a notable absence of hams, we find a most remarkable group of hams, absolutely excluding tons. To understand the occurrence of this group on the Rother—the eastern Rother—and its tributaries, it is essential to remember the great change that has here taken place in the coast line. Unfortunately Dr. Guest, who first discussed the settlement of Sussex, entirely ignored this important change, and his followers have done the same. The late Mr. Green, for instance, in his map, follows the coast line given by Dr. Guest. Thus they wholly overlooked that great inlet of the sea, which formed in later ages the harbours of Winchelsea and Rye, and which offered a most suitable and tempting haven for the first Saxon settlers. The result of so doing was that they made the earliest invaders pass by the whole coast of Sussex before finding, at Selsea Bill, one of those marshy inlets of the sea, where they could make themselves at home. Therefore, argued Mr. Grant Allen,[13] “the original colony occupied the western half of the modern county; but the eastern portion still remained in the hands of the Welsh.” The orthodox hypothesis seems to be that the settlers then fought their way step by step eastwards, that is, towards Kent, reaching and capturing Pevensey in 491, fourteen years after their first landing.[14] As against this view, I would suggest that the distribution of Sussex place-names is in favour of vertical not lateral progress, of separate settlements up the rivers. And, in any case, I claim for the group of hams at the extreme east of the county the position of an independent settlement, to the character of which I shall return.

I must not wander too far from what is immediately my point, namely, the grouping of the hams and tons not haphazard but with cause. Even those students who discriminate suffixes, instead of lumping them together, like Kemble and his followers, make no distinction, I gather, between hams and tons. Mr. Seebohm, for instance, classes together “the Saxon ‘hams’ and ‘tuns,’”[15] and so does Professor York Powell, even though his views on the settlement are exceptionally original and advanced.[16] There are, however, various reasons which lead me to advance a different view. In the first place, the wide-spread existence, on the Continent, of ham in its foreign forms proves this suffix to be older than the settlement. ‘Ton,’ on the other hand, as is well known, is virtually absent on the Continent, which implies that it did not come into use till after the settlement in England. And as ham was thus used earlier than ton, so ton, one need hardly add, was used later than ham. The cases in Scotland, and in what is known as “little England beyond Wales,” will at once occur to the reader. Canon Taylor states of the latter that the Flemish names, such as Walterston, “belong to a class of names which we find nowhere else in the kingdom,” formed from “Walter and others common in the 12th century.”[17] But in Herefordshire, for instance, we have a Walterston; and in Dorset a Bardolfston, a Philipston, a Michaelston, and a Walterston, proving that the same practice prevailed within the borders of England. Nor need we travel outside the two counties I am specially concerned with to learn from the ‘Ælfelmston’ of Essex or the Brihtelmston of Sussex that we find ton compounded with names of the later Anglo-Saxon period. A third clue is afforded by the later version, found in the Liber de Hyda, of Alfred’s will. For there we find the ham of the original document rendered by ton. It is clear, therefore, I contend, that ton was a later form than ham. Now the map of England as a whole points to the same conclusion; for ton is by no means distinctive, like ham, of the districts earliest settled. And if we confine ourselves to a particular county, say this of Sussex, we discover, I maintain, an appreciable difference between the distribution of the hams and the tons. While the hams follow the course of the rivers, the scene of the first settlements, the tons are largely found grouped away on the uplands, as if representing a later stage in the settlement of the country. In connection with this I would adduce the “remarkable passage,” as Mr. Seebohm rightly terms it, in one of King Alfred’s treatises, where he contrasts the “permanent freehold ham” with the new and at first temporary ton, formed by ‘timbering’ a forest clearing in a part not previously settled.[18] It is true that Mr. Seebohm, as I have said, recognises no distinction, and even speaks of this example as “the growth of a new ham”; but it seems to me to confirm the view I am here advancing. It is obvious that if such a canon of research as that ham (not ton) was a mark of early settlement could be even provisionally accepted, it would greatly, and at once, advance our knowledge of the settlement of England. Although this is nothing more than a ‘pioneer’ paper, I may say that, after at least glancing at the maps of other counties, I can see nothing to oppose, but everything to confirm, the view that the settlers in the hams ascended the rivers (much as they seem, on a larger scale, to have done in Germany); and a study of the coast of England from the Tweed to the British Channel leads me to believe that, as a maritime people, their settlements began upon the coast.

I now pass to my second point—the insufficient attention which has hitherto been paid to our minor place-names. Kemble, for instance, working, as he did, on a large scale, was dependent, so far as names still existing are concerned, on the nomenclature of present parishes. And such a test, we shall find, is most fallacious. Canon Taylor, it is true, has endeavoured to supplement this deficiency,[19] but the classification of existing names, whether those of modern parishes or not, has not yet, so far as I can find, been even attempted. Hitherto I have mainly spoken of Sussex, because it is in that county that place-names can be best studied; the Essex evidence is chiefly of value for the contrast it presents. The principal contrast, and one to which I invite particular attention, is this: confining ourselves to the names I am concerned with—the ings, hams, and tons—we find that in Essex several parishes have only a single place-name between them, while in Sussex, on the contrary, a single parish may contain several of these place-names, each of them, surely, at one time representing a distinct local unit. This contrast comes out strongly in the maps I have prepared of the two counties, in which the parishes are disregarded, and each place-name separately entered. I do not pretend that the survey is exhaustive, especially in the case of Sussex, as I only attempt to show those which are found on an ordinary county map, together with those, now obsolete, which can safely be supplied from Domesday. But, so far as the contrast I am dealing with is concerned, it is at least not exaggerated.

As the actual names are not shown, I will now adduce a few examples. In Sussex, Burpham is composed of three tythings—Burpham, Wepham, Pippering; Climping comprises Atherington and Ilesham; Offham is included in South Stoke; Rackham in Amberley; Cootham in Storrington; Ashton, Wellingham, and Norlington in Ringmer; Buddington in Steyning; and Bidlington in Bramber.

In Essex, on the other hand, ‘Roothing’ does duty for eight parishes, Colne for four, Hanningfield, Laver, Bardfield, Tolleshunt, and Belchamp for three each, and several more for two. There are, of course, in Sussex also, double parishes to be found, such as North and South Mundham, but they are much scarcer.

We may learn, I think, a good deal from the contrast thus presented. In the first place, it teaches us that parochial divisions are artificial and comparatively modern. The formula that the parish is the township in its ecclesiastical capacity is (if unconsciously adopted) not historically true. Antiquaries familiar with the Norman period, or with the study of local history, must be acquainted with the ruins or the record of churches or chapels (the same building, I may observe in passing, is sometimes called both ecclesia and capella[20]), which formerly gave townships now merged in parishes a separate or quasi-separate ecclesiastical existence. In Sussex the present Angmering comprises what were once three parishes, each with a church of its own. The parish of Cudlow has long been absorbed in that of Climping. Balsham-in-Yapton was formerly a distinct hamlet and chapelry. Conversely, the chapelries of Petworth have for centuries been distinct parishes.

In Essex we have examples of another kind, examples which remind us that the combination or the subdivision of parishes are processes as familiar in comparatively modern as in far distant times. The roofless and deserted church to be seen at Little Birch testifies to the fact that, though now one, Great and Little Birch, till recently, were ecclesiastically distinct. In the adjoining parish of Stanway, the church, similarly roofless and deserted, was still in use in the last century.

Again, the civil unit as well as the ecclesiastical, the village, like the parish, may often prove misleading. It is, indeed, very doubtful whether we have ever sufficiently distinguished the manor and the village. If we construct for ourselves a county map from Domesday, we shall miss the names of several villages, although often of antiquity; but, on the other hand, shall meet with the names of important manors, often extending into several parishes, often suggesting by their forms a name as old as the migration, yet now represented at most by some obscure manor, and perhaps only by a solitary farm, or even, it may be, a field. In Sussex, for instance, the ‘Basingham’ of Domesday cannot now be identified; its ‘Belingeham’ is doubtful; its ‘Clotinga’ is now but a farm, as is ‘Estockingeham.’ ‘Sessingham’ and ‘Wiltingham’ are manors. In Essex ‘Hoosenga’ and ‘Hasingha’ occur together in Domesday, and are unidentified. Nor have I yet succeeded in identifying ‘Plesingho,’ a manor not only mentioned in Domesday, but duly found under Henry III. Morant, followed by Chisenhale-Marsh, identified it wrongly with Pleshy. Such names as these, eclipsed by those of modern villages, require to be disinterred by archæological research.

Another point on which light is thrown by the contrast of Essex and Sussex is the theory tentatively advanced by Mr. Maitland in the ‘Archæological Review,’ that the Hundred and the township may, in the beginning, have been represented by the same unit.[21] Broadly speaking, he adduced in support of this hypothesis the originally large township of Essex, proved by the existence of a group of villages bearing the same name, comparing it with the small Hundreds characteristic of Sussex. But in Sussex, I think, the small Hundreds were coincident with those many small townships; while in Essex the scattered townships are coincident with larger Hundreds. And this leads me to suggest that the Saxon settlements in Sussex lay far thicker on the ground than those found in Essex, and that we possibly find here some explanation of the admitted silence as to the East-Saxon settlement contrasting with the well-known mention of that in Sussex. It seems to me highly probable that Essex, in those remote times, was not only bordered and penetrated by marshes, but largely covered with forest. It is, perhaps, significant that in the district between Westham and Boreham, some twenty-five miles across as the crow flies, there is not a ham to be found.

From this I turn to the opposite extreme, that group of hams on the ‘Rother’ and its tributaries, thirty-seven in number. Isolated alike from ings and tons, and hemmed in by the spurs of the Andredswald, it is, perhaps, unique in character. Nowhere have I lighted on a group of hams so illustrative of the character of these settlements, or affording a test so admirable of the alleged connection between this suffix and the villa of the Roman Empire.

One of the sections of Mr. Seebohm’s work is devoted to what he terms “the connection between the Saxon ‘ham,’ the German ‘heim,’ and the Frankish ‘villa.’” This, indeed, it may fairly be said, is one of the important points in his case, and one to which he has devoted special research and attention. Now, I am not here dealing with the equation of ‘ham’ and ‘villa.’ If I were, I should urge, perhaps, that, as with the ‘Witan’ of the English and the ‘Great Council’ of the Normans, it does not follow that an equation of words involves their absolute identity of meaning. I confine myself to the suffix ‘-ham.’ “Its early geographical distribution,” Mr. Seebohm has suggested, “may have an important significance.” With this, it will be seen, I entirely agree. But, if the distribution is important, let us make sure of our facts; let us} as I urge throughout this volume, test and try our evidence before we advance to our conclusion. When Mr. Seebohm informs us that “the ‘hams’ of England were most numerous in the south-eastern counties, finding their densest centre in Essex,” the statement must startle any one who has the least acquaintance with Essex, where the termination ‘-ham’ is comparatively rare in place-names. On turning to Mr. Seebohm’s map, one is still further surprised to learn that its “local names ending in ‘ham’” attain in Domesday the enormous proportion of 39 per cent. The clue to the mystery is found in a note that “in Essex the h is often dropped, and the suffix becomes am.” For the whole calculation is based on a freak of my old friend, the Domesday scribe. The one to whom we are indebted for the text of the Essex survey displayed his misplaced scholarship in Latinizing the English names so thoroughly, that not only did Oakley, the first on the list, become ‘Accleia,’ but even in the accusative, “Accleiam tenet Robertus.” Thus we need travel no further than the first name on the index to learn how Mr. Seebohm’s error was caused. Elmstead, Bonhunt, Bentley, Coggeshall, Danbury, Dunmow, Alresford, and many other such names, have all by this simple process been converted into ‘hams.’ I hasten to add that my object in correcting this error is not to criticise so brilliant an investigator and so able a scholar as Mr. Seebohm, but to illustrate the practical impossibility of accomplishing any scientific work in this department of research until the place-names of England have been classified and traced to their origin. I am eager to see this urgent work undertaken county by county, on much the same lines as those adopted by the Government in France. It seems to me to be eminently a subject for discussion at the Annual Congress of Archæological Societies.

If it were the case that the English ham represents the Roman villa, this remarkable group on the borders of Kent and Sussex should indicate a dense Roman settlement; but of such settlement there is, I believe, no trace existing. Conversely, we do not find that the sites of Roman villas are denoted by the suffix ham.[22]

From considering this group as a whole, I advance to two settlements on what is known as the Tillingham River, namely, Billingham and Tillingham. One would not easily find names more distinctive of what Kemble insisted on terming the mark system, or what later historians describe as clan settlement. Parenthetically, I may observe that while ham is common in Sussex, the compound ingham is not. This is well seen in the group under consideration. The same may, I think, be said of Essex, while in North Suffolk ingham begins to assert its predominance. The frequent occurrence in Norfolk and Lincolnshire renders it a note of Anglian rather than Saxon settlement.[23] And now for Billingham and Tillingham. Billing is one of the most common of the so-called patronymics; and there is a Tillingham in Essex. Whether we turn to the specialist works of such writers as Stubbs and Green, or to the latest compendia of English history as a whole, we shall virtually always read that such names as these denote original settlement by a clan.[24]

In venturing to question this proposition, I am striking at the root of Kemble’s theory, that overspreading theory of the Mark, which, as it were, has shrunk from its once stately splendour, but in the shadow of which all our historians since his time have written. Even Professor York Powell, although he rejects the mark theory,[25] writes of “the first stage” of settlement: “We know that the land was settled when clans were powerful, for the new villages bear clan names, not personal names.”[26] The whole theory rests on the patronymic ing, which Kemble crudely treated as proving the existence of a mark community, wherever it occurs in place-names.[27]

Now the theory that ing implies a clan, that is, a community united by blood or by the belief in a common descent,[28] may be tested in two distinct ways. We may either trace its actual use as applied to individuals or communities; or we may examine the localities in the names of which it occurs. I propose to do both. The passage usually adduced to prove the ‘clan’ meaning is the well-known genealogy in the Anglo-Saxon Chronicle: “Cerdic was Elesing, Elesa was Esling, Esla was Gewising,”[29] etc. Even Mr. Seebohm reluctantly admits, on this “evidence of the Anglo-Saxon Chronicle,” that ing was used as alleged. But it always seemed obvious to me that this passage, so far from proving the ‘clan’ meaning, actually proved the opposite, namely, that the patronymic changed with every generation. Again, if we turn from the Chronicle to the Anglo-Saxon charters, we find inga normally used to denote the dwellers at a certain place, not the descendants of a certain man. It is singular that Kemble, although he was the first to make an exhaustive study of these charters, classed such names with the other ings, from which they were quite distinct.[30] His enthusiasm for the ‘mark’ carried him away. In Sussex, we have, as it seems to me, a very excellent illustration; the name of Angmering, the present form, occupies, as it were, a medial position between the “Angemare” of Domesday and the “Angmeringatun” of Alfred’s will. Here, surely, the Angmeringas were those who dwelt at Angmer, not a ‘clan’ descended from a man bearing that name.

I will not, however, dwell on this side of the argument, more especially as I would rather lay stress on the other line of attack. For this is my distinctive point: I contend that, in studying the place-names into which ing enters, attention has hitherto exclusively, or almost exclusively, been devoted to those now represented by towns or villages. With these it is easy to associate the idea of a clan settlement. But what are we to make of such cases as our Sussex Billingham and Tillingham? We shall search for them in vain in Lewis’ Topographical Dictionary; and yet they are names of the same status as fully developed villages. As a Sussex antiquary has observed (though I cannot accept his explanation): “In the names of many farms we shall likewise find names which also mark whole parishes in the county.” Canon Taylor has unconsciously recorded, in the adjoining county of Kent, evidence to the same effect, observing that “the lone farmhouses in Kent, called Shottington, Wingleton, Godington, and Appleton, may be regarded as venerable monuments, showing us the nature of the Saxon colonization of England.”[31] I say that this evidence is unconscious, for the Canon applies it only to the evolution of the ton, and seems not to have observed its bearing on that compound ing, which he, like Kemble, fully accepts as proof of a clan community. From Shottington and Godington, as from Billingham and Tillingham, Kemble would have confidently deduced the settlement of a ‘mark’ or clan community; and yet, when we learn what the places are, we see that they represent a settlement by households, not by communities.

Here, then, is the value of these cases of what we may term arrested development: they warn us against the rashness of assuming that a modern or even a mediæval village has been a village from the first. The village community may be so far from representing the original settlement as to have been, on the contrary, developed from what was at first but a farmstead. The whole argument of such scholars as Professor Earle here and Dr. Andrews in America is based on the assumption that the land was settled by communities, each of them sufficiently large to have a head, whether civil or military. To that supposition such names as I have mentioned are, I think, fatal.

Yet another point must be touched on as to this alleged patronymic. To Kemble, as I have said, it was of small moment what suffix his ‘marks’ bore. Indeed, those that denoted forest were to him specially welcome, because he associated the idea of a ‘mark’ with that of a forest clearing. But we who have seen that such suffixes as -field, -hurst and -den are distinctive of those districts untouched by the early settlers cannot recognise such names, for instance, as the Itchingfield or Billingshurst of Sussex as denoting village communities. Again, in the Anglo-Saxon charters the characteristic den of Kent is frequently preceded by ing; and if these dens were clearly from the context only forest pastures for swine, we must here also reject the ing as proof of a clan community. One may also glance in passing at such names as the “Willingehala” of Essex, now “Willingale,” and ask whether a clan community is supposed to have settled in a hall?[32]

I trust that I have now sufficiently shown that even where ing genuinely enters into the composition of a place-name it is no proof of settlement by a clan. Kemble looked on the typical ‘mark’ as “a hundred heads of houses,” which he deemed “not at all an extravagant supposition.”[33] I think that even at the present day a visit to the hams and tons of Sussex, and, in some cases, to the ings, would lead us in practice to the opposite conclusion, and would throw the gravest doubt on the theory of the village community. I was trained, like others of my generation, to accept that theory as an axiomatic truth; but difficult as it is to abandon what one has been so taught, the solitary manor house, the lonely farm, is a living protest against it. The village community of the class-room can never have existed there. On paper it holds its own: solvitur ambulando.

But the fact that a place bearing a typical clan name may prove to have been but a single homestead takes us farther than this. Ing, which Canon Taylor has described as “the most important element which enters into Anglo-Saxon names,” has been held to denote settlement not merely by a clan, but by a portion of a tribe bearing, both in England and abroad, one common name. Kemble insisted strongly upon this,[34] and is duly followed by Canon Taylor[35] and others. On the same foundation Mr. Andrew Lang has erected yet another edifice, tracing the occurrence in scattered counties of the same clan name to the existence of exogamy among our forefathers. And this ingenious suggestion has been adopted by Mr. Grant Allen.[36] But the very first instance he gives, that of the Hemings, will not stand examination.[37]

As yet I have been dealing with those ‘clan names’ in which the presence of the ing is genuine; and I have been urging that it is not proof, as alleged, of settlement by a clan. I now pass to those place-names in which the ing is not genuine, but is merely a corruption. That such names exist has always, of course, been admitted,[38] but their prevalence has not been sufficiently recognised. And not only are there large deductions, in consequence, to be made from the so-called clan names, but even in cases where the ing is genuine the prefix is often so corrupt that the name of the clan deduced from it is altogether wrong.

Let us take some instances in point. Kemble deduced the existence of the Brightlings (‘Brightlingas’) from Brightling in Sussex and Brightlingsea in Essex. Nothing, at first sight, could seem clearer. And yet, on turning to Domesday, we find that the Sussex Brightling is there entered as Brislinga—suggesting that Somerset Brislington from which Kemble deduced the Brislings—while Brightlingsea appears in the Essex Domesday as ‘Brictriceseia,’ and in that of Suffolk as ‘Brictesceseia,’ from which forms is clearly derived the local pronunciation ‘Bricklesea.’ So much for the Brightlings. Yet more striking is the case of an Essex village, Wormingford. Kemble, of course, detects in it the name ‘Wyrmingas.’ Yet its Domesday name is Widemondefort,’ obviously derived from ‘Widemond,’ the name of an individual.[39] Here the corruption is so startling that it is well to record the transition form ‘Wiremundeford,’ which I find in the 13th century.[40] Now, as I have often to point out in the course of my historical researches, however unpopular it may be to correct the errors of others, those errors, if uncorrected, lead too often to fresh ones. Thus, in this case, the ‘Wyrmingas,’ wrongly deduced from Wormingford, have been claimed by scholars as sons of the ‘worm,’ and, therefore, as evidence that ‘Totemism’ prevailed among the Anglo-Saxons. It would take me, I fear, too far afield to discuss the alleged traces of Totemism; but when we find Mr. Grant Allen asserting that “the oak has left traces of his descendants at Oakington in Cambridge” (shire), one has to point out that this place figures in Domesday as ‘Hochinton(e)’[41] in no fewer than five entries, although Kemble derives from it more suo the ‘Æcingas.’ But a few more instances of erroneous derivation must be given in order to establish clearly the worthlessness of Kemble’s lists. How simple it seems to derive, with him, the ‘Storringas’ and ‘Teorringas’ from Storrington, Sussex, and Tarrington, Herefordshire, respectively. Yet the former, in Domesday, is ‘Storgetune’ or ‘Storchestune,’ while the latter is ‘Tatintune’ in both its entries. It might be suggested that the error is that of the Domesday scribe, but in this case I have found the place entered in several documents of the next century as Tadinton or Tatinton, thus establishing the accuracy of Domesday. Indeed, in my experience, the charters of the 12th century prove that Domesday nomenclature is thoroughly deserving of trust. The climax of Kemble’s derivations is reached perhaps in Shillingstone, from which Dorset village he duly deduces the ‘Scyllingas.’ For, as Eyton has shown, its name was ‘Acford,’ but, from its Domesday tenant, Schelin, it became known as Ockford Eskelling, Shilling Ockford, and finally, by a yet bolder corruption, Shillingstone.[42] As if to make matters worse, Kemble treats ‘Shilling-Okeford’ and ‘Shillingstone’ as two distinct places. Could anything, one asks, be more unfortunate than this? Alas, one must answer Yes. The great clan of the ‘Cypingas’ is found in eight counties: at least so Kemble says. I have tested his list and discovered that the names which prove the existence of his clan are Chipping Ongar, Chipping Barnet, Chipping Sodbury, Chipping Campden, Chipping Wycombe, Chipping Warden, and Chipping Norton. Even the historical tyro would avoid this wild blunder; he would know that Chipping was about as much of a clan name as is Cheapside. After this final example, it can hardly be disputed that Kemble’s lists are merely a pitfall for the unwary.

Yet we still follow in his footsteps. Take such a case as that of Faringdon, which Mr. Grant Allen, we have seen, selected as a typical instance of the ing patronymic in place-names.[43] If we turn to Domesday, we find in Berks a ‘Ferendone,’ in Northants a ‘Ferendone’ or ‘Faredone,’ in Notts a ‘Ferendone’ or ‘Farendune,’ in Hants a ‘Ferendone.’ These names were all the same; and yet they have become ‘Farndon’ in Notts and Northants, ‘Faringdon’ in Berks, and ‘Farringdon’ in Hants. Farringdon, therefore, is no more a clan name than is the Essex Parndon, the ‘Perenduna’ of Domesday. But, indeed, in Essex itself, there is an even better illustration. We learn from Canon Taylor that “the Thurings, a Visigothic clan, mentioned by Marcellinus ... are found ... at Thorrington in Essex.” Kemble had previously described them as “likely to be offshoots of the great Hermunduric race, the Thyringi or Thoringi, now Thuringians, always neighbours of the Saxons,[44] and claims the Essex Thorrington” as their settlement.[45] Now Thorington in the first place was not a ton, and in the second place had not an ing. Both these forms are corruptions. In Domesday it occurs twice, and both times as ‘Torinduna.’ With this we may compare ‘Horninduna,’ which is the Domesday form of Horndon, and occurs frequently. Therefore Thorington and Thorndon, like Farringdon and Farndon, were both originally the same name and destitute alike of ing.

As to the names ending in ing, with no other suffix, I prefer, for the present, to reserve my opinion. Kemble’s hypothesis, however, that they were the parent settlements, and the hams and tons their filial developments, seems to me to have little support in the facts of their actual distribution. If in that distribution there is a feature to be detected, it is, perhaps, that the ings are found along the foot of the downs. This, at least, is often observable. Another point deserving of attention is that, in its French form, igny, this suffix seems as distinctive of the ‘Saxon’ settlement about Bayeux as it is absent in that which is found in the Boulogne district. But these are only, as it were, sidelights upon the problem; and this, as I said, is nothing more than a ‘pioneer’ paper.

I close with a point that appears to me of no small importance. To the east of Sussex and the south of Sussex there lay that so-called Jutish land, the county of Kent. As I pointed out years ago, in my ‘Domesday Studies,’ the land system of Kent is found in the Great Survey to be essentially distinct from that which prevailed in other counties. It was not assessed in ‘hides,’ but in ‘solins,’ that is, the sulungs of the natives, the land of a suhl or plough. The yokes, or subdivisions, of this unit are also directly connected with the plough. But the hide and virgate of other counties are, as I pointed out, not connected in name with the plough.[46] Now if we work through the land charters printed by Professor Earle, we find that this Domesday distinction can be traced back, clear and sharp, to the earliest times within their ken. We read in an Anglo-Saxon charter of “xx swuluncga,” while in Latin charters the normal phrase is the land of so many ploughs (‘terra trium aratrorum,’ ‘terra decem aratrorum,’ etc.); we even meet with the phrase, “decem aratrorum juxta æstimationem provinciæ ejusdem.”[47] In another charter “v aratra” equates “fifsulung landes.” But in other counties the normal terms, in these charters, for the land units are “manentes” and “cassati,”[48] which occur with similar regularity. A cleavage so ancient and so clear as this, in the vital sphere of land division, points to more than a separate rule and confirms the tradition of a distinct origin.

II
Ingelric the Priest and Albert of Lotharingia

In my paper on “Regenbald, Priest and Chancellor,”[49] I was able to trace, by combining the evidence of Domesday and of charters, the history of a “priest” of Edward the Confessor, who became the “priest” of his successor also, and held of him rich possessions in churches and lands. Another churchman who flourished both before and after the Conquest, and must have enjoyed the favour both of the Confessor and of the Conqueror, was Ingelric, first dean of the house of St. Martin’s-le-Grand, whose lands had passed before Domesday to Count Eustace of Boulogne. Mr. Freeman was interested in Ingelric as a “commissioner for redemption of lands,” but only knew him as a layman. Nor indeed is there anything in Domesday to suggest that he was other. To Mr. W. H. Stevenson belongs the credit of proving that he was a priest by printing “an old English charter of the Conqueror,” confirming the foundation of St. Martin’s-le-Grand, in which the “cujusdam fidelis mei Ingelrici scilicet peticioni adquiescens” is equated by “æfter Ingelrices bene mines preostes.”[50] It was similarly as “minan preoste” that William had described Regenbald.

The charter I shall now deal with was not known to Mr. Stevenson, and has not, I believe, been printed. It is of real historical interest, apart from the fact that among its witnesses we find Ingelric “the priest.”

Mr. Freeman held that the reconciliation between the Conqueror and the Abbot of Peterborough—Brand, the Englishman, whose election had been confirmed, even after the Battle of Hastings, by the ætheling Eadgar—was one of the earliest events after William’s coronation.[51] To that episode I do not hesitate to assign a charter entered in the Peterborough ‘Liber Niger’ belonging to the Society of Antiquaries. It is a general confirmation of the abbey’s possessions, “petente abbate Brand,”[52] and is witnessed thus:

Huic testes affuere: Aldredus Eboracensis archiepiscopus; Wlwinus Lincoliensis episcopus; Merlesuen vicecomes; Ulf filius Topi; Willelmus comes; Willelmus Malet; Ingelri[cus] presbyter.

Here we have first Ealdred, by whom William had been crowned; then Wulfwig, bishop of Dorchester, here described as bishop “of Lincoln.” The mention of Mærleswegen is of special importance, for this great English noble had been left in charge of the North by Harold on the eve of the Battle of Hastings, and rose in revolt against William in the summer of 1068. Here we have evidence of his presence at William’s court, when his movements were unknown to Mr. Freeman. We see, moreover, that he was still sheriff (of Lincolnshire). “Ulf filius Topi,” who appears in other Peterborough charters, had given “Mannetorp,” Lincolnshire, and other lands to the abbey.

It is very remarkable that the Norman witnesses are only entered after these Englishmen, although the first is “earl William,” in whom we must see the Conqueror’s friend, William Fitz Osbern, already, apparently, earl of Hereford. Sufficient attention has hardly been given to this early creation or to the selection of so distant a county as Herefordshire for William’s earldom.

In addition to this charter, there is known to me another, little later probably, the last witness to which is entered as “Ego Ingelricus ad hoc impetrandum obnixe studui.” This brings me to the third charter that I shall deal with in connection with Ingelric. This is the one I mentioned at the outset as granted by the Conqueror at his request, and edited with so much care and learning by Mr. W. H. Stevenson. This, in its stilted, antique form, has much in common with the one preceding, while its style combines those of the two others. I place the three together for comparison:

(1) Ego Willelmus dei beneficio rex Anglorum.

(2) jure hereditario Anglorum patrie effectus sum Basileus.

(3) Ego Willelmus Dei dispositione et consanguinitatis hereditate Anglorum basileus.

Mr. Freeman looked with suspicion on this third charter, which he termed “an alleged charter of William.”[53] His criticism that, though dated 1068, its list of witnesses closes with the two papal legates who visited England in 1070, is a perfectly sound one. Mr. Stevenson ignored this difficulty in his paper; and, on my pointing it out, still failed to explain the positive “huic constitutioni interfui” of Cardinal John. Awkward, however, as the difficulty is, the other attestations are so satisfactory that we must treat these as subsequent additions rather than reject the charter.

The remarks which immediately follow are intended only for students of what is uncouthly known as ‘diplomatic,’ a study hitherto much neglected in England. In this charter, as printed in Mr. Stevenson’s paper, there is appended the clause:

Scripta est hec cartula anno ab incarnatione Domini MLXVIII^o scilicet secundo anno regni mei.

A corresponding clause is found in the old English version of the text which follows it. But in the Latin text the clause is followed by these words:

Peracta vero est hec donacio[54] die Natali Domini; et postmodum in die Pentecostes confirmata, quando Mathildis conjux mea ... in reginam ... est consecrata.

Mr. Freeman somewhat carelessly confused the two clauses:

The charter (sic) is said to have been granted at the Christmas feast of 1068 (evidently meaning 1067), and to have been confirmed at the coronation of the queen at the following Pentecost (iv. 726).

Mr. Stevenson follows him in this confusion, but carries it much further. Speaking of “supplementary confirmations,” as used in William’s chancery, he writes:

We have one in this very charter, which was executed (peracta) on Christmas Day, 1068 (i.e. 1067), but was afterwards confirmed on the occasion of Matilda’s coronation at Whitsuntide, 1068. If we had the original charter, we should probably find that the clause relating to the Whitsuntide confirmation had been added, as in similar continental instances, on a blank space in the charter. Ingelric was, as we know from this grant, one of William’s clerks, and he must have been a man of considerable influence to have obtained a diploma from a king who was so chary in the granting of diplomata, and to have, moreover, obtained the execution of it at so important a ceremony as the king’s coronation, and a confirmation of it at the queen’s coronation.[55]

In the elaborate footnotes appended to this passage there are three points to be dealt with.

The first is “the king’s coronation” as the time when the charter was executed. Mr. Stevenson writes:

Freeman, Norman Conquest, iv. 724, says that the date of the charter, Christmas 1068, evidently means 1067, the date of William’s coronation; etc.... There are good grounds, therefore, for holding that the witnesses were spectators of William’s coronation, which gives the charter its greatest historical importance.[56]

But, as we have seen, it is not the fact that Mr. Freeman spoke of Christmas 1067 as “the date of William’s coronation.” That event took place, as all the world knows, at Christmas, 1066, and so was long previous to this gift and charter. Mr. Stevenson’s error is a strange one.

The second point is that of the “supplementary confirmation.” Mr. Stevenson, referring us to the best parallel, writes:

In the case of the council (or rather placitum) of 1072 concerning the subjection of York to Canterbury, which, like the charter under consideration, received a supplementary ratification, a second text was drawn up for the later action.

I here break off to print, for convenience, the parallel clauses in these documents side by side.

Resuming now Mr. Stevenson’s note on the documents of 1072, at the point where I broke it off, we read:

The originals of both still exist. The first, dated at Winchester at Whitsuntide,[58] is validated only by the crosses of William and his queen, the papal legate, both archbishops and four bishops (Palæographical Society, i. fol. 170). The second ... is dated at Windsor, also at Whitsuntide, and is attested by additional bishops, and by numerous abbots.

As the former document (A.2 of the Canterbury charters, apparently overlooked till some twenty years ago) could not possibly be “dated at Winchester at Whitsuntide,” one turns to the text as given by the Palæographical Society, only to find that these words are sheer imagination on Mr. Stevenson’s part. There is nothing of the kind to be found there. Owing to this incomprehensible error, he has altogether misunderstood these “supplementary confirmations.” The clauses I have printed side by side must not be broken up. The earlier, like the later, is a consistent whole, added at one time.[59]

When, then, was the “Ingelric” charter actually drawn up? Mr. Stevenson, following, we have seen, Mr. Freeman’s loose expressions, tells us that “as the present charter (sic) was peracta at Christmas, 1067, and confirmata at Whitsuntide, it was most probably written at the former date.” But it was the “donacio,” not the “charter,” which was “peracta” at Christmas. The text only tells us of the charter that it was written “anno ab incarnacione Domini MLXVIII^o.” My own view is that the charter was written not at Christmas, 1067 (which was the date of the act of gift), but at (or after) Whitsuntide, 1068. I base this conclusion on the first three witnesses:

• Ego Willelmus rex Anglorum, etc.
• Ego Mathildis regina consensum præbui.
• Ego Ricardus regis filius annui.

Matilda was not “queen” till Whitsuntide, 1068, and was not even in England at Christmas, 1067. If it be urged that, even though found in this position, her name was interpolated afterwards, I reply that the name of William’s eldest son, Robert, would then have been similarly added. The fact that we find, instead, his second son, Richard (afterwards killed while hunting in the New Forest) is to me the strongest possible evidence that Robert had remained behind, as regent, in Normandy when his mother came over to England to be crowned. The most probable date, therefore, for the execution of this charter is that of her coronation at Westminster, 1068. It preserves for us, in that case, the names of the magnates present on that occasion, including Hugh bishop of Lisieux, who may well have escorted her from Normandy, and thus have attended the ceremony.[60].

My third point follows as a corollary from this conclusion. For if the charter was drawn up at Whitsuntide, 1068, not at Christmas, 1067, there is an end of Mr. Stevenson’s argument and conclusion:

The 25th December in the second year of William’s reign was in 1067 according to our reckoning. But the old system of reckoning the year “ab Incarnatione” began the year on 25th December. This was the old English system, and this charter proves that William’s chancery also commenced the year at the Nativity.[61]

The time spent on this important charter has not been wasted. We have found that one who stands in the front rank of English philologists, and for whom the same would, doubtless, be claimed in “diplomatic,” may arrive, in spite of great learning, at quite erroneous conclusions, simply from inexact treatment of the evidence before him.

A word more on Ingelric. According to Mr. Freeman, “that Ingelric was an Englishman seems plain.”[62] Mr. Stevenson, however, who has specially studied the subject of personal names, holds that this was Frankish. The St. Martin’s charter specially speaks of his having acquired his lands under Edward the Confessor. Mr. Stevenson, however, goes further, and states, as we have seen, that it proves him to have been “one of William’s clerks” (sic); and he argues that “if he was a chancery clerk, he may have continued the traditions of Edward’s chancery.” It is remarkable, however, that in an Exeter charter (1069) to which Mr. Stevenson refers us, he again attests, as in two of the charters dealt with above, as “Ingelricus presbyter.” I have chosen, therefore, for this paper the style “Ingelric the priest.”

No question of origin can arise in the case of a third personage, who also enjoyed the favour both of Edward and of his successor, namely, Albert of Lotharingia. Known hitherto as having, it is supposed, given its name to Lothbury—for the “Terra Alberti Loteringi” is mentioned in the list of London wards temp. Henry I.[63]—he occurs in many places on the pages of Domesday. As “Albertus Lothariensis” we find him a tenant-in-chief in the counties of Herefordshire and Beds (186, 216b2), one of his manors in the latter county having been held by him, we read, under Edward the Confessor; and he also occurs by the same style as holding under the latter king at Hatton, Middlesex (129). But, so far, there is nothing to show that Albert was a cleric.

It is a Westminster Abbey charter that supplies the missing clue:

Willelmus rex Anglorum Francis et Anglis salutem. Sciatis me dedisse Sancto Petro Westmonasterii et abbati Gilleberto ecclesias de Roteland et terras pertinentes ad easdem ecclesias sicut Albertus Lotharingius de me tenebat ipsas ecclesias cum omnibus pertinentibus ad ipsas. Teste Hugone de Portu.[64]

Turning to “Roteland” in Domesday, we find that the last name in the list of its tenants-in-chief is that of “Albertus clericus,” who holds the churches of Oakham, Hambleton, and St. Peter’s, Stamford, “cum adjacentibus terris eisdem ecclesiis ... de rege,” the whole forming a valuable estate. Again, we read under Stamford: “Albertus unam æcclesiam Sancti Petri cum duabus mansionibus et dimidia carucata terre quæ jacet in Rotelande in Hemeldune; valet x sol.” (336 b). Following up this clue, we recognise our man in the “Albertus clericus” who holds at “Eddintone,” in Surrey (30, 36 b), and doubtless also in “Albertus clericus” who held land as an under-tenant at Windsor (56 b). Nay, it is difficult to resist the conclusion that he is also the “Albertus capellanus” who, at the end of the Kent Domesday (14 b), has a page all to himself as tenant-in-chief of Newington. Thus in the official index to Domesday we find Albert entered under “clericus,” “Lothariensis,” “Albertus,” and (probably) “capellanus,” and yet, in each case, it is the same man. Regenbald, exactly in the same way, is entered under ‘Cirecestre,’ ‘presbyter’ and ‘Reinbaldus.’ In my ‘Feudal England’ I have similarly identified (p. 167) “Eustachius,” one tenant-in-chief, with “Eustachius vicecomes,” another (and with “Eustachius,” an under-tenant),[65] and “Oger,” a Northamptonshire tenant-in-chief, with Oger “Brito,” a Lincolnshire one (p. 220). In the Eastern counties the Breton founder of the house of Helion is similarly indexed under ‘Britto’ for Essex, ‘Herion’ for Suffolk, and ‘Tehelus’ for Norfolk. Small as these points may seem, their ultimate consequence is great, for they still further reduce the number of tenants-in-chief. When the history of these magnates is more fully known, it will probably be found that those who held in capite per servitium militare, thus excluding, of course, mere serjeants, etc., were a mere handful compared with the vast total given by Ellis and others.

Albert’s Lotharingian origin becomes of special interest now that we know he was a cleric, for Mr. Freeman devoted a special appendix to “Lotharingian churchmen under Edward.”[66] Unfortunately he was not acquainted with the case of Albert. Dr. Stubbs also has dwelt on the importance, for the church, of “the increased intercourse with the empire, and especially with Lorraine,” under Edward the Confessor.[67] He alludes, without committing himself to it, to Mr. Freeman’s somewhat fanciful theory on the subject.

III
Anglo-Norman Warfare

Having devoted special study to the art of war in the Norman period, including therein the subject of castles, I may have, perhaps, some claim to deal with the latest work on a topic which requires for its treatment special knowledge. When a treatise assumes a definite character, and is likely to be permanently consulted, it calls for closer criticism than a mere ephemeral production, and on this ground I would here discuss some points in Mr. Oman’s ‘History of the Art of War’ (1898).

Mr. Oman issued, so far back as 1885, ‘The Art of War in the Middle Ages,’ so that he enjoys, on this subject, the advantage of prolonged study. In 1894 he contributed to ‘Social England’[68] an article on “Norman Warfare,” to which I shall also refer. I should add that in his first (1885), as in his later work (1898), Mr. Oman received the help of Mr. F. York Powell, now Regius Professor of Modern History at Oxford.

The first point I propose to consider is that of the famous English “formation” before the Norman Conquest. Mr. Oman originally wrote as follows:

The tactics of the English axemen were those of the column; arranged in a compact mass, they could beat off almost any attack, and hew their way through every obstacle (‘Art of War,’ p. 24).

This was also the view of the late Professor Freeman, who wrote of the battle of Maldon that—

The English stood, as at Senlac, in the array common to them and their enemies—a strong line, or rather wedge of infantry, forming a wall with their shields.

At the battle of Hastings (“Senlac”) itself he tells us—

The English clave to the old Teutonic tactics. They fought on foot in the close array of the shield wall.

They were ranged, he held, “closely together in the thick array of the shield wall.” He had well observed that “the Norman writers were specially struck with the close array of the English,” and had elsewhere spoken of “the close array of the battle-axe men,” and of “the English house-carls with their ... huge battle axes,” accustomed to fight in “the close array of the shield wall.”[69]

To this formation, it is necessary to observe, the term testudo was applied. At the battle of Ashdown, Freeman wrote:

Asser calls it a testudo or tortoise. This is the shield wall, the famous tactic of the English and Danes. We shall hear of it in all the great battles down to the end.

Florence adopts the same word in describing the formation of the rival hosts on that occasion:

Pagani in duas se turmas dividentes, æquali testudine bellum parant (i. 83).

Ælfred ... Christianas copias contra hostiles exercitus ... dirigens ... testudine ordinabiliter condensata (i. 84).

So, too, at the battle of Ethandun:

Ubi contra Paganorum exercitum universum cum densa testudine atrociter belligerans (i. 96).

Again, in 1052:

Pedestris exercitus ... spissam terribilemque fecit testudinem.

This is an exact description of the host that faced the Normans, fourteen years later, on the hill of Battle. As William of Malmesbury describes it:

Pedites omnes cum bipennibus, conserta ante se scutorum testudine, impenetrabilem cuneum faciunt.[70]

“It is a pleasure,” as I wrote, “to find myself here in complete agreement with Mr. Freeman.”[71] Mr. Freeman saw in this passage “the array of the shield wall,”[72] and aptly compared Abbot Æthelred’s description of the English array at the Battle of the Standard: “Scutis scuta junguntur, lateribus latera conseruntur.”[73] With Mr. Oman also I was no less pleased to find myself in perfect agreement. I myself should speak, as he does, of the “tactics of the phalanx of axemen.”[74] It is particularly interesting to read in his latest work (p. 57), that at Zülpich (A.D. 612), according to Fredegarius:

So great was the press when the hostile masses [phalanges] met and strove against each other, that the bodies of the slain could not fall to the ground, but the dead stood upright wedged among the living.

For precisely the same phenomenon is described at the Battle of Hastings. William of Poitiers says of the English:

Ob nimiam densitatem eorum labi vix potuerunt interfecti.

And Bishop Guy:

Spiritibus nequeunt frustrata cadavera sterni,

Nec cedunt vivis corpora militibus.

Omne cadaver enim, vita licet evacuatum,

Stat velut illæsum, possidet atque locum.[75]

There is nothing strange in this parallel between Zülpich and Hastings, for Mr. Oman observes that:

In their weapons and their manner of fighting, the bands of Angles, Jutes, and Saxons who overran Britain were more nearly similar to the Franks than to the German tribes who wandered south.[76]

At Poictiers “the Franks fought, as they had done two hundred years before at Casilinum, in one solid mass,”[77] for their tactics were “to advance in a deep column or wedge.”[78] We have seen that the “column” of English axemen similarly fought, according to Mr. Oman, “arranged in a compact mass.”

Where the agreement is so complete, I need not labour the point further. In my ‘Feudal England’ (pp. 354–8), I showed that Mr. Archer’s views on the subject could not stand for a moment against those of Mr. Freeman and Mr. Oman, to which they were directly opposed.

In ‘Social England’—just as Mr. Freeman had written that both the English and the Danes stood as a “wedge of infantry forming a wall with their shields”[79]—Mr. Oman writes of their “wedge or column.” It is only in his later work that he suddenly shifts his ground, and flatly contradicts his own words:

The writer’s “of course” is delightful.

This contradiction of himself, however, is as nothing compared with that to which we are now coming.

In his first work Mr. Oman wrote under Mr. Freeman’s influence. The Normans, he held, at the Battle of Hastings, were confronted by “impregnable palisades.” Nine years later, in his second description of the battle, he substituted for these “impregnable palisades” an “impenetrable shield wall.”

The faithful réchauffé of his former narrative only renders the more significant Mr. Oman’s change of “impregnable palisades” to “impenetrable shield wall.” For what had happened in the meanwhile to account for this change being made? In July, 1892, there had appeared in the ‘Quarterly Review’ my well-known article on “Professor Freeman,” in which I had maintained that the English defence consisted, not of impregnable “palisades,” but only of an impenetrable “shield wall.” On the furious and famous controversy upon this topic which followed, it is quite unnecessary to dwell. Mr. Oman, we have seen himself adopted the view I had advanced, and not, I hasten to add, on this point alone, for with his whole description of the battle, as given in ‘Social England,’ I am in complete agreement. The “shield wall” he mentions twice.[82] Of “palisades,” intrenchments, or breastworks there is not a word.

And yet Mr. Oman, now, is not ashamed to write:

I fear that I must plead that I was never converted. This being so, Mr. Round cannot prove that I was.[83]

What is the explanation of Mr. Oman’s statement? Simply that he has again changed his view; and having first adopted that of Mr. Freeman, and then abandoned it to adopt my own, he now, in turn, abandons both, and advances a third (or fourth) at variance with both alike! His Norman knights are still “surging”; but they “surge” against an obstacle which has once more changed its character:

The knights, if unsupported by the bowmen, might have surged for ever against the impregnable breastworks. The archers, unsupported by the knights, could easily have been driven off the field by a general charge. United by the skilful hand of William, they were invincible (‘History of the Art of War,’ p. 164).

What then were these “impregnable breastworks” which now make their appearance in our old familiar passage? They are described on page 154, where we read that “we must not think ... of massive palisading:[84] they were merely

wattled hurdles ... intended, perhaps, more as a cover against missiles than as a solid protection against the horsemen, for they can have been but hastily constructed things, put together in a few hours by wearied men.”

Let us place, side by side, Mr. Oman’s own words in this his latest work:

That they were, to say the least, “not wholly impassable” is evident from the writer’s own description (p. 159) of the Norman knights’ first charge “against the long front of the breastworks, which, in many places, they must have swept down by their mere impetus.” Nay, “before the two armies met hand to hand,” as Mr. Freeman observes,[85] a single horseman—“a minstrel named Taillefer,” as Mr. Oman terms him—“burst right through the breastwork and into the English line” (p. 158).[86] Such, on Mr. Oman’s own showing, were his so-called “impregnable breastworks” (p. 164). A single horseman could ride through them!

We see then that, in this his latest work, he not only adopts yet another view, but cannot adopt it consistently even when he does.

To me there is nothing strange in all this shift and shuffle. It has distinguished each of my opponents on this subject from the first. Not only are they all at variance with one another: they are also at variance with themselves. Alone my own theory remains unchanged throughout. The English faced their foes that day in “the close array of the shield wall.” Other defences they had none.

Mr. Oman has actually advanced four theories in succession:

(1) “The impregnable palisades.”[87]

(2) “The impenetrable shield wall.”[88]

(3) “An abattis of some sort.”[89]

(4) “Wattled hurdles.”[90]

The third of these made its appearance after his description in ‘Social England.’ “I still hold,” Mr. Oman wrote, “to the belief that there was an abattis of some sort in front of Harold’s line.”

But how can he “still” hold to a belief which he has never expressed before or since? For neither the first, second, or fourth of the defences he gives above can by any possibility describe an abattis. The New English Dictionary describes an abattis as

a defence constructed by placing felled trees lengthwise, one over the other, with their branches towards the enemy’s line.

The ‘Encyclopædia Britannica’ gives us a similar description, speaking of this defence as constructed of “felled trees lengthwise ... the stems inwards.”[91] One is driven to suppose that Mr. Oman is quite unable to understand what an abattis really is.

We have now seen that the writer has actually given in succession four entirely different descriptions of the defences of the English front, while he has not the candour to confess that he has ever changed his mind.

At this I am not in the least surprised. As I have observed in ‘Feudal England,’ p. 342:

As for the defenders of the ‘palisade,’ they cannot even agree among themselves as to what it really was. Mr. Archer produces a new explanation only to throw it over almost as soon as it is produced. One seeks to know for certain what one is expected to deal with; but, so far as it is possible to learn, nobody can tell one. There is only a succession of dissolving views, and one is left to deal with a nebulous hypothesis.

Even since these words were published, Mr. Oman has produced his fourth explanation, and has produced it in conjunction with Mr. Archer, who had previously enriched this series of explanations by two further ones of his own. In one of them the “fenestres,” which Wace makes the principal ingredient of the palisade, are rendered by Mr. Archer “windows.”[92] In another he describes the English defence as “a structure of interwoven shields and stakes,” “shields set in the ground and supported by a palisade of stakes,” a defence into which “actual shields have been built.”[93] It is only necessary to add that Mr. Oman, who acknowledges here his “indebtedness to Mr. T. A. Archer,”[94] tacitly, but absolutely, rejects both these phantasies, together with Mr. Archer’s great theory that the English axemen were “shieldless” at the battle,[95] and “could not or did not form the shield wall.”[96] All this Mr. Oman rejects, though, of course, he is careful not to say so; just as Mr. Archer, before him, had rejected views of Mr. Freeman, while professing to defend his account of the battle against me.[97]

I have now shown that my opponents are still as unable as ever to agree among themselves on the subject of the alleged English defence, and that as to Mr. Oman, he contradicts himself, not only in successive works, but even in a single chapter. A little clique of Oxford historians, mortified at my crushing exposé of Mr. Freeman’s vaunted accuracy, have endeavoured, without scruple, and with almost unconcealed anger, to silence me at any cost. And they cannot even wait until they have agreed among themselves.

How entirely impotent they are to stay the progress of the truth is shown by the fact that a German writer, Dr. Spatz, who has independently examined the authorities and the ground, goes even farther than myself in rejecting Mr. Freeman’s narrative, and especially the palisade.[98] Sir James Ramsay also, on similarly independent investigation, has been driven to the same conclusion, which his recently published work embodies. Does Mr. Oman refer to Dr. Spatz, whose work is a well-known one? No, he coolly states that “the whole balance of learned opinion” is against me on this matter,[99] although, as we have seen, neither he nor Mr. Archer accepts Mr. Freeman’s narrative,[100] while their own recorded views hopelessly differ (see pp. 43, 49).

Again, Mr. Oman writes:

I do not see what should have induced him [Wace] to bring the wattled barrier into his narrative, unless it existed in the tale of the fight as it had been told him, etc. (p. 153).

And yet he made use of my ‘Feudal England,’ in which I set forth prominently (pp. 409–416), as I had previously done in the ‘English Historical Review’ (viii. 677 et seq.; ix. 237), my theory that the passage in Wace “is nothing but a metrical, elaborate, and somewhat confused paraphrase of the words of William of Malmesbury,” and that he was clearly misled by the words “conserta ... testudine,” which he did not understand. Mr. Archer discussed this theory, but did not venture to reject it (Ibid.). Mr. Oman finds it safer to ignore it, and to profess that he cannot imagine where Wace got the idea from, except from oral tradition.

It is the same with the arrangement of the English host. In his latest work, Mr. Oman states, as a matter of fact, that the “house carles” formed the centre, and that

the fyrd, divided no doubt according to its shires, was ranged on either flank (p. 155).

There is no authority whatever for this view in any account of the battle, and it is wholly at variance with Mr. Oman’s own view, as stated in his earlier works.

In perfect agreement with these passages, I hold that “the well-armed house carles,” as Mr. Oman terms them, formed the English front, and were “backed” by the rest of the host.[101] Mr. Oman’s later view involves a tactical absurdity, as I have maintained throughout.[102] But here again Mr. Oman finds it the safest plan to ignore an argument he cannot face.

Let me, however, part from his narrative of the great struggle with an expression of honest satisfaction that, even in his latest work, he treats “the English host” as ranged “in one great solid mass” (p. 154). This is the essential point on which I have insisted throughout.[103] “No feature of the great battle is more absolutely beyond dispute”;[104] and it absolutely cuts the ground from under Mr. Archer’s feet.[105]

I may add that the denseness of the English host is similarly grasped by Sir James Ramsay, who has made an independent examination of the battle, and has set forth his interesting and original conclusions in his recently-published ‘Foundations of England.’ The ground plan of the battle in his work should be carefully compared with that which is found in Mr. Freeman’s History. For the two differ so hopelessly that the wholly conjectural character of Mr. Freeman’s views on the matter will at once be vividly shown. The bold conclusion of Sir James Ramsay that the English host held only the little plateau at the summit of the Battle hill, is at least in harmony with their dense array, and is very possibly correct.[106]

I now turn from battles to castles—those castles which played so prominent a part in Anglo-Norman warfare.

Let us first glance at the moated mound, and then at the rectangular keep. I do not desire, on the moated mound, to commit myself to all Mr. Clark’s views; but practical archæologists, I need scarcely say, are aware that the outer works of these most interesting strongholds were normally of horseshoe or crescent form, the mound being “placed on one side of an appended area.”[107] Mr. Oman, while acknowledging in his book, and in the columns of the ‘Athenæum,’ his indebtedness to Mr. Clark’s “admirable account of the topographical details of English castles,” describes the old English burhs as “stake and foss in concentric rings enclosing water-girt mounds” (p. 111). I pointed out in the ‘Athenæum’[108] that “Mr. Clark, who did more than any one for our knowledge of these burhs, was careful to explain,” in his plans,[109] that their outer defences were not concentric, as Mr. Oman asserts.

Determined never to admit a mistake, Mr. Oman retorted:

Of course, I am quite aware that in many burhs the outer works are not purely concentric; but the concentric form is the more typical. An admirable example of such a stronghold may be seen on p. 21 of Mr. Clark’s book, where he gives the plan of Edward’s burh of Towcester built in 921.[110]

Yet, in dealing with the Norman shell keeps on these “old palisaded mounds,” Mr. Oman actually, in his own book, admits, of their “outer defences,” that

as a general rule, the keep lies not in the middle of the space, but at one end of it, or set in the walls ... as a general rule the keep stands at one end of the enclosed space, not in its midst.[111]

This is the feature of these striking works for which I myself contended, and which, on that account, Mr. Oman at once denied.

As to the Towcester burh, I will place side by side my criticism and Mr. Oman’s reply:

Any archæologist comparing the plans will see at once that my statement is correct, and that the plan (compare the section) shows absolutely nothing beyond the actual ditch of the mound. I offered to submit the question to Mr. St. John Hope’s decision,[114] but Mr. Oman would submit it to no one but his friend and coadjutor, Mr. York Powell, who is not known as an authority on these works, and who is hostile to myself because I exposed Mr. Freeman![115]

Having now shown that, in his own words, Mr. Oman “cannot read the simplest military sketch,” I pass to the siege of Rochester Castle, famous for its rectangular keep, in 1264. This was an event that deserves attention in a ‘History of the Art of War,’ for John had breached the keep by mining half a century before, and the stately structure had now to stand an energetic siege at the hands of Simon de Montfort. A striking passage in Rishanger’s Chronicle tells us that, advancing from London,

comes autem de Leycestria, vir in omnibus circumspectus, machinas et alia ad expugnationem castri necessaria secum a civitate Londoniarum per aquam et per terram transvehi præcepit, quibus inclusos vehementer impugnavit, nec eos indulgere quieti permisit; exemplum relinquens Anglicis qualiter circa castrorum assultationes agendum sit qui penitus hujusmodi diebus illis fuerant ignari.[116]

The barons promptly stormed the ‘outer bailey’ of the castle (April 19),[117] and strove desperately to gain the keep, till, a week later, they fled suddenly at the news of the king’s advance on London.[118] But so vigorous were the siege operations by attack, battery, and mining, that they were on the point of succeeding when they had to raise the siege.[119]

Surely a ‘History of the Art of War’ should mention the above remarkable allusion to Simon’s mastery of siege operations, and to his teaching the English, who were then ignorant of the subject. But all that Mr. Oman tells us is that—

the massive strength of Gundulf’s Norman keep was too much for such siege appliances as the earl could employ. The garrison under John de Warenne, the Earl of Surrey, held their own without difficulty (p. 416).

We have seen that, on the contrary, the keep was on the point of being taken. But what are we to say to the words, “Gundulf’s Norman keep”? “It was long the custom,” as Mr. Clark wrote, “to attribute this keep to Gundulf, making it contemporaneous, or nearly so, with the Tower of London”; but, more than thirty years ago, it was shown by Mr. Hartshorne (in the ‘Archæological Journal’) that it was built in later days under William of Corbeuil (1126–1136).[120] No one, in the present state of our knowledge, could suppose that Gundulf was its builder; and it is obvious that a writer who does must have yet everything to learn on Norman military architecture.

I must lastly deal as briefly as possible with the subject of knight service. The view of modern historians has been that this was gradually evolved during the Norman period out of a pre-conquestual obligation to provide one armed man for every five hides held. As against this I have advanced the theory[121] that the whole arrangement was introduced de novo at the Conquest, when the Conqueror assessed the fiefs he granted in terms of the five-knight unit irrespective of hidation. Put in a less technical form my theory is that the Conqueror called on the holder of every considerable fief to furnish a contingent of five knights, or some multiple of five, to the feudal host.[122] And this he did arbitrarily, without reckoning the ‘hides’ that might be contained in the fief. Further, by the argumentum ad absurdum, I showed that if every five hides had to provide a knight, there would be nothing, or less than nothing, left for the tenant-in-chief.[123] It was of this new theory that Professors Pollock and Maitland observe, in their history of English Law (i. 238–9), that they regard it “as having been proved by Mr. Round’s convincing papers.”

Mr. Oman, however, leans to the now exploded theory, and holds that under Norman rule “the old notion that the five hides must provide a fully armed man was remembered;”[124] and that though “some lay tenants-in-chief” got off easily, “the majority were obliged to supply their proper contingent.”[125] He then proceeds:

It has been clearly shown of late, by an eminent inquirer into early English antiquities, that the hidage of the townships was very roughly assessed, and that the compilers of Domesday Book incline towards round numbers.

Now apart from the fact that this “eminent inquirer,” my friend Professor Maitland to wit, gives me full credit for having been first in the field[126]—a fact which Mr. Oman, with my book before him, of course carefully ignores—his words show that he cannot understand the simplest historical theory. Professor Maitland and I have dwelt on the antiquity of this assessment, with which “the compilers of Domesday Book” had no more to do than Mr. Oman himself, and which indeed the compilation of that book has almost utterly obscured.

From the fact of the five-hide unit Mr. Oman argues “that there was little difficulty in apportioning the military service due from the tenants-in-chief who owned them,”[127] though such apportionment, as I have shown, would result in an actual absurdity.[128] Indeed, Mr. Oman himself observes that the tenant-in-chief, to discharge his obligation, “might distribute the bulk of his estate in lots roughly averaging five hides to subtenants, who would discharge the service for him,”[129] although a moment’s consideration will show that this process would absorb not “the bulk,” but the whole of his estate.

But all this is insignificant by the side of Mr. Oman’s double error on the vetus feoffamentum. This begins on p. 359, which is headed “The old enfeoffment,’” and which describes the distribution of fiefs by William among the tenants-in-chief. On the next page he writes of “the knights of ‘the old enfeoffment,’ as William’s arrangement was entitled,” and proceeds to vouch my ‘Feudal England’ as his authority for this statement! On the same page we read of the landholder’s “servitium debitum according to the assessment of the vetus feoffamentum of the Conqueror”; and further learn that Henry II.

demanded a statement as to the number of knights whom each tenant-in-chief owed as subtenants, how many were under the ‘old enfeoffment’ of William I., and how many of more recent establishment.

We also read that—

the importance of King Henry’s inquest of 1166 was twofold. It not only gave him the information that he required as to the proper maintenance of the debitum servitium due under the ‘old enfeoffment’ of the Conqueror, but showed him how many more knights had been planted out (sic) since that assessment (p. 363).

Again, on page 364 we read of “the ‘old enfeoffment’ of the eleventh century,” and the phrase (which Mr. Oman quite properly places within quotation marks) occurs in at least three other passages.

It is quite evident that Mr. Oman imagines the vetus feoffamentum to be (1) the original distribution by the Conqueror (2) among the tenants-in-chief. Both ideas are absolutely wrong. For (1) it had nothing to do with “William’s arrangement”—which determined the servitium debitum, a very different matter; and (2) it referred to the sub-enfeoffment of knights by tenants-in-chief. The dividing line between the “old” and the “new” feoffments, was the death of Henry I. in 1135. All fees existing at that date were of the antiquum feoffamentum; all fees created subsequently were of the novum feoffamentum. This essential date is nowhere given by Mr. Oman, who evidently imagined that the latter were those “of more recent establishment” than “the old enfeoffment of William I.”

The frightful confusion into which Mr. Oman has been led by his double blunder is shown by his own selected instance, the carta of Roger de Berkeley in 1166. According to him, “Roger de Berkeley owed (sic) two knights and a half on the old enfeoffment.”[130] Two distinct things are here hopelessly confused.

(1) Roger “owed” a servitium debitum (not of 2½, but) of 7½ knights to the Crown; and his fief paid scutage[131] accordingly in 1168, 1172, and 1190.

(2) Roger “has” two and a half knights enfeoffed under the old feoffment[132] (that is, whose fiefs existed in 1135), the balance of his servitium debitum being, therefore, chargeable on his demesne,[133] as no knights had been enfeoffed since 1135.

It is difficult to understand how the writer can have erred so grievously, for it was fully recognised by Dr. Stubbs and by myself (‘Feudal England,’ pp. 237–239) that 1135 was the dividing point.[134] It may be as well to impress on antiquaries that fees “de antiquo feoffamento” were fees which had been in existence in 1135, at the death of Henry I., just as tenures, in Domesday Book, ‘T.R.E.,’ were those which had existed in 1066, at the death of Edward; for with these two formulas they will frequently meet. It is the “servitium debitum,” not the “antiquum feoffamentum,” which “runs back,” as Mr. Oman expresses it, to the Conquest.

The result of his confusion is that his account of the origins (in England) of knight service is not only gravely erroneous, but curiously topsy-turvy. This is scarcely wonderful when we find on page 365 that he is hopelessly confused about knights and serjeants, not having grasped the elementary distinction between tenure by serjeanty and tenure by knight service. From what I have seen of the author’s account of the battle of Bannockburn, his errors, I imagine, are by no means restricted to the subjects I have here discussed. A curious combination of confidence and unwillingness to admit his mistakes, with a haste or confusion of thought that leads him into grievous error, is responsible, it would seem, for those misconceptions which render untrustworthy, as it stands, his ‘History of the Art of War.’

IV
The Origin of the Exchequer

Historians have rivalled one another in their witness to the extraordinary interest and importance of the twelfth-century Exchequer. “The whole framework of society,” writes the Bishop of Oxford, “may be said to have passed annually under its review.... The regular action of the central power of the kingdom becomes known to us first in the proceedings of the Exchequer.” Gneist insists on “its paramount importance” while “finance is the centre of all government”; and in her brilliant monograph on Henry the Second, Mrs. Green asserts “that the study of the Exchequer is in effect the key to English history at this time.... It was the fount of English law and English freedom.” One can, therefore, understand Mr. Hall’s enthusiasm for “the most characteristic of all our national institutions ... the stock from which the several branches of the administration originally sprang.” Nor does this study appeal to us only on account of its importance. A glamour, picturesque, sentimental it may be, and yet dazzling in its splendour, surrounds an institution possessing so immemorial an antiquity that “Barons of the Exchequer” meet us alike in the days of our Norman kings and in those of Queen Victoria. Its “tellers,” at least coeval with the Conquest, were only finally abolished some sixty years ago, while the Chancellor of the Exchequer is believed to represent that “clericus cancellarii” whose seat at the Exchequer of the second Henry was close to that of the official ancestor of the present secretary to the Treasury. Yet, older than these, older even than the very name of the Exchequer, was its wondrous system of wooden tallies, that hieroglyphic method of account which carries us back to a distant past, but which, Sir John Lubbock has observed, was “actually in use at the Exchequer until the year 1824.” Of all survivals of an archaic age this was, probably, the most marvellous; it is not easy to realize that even in the present century English officials were keeping their accounts with pieces of wood which “had attained the dimensions, and presented somewhat the appearance, of one of the wooden swords of the South Sea Islanders.” It was an almost tragic feature in the passing of “the old order” that when these antique relics were finally committed to the flames, there perished, in the conflagration said to have been thus caused, that Palace of Parliament which, like themselves, had lingered on to witness the birth of the era of Reform.

But what, it may be asked, was the Exchequer, and why was it so named? The earliest answer, it would seem, is that of William Fitz Stephen, who, in his biography of Becket, tells us that, in 1164, John the Marshal was in London, officially engaged “at the quadrangular table, which, from its counters (calculis) of two colours, is commonly called the Exchequer (scaccarium), but which is rather the king’s table for white money (nummis albicoloribus), where also are held the king’s pleas of the Crown.”[135] The passage is not particularly clear, but I quote it because it is not, I believe, mentioned by Mr. Hall,[136] and because William Fitz Stephen knew his London well. The questions I have asked above are those which avowedly are answered in the first chapter of the famous ‘Dialogus de Scaccario’ (circ. 1178). I need not, however, repeat in detail the explanations there given, for they should be familiar from the works of Dr. Stubbs and of every writer on the subject. Suffice it to say that while, in shape, the ‘Exchequer’, with its ledge, as Mr. Hall observes, was not unlike a billiard table, “it derived its name from the chequered cloth” which, says Dr. Stubbs, covered it, and which gave it a resemblance to a chess board (scaccarium). Antiquaries have questioned this, as they will question everything; but the fact remains that the symbol of the Exchequer, of which types have been depicted by Mr. Hall, is that which swings and creaks before the wayside ‘chequers,’ which once, in azure and gold, blazed upon the hill of Lewes, and which still is proudly quartered by the Earl Marshal of England.

In the present paper I propose to consider the origin and development of the institution, and to examine critically some of the statements in the famous ‘Dialogus de Scaccario,’ of which the authority has hitherto been accepted almost without question.

It is alleged that a cruel hoax was perpetrated on the Royal Society by that ‘merry monarch’ Charles II., who called on its members to account for a phenomenon which existed only in his own imagination. Antiquaries and historians have, with similar success, been hoaxed by Richard the son of Nigel, who stated as a fact in his ‘Dialogue on the Exchequer,’ that there is no mention of a ‘blanch’ ferm to be found in Domesday Book. Richard proceeded to infer from this that those who spoke of ‘blanch’ ferm existing before the Conquest must be mistaken.[137]

Dr. Stubbs actually accepts the statement that “the blanch-ferm is not mentioned in Domesday,” but declares that Stapleton, in his well-known argument,[138] has clearly shown it to have had “its origin in a state of things that did not exist in Normandy, and was ‘consequent upon the monetary system of the Anglo-Saxons.’ The argument,” he writes, “is very technical, but quite conclusive.” Sir James Ramsay also, though writing as a specialist on finance, contents himself with citing Stapleton, through Stubbs, and with adding a reference to “white silver” in the Laws of Ælfred,[139] and ignores the evidence in Domesday Book.

Now the index to the Government edition of Domesday is a very imperfect production, but we need travel no farther than its pages to discover that there is no difficulty to solve; for the “alba firma” is duly entered under an Isle of Wight manor (i. 39 b). Moreover, we read on the same folio of “lx solidos albos” and “xii libras blancas” in a way that suggests the identity of the two descriptions. But, further, we find, scattered over Domesday, ‘Libræ albæ,’ ‘blancæ,’ and ‘candidæ,’ together with ‘libræ de albis denariis’ or ‘de candidis denariis,’ and ‘libræ alborum nummorum’ or ‘candidorum nummorum.’ The ‘blanch’ system, therefore, was already quite familiar. This, however, is not all. On the folio mentioned above (i. 39 b) we read of another manor: “T. R. E. xxv lib. ad pensum et arsuram.” This can only refer to that payment in weighed and assayed money, the method of which is described in the ‘Dialogue’ under ‘Quid ad militem argentarium’ and ‘Quid ad fusorem’ (I. vi.). All this elaborate system, therefore, must have been already in operation before the Conquest.

But the ‘Dialogue’ asserts in its next and very remarkable chapter—“A quibus vel ad quid instituta fuerit argenti examinatio”—that this system was first introduced by the famous Roger, bishop of Salisbury, the writer’s great-uncle, after he had sat at the Exchequer for some years, and had discovered the need of introducing it.[140] Between this statement and the evidence of Domesday the contradiction is so absolute that a grave question at once arises as to the value of the writer’s assertions on the early Norman period. Like the men of his time, he revelled in texts, and loved to drag them in on every possible occasion. One is, therefore, only following his example in suggesting that his guiding principle was, “I magnify my office.” The greatness and the privileges of a seat at the Exchequer were ever present in his mind. But to this he added another principle, for which insufficient allowance, perhaps, has hitherto been made. And this was, ‘I magnify my house.’ Nor can one blame the worthy treasurer for dwelling on his family’s achievements and exalting his father and his great-uncle as the true pillars of the Exchequer. He was perfectly justified in doing this; but historians should have been on their guard when he claims for Bishop Roger the introduction of a system which Domesday Book shows us as already in general operation.[141]

Enlightened by this discovery, we can more hardily approach a statement by the writer in the same chapter, which has been very widely repeated. One need only mention its acceptance by such specialists as Stapleton, in his work on the Norman Exchequer, and Mr. Hubert Hall, who, in his work on the ‘Antiquities and Curiosities of the Exchequer,’ refers to it four times.[142] He first tells us that

for half a century after the Conquest there could have been very little need of a central treasury at all, since the greater part of these provisions formed an intrinsic portion of the revenue itself ... which was still payable in kind. This point is both important and interesting, and has been hitherto somewhat overlooked by economic writers. The fact (which is probable enough in itself) rests on high authority—that of the famous treasurer of the first two Plantagenet kings (p. 4).

Again, he writes on p. 161:

We have seen that in the earliest times—previously, that is, to the reorganization of the Exchequer under Henry I.—the revenue of the sovereign was answered in two forms, namely, in specie and in kind, the former drawn from judicial fines and farms of towns, and the latter rendered, at an arbitrary assessment, by the cultivators of the royal demensne.[143]

The passage itself in the ‘Dialogus,’ which Mr. Hall translates in extenso (pp. 180–182), requires careful examination. The “high authority” of which he speaks proves to be, in fact, only tradition, for the opening words of the passage run: “Sicut traditum est a patribus.” Now one would not strain unduly the words of the Dialogue’s author, but his meaning may be fairly understood to be that the rents of the royal demesne were not only paid in kind (for that he clearly asserts), but were also valued in kind alone. For he thus describes the change introduced under Henry I.:

Destinavit [rex] per regnum quos ad id prudentiores et discretiores cognoverat, qui circueuntes et oculata fide fundos singulos perlustrantes, habita æstimatione victualium, quæ de hiis solvebantur, redegerunt in summam denariorum.

This can only imply the substitution of a money valuation for a rent payable in kind. And yet we have to go no further than this very chapter to learn that these rents had previously been reckoned in money (not in kind). For if, as stated in the note below, they had, when they were paid in kind, to be reduced by the king’s officers to a money standard, it could only be because their amounts were due, not in kind, but in money.[144] Fortunately, however, we are not dependent on this obvious contradiction, for the evidence of Domesday makes it certain that, just as the assay was employed under the Conqueror, and indeed under the Confessor, instead of being first introduced under Henry I., so the valuation in money of the rents from the royal demesne was not a reform effected, as alleged, by the latter king, but was the rule under William I.; and, indeed, almost as much the rule before the Conquest.[145] We gather from Domesday that the Conqueror advanced the commutation of the old “firma unius diei,” etc., for a sum of money; but even under his predecessor there were only a few localities in which the archaic system had lingered on.

I have said something in ‘Feudal England’[146] of the “Firma unius noctis,” and I would now add to the evidence that I there adduced on this curious and interesting subject.

In Devonshire we meet with a singular feature, which, I think, has escaped attention. Exeter, we read, “reddit xviii. lib. per annum.” I have elsewhere[147] discussed this payment, and shown that it was strangely small; but I now proceed to a new point, namely, that the figure 18 may prove highly significant. Lidford, Barnstaple, and Totnes, we read,[148] “rendered” between them the same amount of (military) service as Exeter “rendered”; and this service was equally divided between them.[149] Now, if we turn from the service to the payments made by this group of boroughs, we find that the “render” of each was £3 a year, so that the whole group paid £9, exactly half the “render” of Exeter.[150]

If we follow the clue thus given us, and turn to the manors which Queen Edith and Harold’s mother and Harold himself had held, but which, in 1086, had passed to the king,[151] we find these remarkable figures: £15, £30, £45, £18, £48, £1½, £48 (formerly £23), £2, £6, £23 (formerly £18), £24, £3, £18, £3, £18, £12, £18, £24, £4 (?), £24, £1 (?), £7, £6, £6, £12, £8, £2, £3, £18, £20 (formerly £24). It is evident enough that these “renders” are based on some common unit, like the ‘renders’ of the comital manors in Somerset.[152] Moreover, we can trace, in Cornwall, something of the same kind. The manor of royal demesne which heads its survey “reddit xii lib. ad pondus et arsuram,”[153] and this is followed by renders of £8, £5, £6, £3 (‘olim’), £18, £6, £3, £7, £6, £6, £4, £5. Even a ‘render’ of £8 was duodecimal in a way; for on fo. 121 b it occurs four times as £8 and thrice as “xii markæ.”

Not only is the rent of these manors distinguished from that of those in private hands by the form ‘reddit,’ instead of ‘valet,’ but the render is stereotyped, being normally unchanged, while the ‘valet’ ever fluctuates. The explanation I suggest for these archaic “renders” is that they represent the commutation of some formerly existing payment in kind similar to the “firma unius noctis.” If the unit of that payment was commuted at a fixed rate, it would obviously produce that artificial uniformity of which we have seen the traces in Devon and Cornwall. We may thus penetrate behind these “renders” to an earlier system then extinct.

This conclusion is confirmed, I think, by some striking instances in Hampshire.[154] Of ‘Neteham’ we read, “T.R.E. et post valuit lxxvi lib. et xvi sol. et viii den.” (i. 38); and of ‘Brestone,’ similarly, “T.R.E. et post valuit lxxvi lib. et xvi sol. et viii den.” (i. 38 b). The explanation is found in these two entries on the latter fo.:

That is, I take it that the half-day’s ferm “rendered” T.R.E. was worth £38 8s. 4d., so that the two other manors, for each of which the sum was £76 16s. 8d., must originally have rendered a whole ‘firma.’ This gives us the value of the ‘firma’ for the other Hampshire manors which “rendered.”[155]

We will now return to the ‘Dialogus’ and its statements on the “firma comitatus.”

It is distinctly asserted, in the above passage, that the ‘firma comitatus’ only dated from this reform under Henry I.[156] This is at variance with the strong evidence set forth in my ‘Geoffrey de Mandeville,’ that Geoffrey’s grandfather, who was dead before this alleged reform, held Middlesex, Essex, and Herts at farm, the very amount of the farm due from him being mentioned. But, indeed, in Domesday itself there are hints, if not actual evidence, that the ‘firma’ was more or less in existence. In Warwickshire, for instance, “T.R.E. vicecomitatus de Warwic cum burgo et cum regalibus Maneriis reddebat lxv libras,” etc., etc. In Worcestershire, also, “vicecomes ... de Dominicis Maneriis regis reddit cxxiii lib. et iiii sol. ad pensum.” Here we have exactly that “summa summarum” of which the ‘Dialogus’ speaks as a novelty introduced under Henry I.[157] Again, in at least one passage (i. 85), we recognise a distinct allusion to the “terræ datæ” system:

De hoc Manerio tenet Giso episcopus unum membrum Wetmore quod ipse tenuit de rege E. Pro eo computat Willelmus vicecomes in firma regis xii lib. unoquoque anno.

Now we know the history of this manor, which had been detached from the royal demesne about a quarter of a century before, when Edward gave it to bishop Giso on his return from his visit to Rome. It follows, therefore, that £12 must have been, ever since, annually credited to the sheriff, in consideration of the Crown having alienated this manor.[158] We thus carry back to a period before the Conquest that Exchequer practice of the 12th century, which is thus alluded to in Stephen’s charter to Geoffrey earl of Essex (1141):

Ita tamen quod dominica quæ de prædictis comitatibus data sunt ... a firma prædicta subtrahantur et ... ad scaccarium computabuntur.[159]

I hasten to add that the Charter of Constance, the Conqueror’s daughter, quoted by Stapleton from the Cartulary of Holy Trinity, Caen, affords an exact parallel in the words: “et ei computabitur in suo redditu cum dica.” But the fact remains that we can prove the existence, under Edward the Confessor, of characteristic features of the later Exchequer system, of which one, at least, as Stapleton explained, must have been of English origin.

What then was the change that took place on the introduction of the Exchequer? How did it modify the system previously in existence? Our only clue is found in the well-known words of the ‘Dialogus’: “Quod autem hodie dicitur ad scaccarium, olim dicebatur ad taleas.” Writing as a specialist on Exchequer history, Mr. Hall contends that “this expression in itself denotes the actual place of receipt and issue of the revenue rather than a court or council chamber.”[160] But one cannot see that ‘scaccarium’ in itself denotes a court or council chamber more than does ‘talea.’ The one was a chequered table, the other a wooden tally. My own view is that the change really consisted of the introduction of the chequered table[161] to assist the balancing of the accounts. Previously, tallies alone would be used, and it is noteworthy that even after the ‘Exchequer’ system was in full operation, the deduction for the loss involved by ‘combustion’ was still effected by tally.[162] I have little doubt that the ‘combustion’ tally was in use in the 11th century for payments “ad arsuram et pensum.”

Instead, then, of the sheriffs’ accounts being balanced by the cumbrous system of tallies, the introduction of the Exchequer table, very possibly under Henry I., enabled them to be depicted to the eye by an ingenious system of counters. To the modern mind it is strange, of course, that, while the reformers were about it, they did not substitute parchment, and work out the accounts on it. But, doubtless for the benefit of unlearned sheriffs, the old system of ocular demonstration was still adhered to, and the Treasurer’s Roll merely recorded the results of the ‘game’ by which the accounts had been worked out upon the table.

Mr. Hall’s belief is best set forth in an article he contributed to the ‘Athenæum’ (November 27, 1886), and of which he reprinted this passage, subsequently, in ‘Domesday Studies’ (1891):

There is every reason for believing that the audit machinery of the ancient Treasury at Winchester was sufficient for the purpose.... It is true, indeed, that the earliest germ of the Exchequer is perceptible in these accounts, which were, however, audited not ‘ad scaccarium,’ but ‘ad taleas,’ i.e. in the Treasury or Receipt at Winchester.... We find in the Pipe Rolls the old Treasury at Winchester used as a permanent storehouse for the reserve of treasure, regalia, and records, and we even find Exchequer business transacted there by way of audit of accounts, which formed a special office or ‘ministerium’ as late as 1130 (Pipe Roll 31 Hen. I).[163]

The purchase of the ‘ministerium thesauri Wintoniæ,’ recorded in the Pipe Roll of 1130,[164] does not affect the question of audit. There can be no question that the national Treasury, in 1130, was at Winchester, or that the Treasurer’s official residence was there also.[165] The really important passages on the roll, passages which I venture to think have been generally misunderstood, are these:

Et in præterito anno quando comes Gloecestriæ et Brientius filius Comitis audierunt compotum de thesauro apud Wintoniam.

De istis habuit Willelmus de Pontearc’ xxx li., de quibus reddidit compotum quando comes Gloecestriæ et Brientius audierunt compotum de thesauro apud Wintoniam.

It has been assumed that these entries refer to the Exchequer business of balancing the sheriffs’ accounts, and Madox even went so far as to draw the conclusion, from their wording, that, at the time of the Roll, Brian Fitz Count was Treasurer. The true meaning was exactly contrary, and an interesting allusion is thus obscured.

For the Pipe Rolls do not, as is sometimes imagined, display the national accounts. They probably do not exhaust the receipts (for some, it is believed, were paid ‘in camera’), and they certainly only record a portion of the royal expenditure. What became of the money which is so continually entered as paid ‘in Thesauro’? It found its way into the national treasury, whence it was paid out as was required by writ of ‘Liberate’ addressed to the Treasurer and chamberlains.[166] Of these outgoings, in the 12th century, there is, it would seem, no record; but they were certainly audited from time to time, the king calling on the Treasurer to account for the money in his charge, as, at the Exchequer, the Treasurer himself had called on the sheriffs to account for the sums for which they were liable. To this ‘generalis compotus,’ associated with the Winchester Treasury, there are, in the ‘Dialogus,’ several allusions which may have been somewhat overlooked.

Quod thesaurarius a vicecomite compotum suscipiat, hinc manifestum est, quod idem ab eo cum regi placuerit requiritur.... Sunt tamen qui dicunt thesaurarium et camerarios obnoxios tantum hiis quæ scribuntur in rotulis ‘in thesauro,’ ut de hiis compotus ab eis exigatur (i. 1).

Raro inquam, hoc est, cum a rege, vel mandato regis, a magnis regni[167] compotus a thesaurario et camerariis regni totius recepta suscipitur (i. 5).

Thesaurarius et camerarii, nisi regis expresso mandato vel præsidentis justiciarii, susceptam pecuniam non expendunt: oportet enim ut habeant auctoritatem rescripti regis de distributa pecunia, cum ab eis compotus generalis exigitur (i. 6).

[De combustione] ... ut de summa ejus thesaurarius et camerarii respondeant (ib.).

These are sufficient allusions to the Treasury, as distinct from the Exchequer, account. I invite particular attention to this Treasury audit, because, so far as I can find, it has hitherto escaped notice. The second extract refers to the use of the £10,000 space on the chequered table, and therefore proves the use of such a table for the Treasury account as well.

Now my point is that the earl of Gloucester and Brian ‘Fitz Count,’ in 1130, were magnates (magni regni) delegated by the king, as described in the second passage,[168] to audit the Treasurer’s account. And this view is confirmed by the fact that William de Pont de l’Arche, who here accounts to them, is styled by Dr. Stubbs “the Treasurer,” and is, in any case, subsequently described as “custos thesaurorum regalium.” Their mission had nothing, I hold, to do with that audit of the sheriffs’ accounts, which was the annual function of the Exchequer.

There is a remarkable entry on the roll of 1187 which alludes to an overhauling of the national treasure at Winchester, at the beginning of that year, the date proving that it was wholly unconnected with either session of the Exchequer:

Et in custamento numerandi et ponderandi thesaurum apud Wintoniam post Natale, et pro forulis novis ad reponendum eundem thesaurum et pro aliis minutis negociis ad predictum opus, etc.... Et pro carriando thesauro a Wintoniâ ad Saresburiam et ad Oxinford’ et ad Geldeford’ et ad plura loca per Angliam £4 8s. 3d.

One might compare with these phrases the ‘Dialogus’ language as to the knights, ‘qui et camerarii dicuntur, quod pro camerariis ministrant.’

Item officium horum est numeratam pecuniam, et in vasis ligneis per centenos solidos compositam, ponderare, ne sit error in numero, tunc demum in forulos mittere, etc. (i. 3).

Also the description of the usher’s office:

Hic ministrat forulos ad pecuniam reponendam, etc. (ib.).

But the latter part of the entry (which is duly quoted by Eyton[169]) is also of much importance. For in Mr. Hall’s work, under 1187, we only read, ‘Treasure conveyed abroad from Winchester.’[170]

It is an essential part of Mr. Hall’s theory, which makes the “Westminster Treasury ... the principal Treasury of the kingdom,”[171] that the Winchester Treasury was merely “an emporium in connection with the transport of bullion (and especially of the regalia and plate), as well as other supplies, viâ Southampton, or other seaports, to the Continent.”[172] But the above passage shows us, on the contrary, treasure sent thence to Salisbury, Oxford, and Guildford. It is manifest that treasure, despatched from Westminster to Oxford or Guildford would not be sent viâ Winchester. From this it follows that Winchester was still a central Treasury, and not a mere ‘emporium’ en route to the south. It is certain that under Henry I., some sixty years before, the session at Westminster of the Barons of the Exchequer did not, as Stapleton observed, affect the position of the national Treasury at Winchester. It is, then, equally certain that the money received at that session must have been duly transmitted to the Winchester Treasury. For that was where the treasure (in coined money) was kept when Stephen succeeded at the close of 1135.

The whole difficulty has arisen from Mr. Hall’s inability to distinguish between the ‘Receipt’ at Westminster, where the money was paid in, and the national Treasury at Winchester in which it was permanently stored. This is, roughly speaking, like confusing a man’s investments with his balance at his bankers. The steadily growing importance of Westminster and the concurrent decadence of Winchester led, of course, eventually, to the shifting of the central Treasury, but at the time of the ‘Dialogus,’ in the days of Henry II., it is clear that the Exchequer was not looked on as the seat of a permanent Treasury. For the storage of treasure is always implied by the payment for the light of the night watchman; and as to the watchman and his light, the evidence of the ‘Dialogue’ is clear:

Vigilis officium idem est ibi quod alibi; diligentissima scilicet de nocte custodia, thesauri principaliter, et omnium eorum quæ in domo thesauri reponuntur.... Sunt et hiis liberationes constitutæ dum scaccarium est, hoc est a die qua convocantur usque ad diem qua generalis secessio.... Vigil unum denarium. Ad lumen cujusque noctis circa thesaurum, obolum (i. 3).

There is absolutely no escaping from these words: a watchman is only provided for the treasure “while the Exchequer is in session”; its treasury is temporary, not permanent. The whole passage, as it seems to me, is absolutely destructive of Mr. Hall’s hypothesis of “the existence of a permanent financial staff under the Treasurer and chamberlains of the Exchequer at Westminster.”[173]

The change from the “Treasury” to the “Exchequer” was, I hold, a gradual process. Careful study of the annual revenues bestowed by our sovereigns on the foreign houses of Tiron, Fontevrault, and Cluny[174] proves clearly how insensibly the “Treasury at Winchester” was superseded by the “Exchequer at London” as the place of payment. This is especially the case with Tiron, where Henry I.’s original grant, made about the middle of his reign, provides for payment “de thesauro meo, in festo Sancti Michaelis, Wintonie.”[175] Under Richard I. this becomes payable “at Michaelmas from his exchequer at London.”[176] Documents between the two show us intermediate stages.

Precisely the same gradual process is seen in the parallel development of the chamberlainship of the “Exchequer” from that of the “Treasury.” Just as Henry II., shortly before his accession, confirmed the grant to Tiron as “de thesauro Wintonie,”[177] so he restored to William Mauduit, at about the same time, “camerariam meam thesauri,” which office was held by his descendants as a chamberlainship of the Exchequer.

The ‘Dialogus’ shows us the Treasurer and the two chamberlains of the Exchequer as the three inseparable Treasury officers. Domesday connects the first with Winchester by showing us Henry “thesaurarius” as a tenant-in-chief in Hampshire. I propose to show that it also connects one of the chamberlains with that county. In that same invaluable but unprinted charter of which I have spoken above, which was granted at Leicester (1153) to William Mauduit, Duke Henry says:

Insuper etiam reddidi eidem camerariam meam thesauri cum liberatione[178] et cum omnibus pertinentibus, castellum scilicet de Porcestra ut supradiximus, et omnes terras ad predictum camerariam et ad predictum castellum pertinentes, sive sint in Anglia sive Normannia, sicut pater suus illam camerariam cum pertinentibus melius habuit et sicut Robertus Maledoctus frater suus eam habebat die quo vivus fuit et mortuus.

This carries back the ‘cameraria thesauri’ (‘illam camerariam’) to the Domesday tenant, whose son Robert occurs in the earlier Winchester Survey, and, though dead in 1130, is mentioned on the Roll of that year (p. 37), in connection with the Treasury in Normandy.

The history of Porchester, in the Norman period, has yet to be worked out. Mr. Clark, for instance, tells us that the castle was “always in the hands of the Crown,”[179] yet we find it here appurtenant to the chamberlainship, and in Domesday (47 b) it was a ‘manor’ held by William Malduith. The above charter, in my opinion, was one of those which Duke Henry granted without intending to fulfil.[180] Porchester had clearly been secured by the Crown, and Henry was not the man to part with such a fortress. Of William Mauduith’s Domesday fief, Hartley Mauditt (‘Herlege’) also was held by the later Mauduits; but they held it still “per serjanteriam camar[ariæ] Domini Regis”[181] or “per camerariam ad scaccarium.”[182]

It should be added that the other chamberlainship of the Exchequer was similarly a serjeanty associated with land. It cannot, however, be carried back beyond 1156, when Henry II. bestowed on Warin Fitz Gerold, chamberlain, lands in Wiltshire worth £34 a year, and in Berkshire to nearly the same amount.[183] The former was the chamberlainship estate, and reappears as Sevenhampton (near Highworth) in his brother’s carta (1166), where it is expressly stated to have been given to Warin by the king.[184] It was similarly held by his heir and namesake (with whom he is often confused), under John,[185] and by the latter’s heir, Margaret ‘de Ripariis,’ under Henry III.[186]

This estate must not be confused with that of Stratton, Wilts, which was bestowed by John (to whom it had escheated) on the later Warin Fitz Gerold, to hold at a fee-farm rent of £13 a year.[187] It is necessary to make this distinction, because Mr. Hall, in dealing with the subject, speaks of it as “held apparently by the Countess of Albemarle as pertaining to the (sic) chamberlainship of England” (sic).[188] On the same page he speaks of a deed, on page 1024 of the same volume, whereby she “secures to Adam de Strattone, clerk, an annuity of £13, charged on the farm of Stratton.” Reference to page 1024 shows that, on the contrary, what she did was to make herself and her heirs responsible to the Exchequer for the annual £13, which was “the farm” of Stratton (so that Adam might hold Stratton quit therefrom). This is a further instance of Mr. Hall’s unhappy inability to understand or describe accurately the documents with which he deals.[189]

I have now traced for the first time, so far as I can find, the origin of the two chamberlainships of the Exchequer. That of Mauduit can be traced, we see, to a chamberlainship of the ‘Treasury,’ existing certainly under Henry I., and possibly under the Conqueror. Of the other the existence is not proved before 1156. Both, I have shown, were associated with the tenure of certain estates.

It is very strange that, in his magnum opus,[190] Madox not only ignores, it would seem, this descent of the office with certain lands, but gives a most unsatisfactory account of those who held the office, confusing it, clearly, with the chamberlainship of England, and not distinguishing or tracing its holders.

For the different standards of payment in use at the Exchequer, our authority, of course, is the ‘Dialogus,’ but the subject, I venture to think, is still exceedingly obscure. Even Mr. Hall, who has studied so closely the ‘Dialogus,’ seems to leave it rather doubtful whether payment in ‘blank’ money meant a deduction of 6d. or of 12d. on the pound.[191] It will be best to leave the ‘Dialogus’ for the moment, and take an actual case where the charters and the rolls can be compared, and a definite result obtained.

In Lans. MS. 114, at fo. 55, there is a series of extracts transcribed from a Register of Holy Trinity (or Christchurch) Priory, London, in which are comprised the royal charters relating to Queen Maud’s gift of two-thirds of the revenues (ferm) of Exeter. First, Henry I. confirms it, late in his reign,[192] as “xxv libras ad scalam,” the charter being addressed to William bishop of Exeter, and Baldwin the sheriff (sic). Then we have another charter from him addressed “Rogero episcopo Sar[esbiriensi] et Baronibus Scaccarii,” and witnessed, at Winchester, by Geoffrey de Clinton, in which it is “xxv libras blancas.” Stephen’s charter follows, addressed to William bishop of Exeter, and Richard son of Baldwin, the sheriff, in which again we have “xxv lib. ad scalam.” Lastly, we come to an important entry that seems to have remained unknown:

In 1180, on St. Martin’s Day, king Henry issued (fecit currere) his new money, in the 26th year of his reign, and as the sheriff of Exeter (Exon’) would not pay the prior of Christchurch, for Michaelmas term, £12 16s. 3d. “secundum pondus blancum,” Prior Stephen obtained from the king the following writ.

Then follows a writ which clearly belongs not to 1180, but to an earlier period. It is addressed “prepositis et civibus Exonie,” and directs that the canons are to enjoy their rents as in his grandfather’s time (‘Teste Manessero Biset dapifero, apud Wirecestriam’). Next comes a passage so important that it must be quoted in the original words, although, like the whole of the transcript, it seems slightly corrupt.